How Japanese Carpenters Built Without Nails
Published:
Some of the most impressive engineering solutions do not look like engineering at first. Read more
Blog posts
2026
Google Introduces the Pixel 11 Generation
Published:
Google unveiled the Pixel 11 family on August 12, 2026, maintaining the late-summer announcement window that Google established with Pixel 9 (August 2024) and Pixel 10 (August 2025). The lineup — Pixel 11, 11 Pro, 11 Pro XL, and Pro Fold — is built around the Tensor G6, the sixth generation of Google’s custom SoC (first introduced with Pixel 6 in October 2021). Tensor G6 continues the custom silicon strategy Google started to give it control over the NPU (Neural Processing Unit) for on-device AI inference, the ISP (Image Signal Processor) for computational photography pipelines, and the modem for radio performance — components that third-party Qualcomm Snapdragon chips would optimize for a broader market rather than for Google’s specific software features. The Tensor lineup has moved through successive silicon partners and manufacturing nodes since Tensor G1 (Samsung 5nm), with each generation improving the NPU throughput and energy efficiency available for on-device AI tasks. Read more
Kaizen Small Improvements Big Results
Published:
There is a peculiar kind of problem that almost never gets fixed. It is not serious enough to become an emergency. It does not crash the system, stop production, or ruin the day. It simply wastes a few seconds, creates a little frustration, or requires one unnecessary step. So we tolerate it. Then we tolerate it again tomorrow. And the day after that. Read more
A Personal Website Is Never Really Finished
Published:
A personal website is a strange project because there is no real final version. Every other software project has a definition of done — a feature ships, a bug is fixed, a version is released. A personal website has none of those endpoints. You can always change a sentence, move a section, update a post from five years ago, add a page you have been meaning to build, or remove something you are no longer sure should be there. Read more
Why I Still Want A Web I Can Own
Published:
The modern internet has made publishing effortless. Creating an account, writing something, and reaching readers within minutes is genuinely easier than at any point in history. The platform handles servers, design, discovery, and the distribution mechanics that would otherwise require technical expertise most people do not have or want. Read more
AMD Advancing AI 2026 Shows a Broader Accelerator Ecosystem
Published:
AMD held its Advancing AI 2026 event in July 2026, presenting the Instinct MI450 accelerator alongside a rack-scale system strategy, updated ROCm 7 software stack, and deployment partnership announcements from Microsoft Azure, Meta, and Oracle Cloud Infrastructure. The MI450 used AMD’s CDNA 4 architecture on TSMC’s 3nm process, with 288 GB of HBM3e memory per GPU — matching the NVIDIA B200 Ultra’s memory capacity — and Infinity Fabric 4 connections providing 896 GB/s inter-GPU bandwidth in an 8-GPU rack node configuration. AMD positioned the MI450 specifically against NVIDIA Blackwell Ultra systems for large-scale LLM inference and training. Read more
The Gaming PC I Would Build in 2026
Published:
If I were building a gaming PC on July 1, 2026, this is the machine I would actually want. I am not simply choosing the most expensive part in every category. The priority is gaming performance, then reliability, cooling and enough headroom to keep the system useful for years. Read more
The Skills Computer Science Students Need Beyond Coding
Published:
Students entering computer science programs tend to focus on programming languages, frameworks, and algorithms. This is understandable — these skills are concrete, teachable, and immediately demonstrable. You write code, it does something, and the result is visible. Read more
Apple Unveils Siri AI at WWDC26
Published:
Apple previewed a redesigned Siri at WWDC26 on June 8, 2026, alongside iOS 27, iPadOS 27, macOS 27 (Tahoe), watchOS 27, visionOS 27, and tvOS 27. The updated Siri replaced the older intent-parsing pipeline with a model capable of maintaining multi-turn conversational context — remembering what a user said two exchanges earlier and acting on cumulative instructions without requiring each request to be self-contained. Apple demonstrated examples such as “add the restaurant my sister mentioned in her last email to my Maps list,” a task requiring Siri to access Mail, parse unstructured text, identify an entity, and act across two apps with a single spoken instruction. Read more
Games as a Way to Understand Algorithms
Published:
Games are full of algorithms, but we rarely notice them because we are busy trying to win. A path has to be found. An opponent has to decide what to do. A score has to be updated. Behind many small game actions is a computer science problem that has a real name and a real solution. Read more
What AI Agents Still Get Wrong
Published:
AI agents have moved from demo to daily use over the past year. The pattern is familiar: an agent receives a goal, plans steps, calls tools — reading files, executing code, querying APIs, sending messages — and continues until the task is complete or it encounters something it cannot handle. The step from “answer a question” to “carry out a task with multiple steps and side effects” seemed like a natural progression. The experience of actually using these systems has revealed where the progression is less natural than it looked. Read more
Google I-O 2026 Expands Agentic AI Across Products
Published:
Google I/O 2026 ran May 19–20 in Mountain View, with Sundar Pichai’s keynote focused almost entirely on agentic AI applications. Google announced Gemini 3.5 Flash — a lightweight, low-latency model optimized for real-time interaction — and Gemini Omni, positioned as a model that could natively generate, edit, and reason across text, images, audio, video, and code in a single model pass rather than routing to separate specialist models for each modality. Gemini Omni was available initially to Gemini Advanced subscribers and through Vertex AI. Read more
Ubuntu 26.04 LTS Arrives
Published:
Canonical released Ubuntu 26.04 LTS “Resolute Raccoon” on April 23, 2026, following the two-year LTS cadence from Ubuntu 24.04 LTS “Noble Numbat” (April 2024) and five years after Ubuntu 22.04 LTS “Jammy Jellyfish” (April 2022). As the eleventh long-term-support release, 26.04 carries the standard five-year free security and maintenance update commitment from Canonical for desktop and server, extensible to ten years through the Ubuntu Pro subscription for enterprise users. The kernel version shipped as the 26.04 HWE (Hardware Enablement) base, with a GA kernel selected for stability and broad driver compatibility across the hardware range certified through Canonical’s Ubuntu Certified programme — spanning AMD Ryzen and EPYC, Intel Core and Xeon, Ampere Altra, and the ARM server lineup that had grown substantially in cloud deployments since Graviton2. GNOME 49 on the desktop introduced refinements to fractional scaling (HiDPI displays at non-integer ratios, a long-standing pain point), improved color management pipeline integration, and updated Wayland protocol support across graphics hardware. PipeWire had fully replaced PulseAudio for audio and JACK for pro-audio use cases, completing the audio subsystem transition that began with Ubuntu 22.10. Read more
Why I Still Enjoy Plain HTML, CSS, and JavaScript
Published:
In March 2026, the modern web development landscape includes dozens of frameworks, meta-frameworks built on top of frameworks, build tools that compile JavaScript that compiles to JavaScript, AI-assisted scaffolding that generates entire project structures before you have written a line of code, and package managers competing to install those structures most efficiently. Read more
NVIDIA Unveils Vera Rubin at GTC
Published:
NVIDIA held GTC 2026 in San Jose in March, with Jensen Huang’s keynote presenting the Vera Rubin platform as a fully integrated AI computing system rather than a discrete GPU. Vera Rubin paired the Rubin GPU with the Vera CPU — NVIDIA’s first custom Arm-based server processor, moving away from the AMD EPYC CPUs used in previous DGX systems. Vera CPUs were connected to Rubin GPUs using NVLink-C2C, the same chip-to-chip interconnect used internally within Grace Hopper Superchips, providing bandwidth of 900 GB/s between CPU and GPU — roughly 7× PCIe 5.0. This allowed Vera Rubin compute nodes to treat CPU and GPU memory as a single flat address space rather than copying data across a PCIe bus. Read more
Learning by Simulation
Published:
Some ideas become easier the moment they start moving. A static diagram can explain a binary search tree, but watching a node rotate during an AVL rebalancing — seeing the pointers change in sequence — creates a different kind of understanding. A paragraph can describe a CPU scheduling algorithm, but watching processes arrive, wait in the queue, and get CPU time in a step-through simulation makes the policy feel concrete rather than abstract. Read more
Anthropic Releases Claude Opus 4.6
Published:
Anthropic released Claude Opus 4.6 on February 5, 2026, available through the Claude API at $15 per million input tokens and $75 per million output tokens, with a 200K-token context window in standard usage and a 1M-token context window in beta for enterprise customers with large codebase or document-analysis requirements. Claude Opus 4.6 was positioned as Anthropic’s most capable model for agentic tasks — long-running workflows where the model executes sequences of tool calls, code execution, web searches, and file operations across sessions measured in minutes or hours rather than single-turn exchanges. Read more
The Return Of Small Software
Published:
Software has become incredibly capable, and that capability has brought weight. A task that should take thirty seconds — resizing an image, converting a file format, extracting text from a PDF — can now require creating an account, accepting a terms of service agreement, waiting for an upload, and navigating an interface designed for users who do this professionally every day. Read more
Microsoft Introduces the Maia 200 AI Accelerator
Published:
Microsoft introduced Maia 200 at its January 2026 Azure hardware event as the successor to the first-generation Maia 100 accelerator (announced November 2023, deployed in Azure in 2024). Maia 200 is manufactured on TSMC’s 3nm process and contains 216 GB of HBM3e memory across six stacks, delivering approximately 7 TB/s of memory bandwidth — roughly 1.75× the bandwidth of an NVIDIA H100 SXM5 (3.35 TB/s) and competitive with the H200 (4.8 TB/s). The chip includes 272 MB of on-chip SRAM, 10+ petaFLOPS at FP4, and 5+ petaFLOPS at FP8 for matrix operations used in transformer inference. Read more
2025
OpenAI Releases GPT-5.2
Published:
OpenAI released GPT-5.2 on December 11, 2025 as an incremental update to GPT-5 (which had launched in May 2025 as a unified model intended to consolidate the separate GPT-4o and o-series reasoning models into a single system). GPT-5.2 focused on reliability improvements for long-running agentic tasks: more consistent tool use (fewer dropped function calls in multi-step sequences), improved handling of very long context documents (128K+ token inputs with better attention to content in the middle of the context window — a known weakness in all transformers), and tighter code generation that produced fewer subtle logic errors in multi-file software tasks. OpenAI’s internal and third-party evaluations on agentic benchmarks (SWE-bench for software engineering, GAIA for general assistant tasks, and internal multi-step research task suites) showed measurable improvements over GPT-5.0, though the improvements were characterized as refinements rather than capability jumps. Read more
Can AI Make Us Better Programmers or Just Faster Ones?
Published:
AI tools have made certain parts of programming faster in ways that are hard to argue with. Generating a function from a description, explaining what an error message means, writing test cases for a known interface, translating code between languages — these tasks take a fraction of the time they did two years ago. That efficiency gain is real and significant. Read more
Google Releases Gemini 3
Published:
Google released Gemini 3 on November 18, 2025, making it available the same day in the Gemini app (replacing Gemini 2.0 Flash as the default), Google AI Studio, Vertex AI, and Google Search’s AI Mode — the generative answer panel that had been gradually expanding across Search since mid-2024. Gemini 3 came in Ultra, Pro, and Flash tiers; the Flash variant was designed for low-latency applications at lower per-token cost, while Ultra targeted complex reasoning and multimodal tasks. Read more
Remembering Atatürk on November 10, 2025
Published:
Every November 10, Turkiye remembers Mustafa Kemal Ataturk. Read more
Ubuntu 25.10 Advances the Linux Desktop
Published:
Ubuntu 25.10 “Questing Quokka” was released on October 23, 2025, the interim release immediately preceding Ubuntu 26.04 LTS, scheduled for April 2026 with five years of standard enterprise support. Like all non-LTS Ubuntu releases, 25.10 received only nine months of security maintenance (through July 2026), targeting developers, enthusiasts, and infrastructure teams who want current software versions rather than the stability-optimized LTS freeze. The release shipped GNOME 48, which introduced a tiled window management feature (GNOME’s response to years of community requests for built-in tiling without extensions), updated color management, and improved notification grouping. The included Linux kernel was 6.12, which had reached upstream LTS status in December 2024, adding improved scheduler latency for interactive workloads and continued refinement of the SCHED_EXT extensible scheduler framework that allows user-space BPF programs to implement custom scheduling policies. Read more
Welcome to Our Department
Published:
Welcome to Computer Engineering in 2025. Read more
What Makes a Good Programming Assignment?
Published:
A programming assignment looks simple to describe: write a program that does something, test it, submit it. But the difference between an assignment that produces genuine learning and one that produces submitted code is significant, and it is mostly in the design. Read more
Apple Introduces Its 2025 iPhone Generation
Published:
Apple announced the iPhone 17, iPhone 17 Plus, iPhone 17 Pro, and iPhone 17 Pro Max on September 9, 2025, with sales beginning September 19. The standard iPhone 17 received the most significant redesign in years: a slimmer aluminum-and-glass form factor and an A18 chip (previously exclusive to Pro models in 2024), bringing faster Neural Engine performance and improved camera processing to the base lineup. iPhone 17 Air — replacing the Plus as the thin-and-light option — launched at 5.5mm thin, the slimmest iPhone ever, at the cost of a smaller battery. Read more
Why Choose Computer Engineering in 2025?
Published:
In 2025, computers can already generate code, explain errors, summarise documents, create images, and assist with many tasks that once required substantial manual work. It is reasonable for a student to ask: if AI can do more of the work, why study Computer Engineering? Read more
The Beauty of Simple Algorithms
Published:
I like clever algorithms, but I often enjoy simple algorithms more. There is something satisfying about a solution that looks almost obvious after you understand it — where the entire idea fits in your head at once. Read more
OpenAI Releases GPT-5
Published:
OpenAI released GPT-5 on August 7, 2025, available to ChatGPT users across Free, Plus ($20/month), and Pro ($200/month) tiers and via the API. Rather than launching as a separate model requiring users to switch manually from GPT-4o or o3, GPT-5 was positioned as a unified system: it handled conversational and creative tasks with the low latency of earlier models while applying extended reasoning — essentially the o-series chain-of-thought approach — automatically when the system judged a problem hard enough to warrant it. Read more
OpenAI Introduces ChatGPT Agent
Published:
OpenAI launched ChatGPT Agent in July 2025, combining GPT-5’s language understanding with a set of tools — web browser control, Python code execution via a sandboxed interpreter, file reading and writing, and external API calls — into a product designed to complete multi-step tasks with minimal user prompting per step. Unlike the earlier Operator (launched January 2025 for browser-only tasks) or Code Interpreter (launched June 2023 for data analysis in a stateless sandbox), ChatGPT Agent maintained state across a full session: it could browse the web, download a CSV file, write a Python script to process it, debug an error in the script, generate a chart, and return the result in one continuous workflow without the user re-explaining context between steps. Read more
The Gaming PC I Would Build in 2025
Published:
If I were building a gaming PC on July 1, 2025, this is the machine I would actually want. I am not simply choosing the most expensive part in every category. The priority is gaming performance, then reliability, cooling and enough headroom to keep the system useful for years. Read more
The Value of Building Things Nobody Asked For
Published:
Some of my favorite personal projects started with a thought that is hard to justify: I want to see what happens if I make this. No one asked for it. No deadline. No user waiting. Sometimes no clear reason beyond the fact that the idea kept returning to my mind until I decided to try it. Read more
Apple WWDC Expands On-Device AI Development
Published:
Apple’s WWDC 2025 (June 9–13, Cupertino) introduced iOS 19, iPadOS 19, and macOS 16, with a significant portion of the developer keynote dedicated to expanding the Foundation Models framework that had debuted in iOS 18/macOS 15 (shipped September 2024). The Foundation Models framework allowed third-party applications to call Apple’s on-device approximately-3-billion-parameter language model without packaging model weights in the app bundle or routing requests to external APIs. iOS 19’s expansion added a Streaming API (returning partial tokens as they were generated rather than waiting for the full response), a multi-modal input API (accepting image and text prompts for the same model), and task-specific Adapter APIs — lightweight fine-tuned adapter layers (roughly 100 MB each) that developers could download and attach to the system model to specialize it for domain-specific tasks like legal document analysis or code review without replacing the base model. Read more
What Teaching Programming Taught Me About Patience
Published:
Programming looks very logical from the outside. Teaching programming is much more emotional. A student can understand a difficult idea in five minutes and then spend half an hour looking for a missing semicolon. Another student may struggle with the same topic for days and suddenly understand it after one small example — the third or fourth version of an explanation you were not sure was any different from the others. Read more
Google I-O Pushes Gemini Deeper Into Search and Android
Published:
Google I/O 2025 was held May 20–21, 2025, at the Shoreline Amphitheatre in Mountain View, and the two-day developer conference was organized almost entirely around Gemini integration across Google’s product surface. The headlining model announcement was Gemini 2.5 Pro, which Google described as its most capable model to date, alongside Gemini 2.5 Flash for high-throughput lower-cost applications. AI Mode in Google Search — an opt-in interface replacing the standard ten-blue-links results page for complex queries — was expanded from its experimental state to a broader US rollout, presenting multi-step AI-generated responses with citations and follow-up query suggestions. Project Astra, Google’s universal AI assistant research prototype using a smartphone camera and screen as real-time context (demonstrated internally in 2024), was announced as the foundation for Gemini Live’s expanded capabilities: the live version of Gemini on Android could now see and respond to what the phone’s camera was pointed at, maintaining a spoken conversation while analyzing video frames in real time. Jules, an autonomous coding agent integrated with GitHub, was announced as capable of reading a GitHub issue, writing a fix across multiple files, running tests, and opening a pull request — competing directly with Anthropic’s Claude Code and GitHub Copilot Workspace in the agentic developer tooling space. Read more
Why Data Structures Still Matter in the Age of AI
Published:
AI can write code very quickly now. It can produce a function, explain an error, or suggest a solution in seconds. Because of this, I sometimes hear a simple question: if AI can write the code, do students still need to learn data structures? Read more
Meta Releases a New Llama Generation
Published:
Meta released Llama 4 Scout and Maverick on April 5, 2025, marking the Llama family’s shift from dense transformers to mixture-of-experts (MoE) architecture. Scout had 17 billion active parameters out of 109 billion total (16 experts, one selected per token), and a 10 million token context window — the longest of any openly released model at the time. Maverick had 17 billion active parameters out of 400 billion total (128 experts) and achieved competitive scores against GPT-4o and Gemini 2.0 Flash on coding and reasoning benchmarks. A third model, Llama 4 Behemoth at over 2 trillion total parameters, was announced as still in training. Read more
NVIDIA GTC Focuses on Blackwell Ultra and Rubin
Published:
NVIDIA held GTC 2025 in San Jose from March 17–21, with Jensen Huang’s keynote detailing the Blackwell Ultra accelerator and announcing the Rubin architecture roadmap. Blackwell Ultra — the B200 Ultra — raised HBM3e memory to 288 GB per GPU (up from 192 GB on the B200) and increased FP4 throughput to support the larger context windows and reasoning chains frontier models were beginning to require. NVLink 6 connected up to 576 B200 Ultra GPUs within a single NVLink domain, enabling the full weight matrix of a large model to reside in fast GPU memory rather than requiring frequent host-memory transfers. Read more
OpenAI Expands Agentic Research Tools
Published:
OpenAI released Deep Research on February 2, 2025, initially for ChatGPT Pro subscribers ($200/month). The feature used an internal version of the o3 reasoning model to autonomously conduct multi-step research tasks: planning a search strategy, executing dozens of web searches, reading and synthesizing source documents, tracking intermediate findings, and producing a 2,000–5,000 word report with inline citations. A typical Deep Research task ran for 5–30 minutes without user intervention, producing outputs comparable in depth to what a research analyst might produce in a few hours. The model was given access to Bing web search and document reading tools; it could adapt its search plan based on what it found at each step, a form of “reflection” where the model assessed the quality of evidence and decided whether to investigate further or broaden scope. Google had launched Gemini Deep Research on December 9, 2024 (powered by Gemini 1.5 Pro with Google Search), making this the first major head-to-head competition between deep research agents within weeks of each other. Read more
How AI Shape The Future Of Coding
Published:
In February 2025, AI code generation has moved from novelty to daily practice for a significant fraction of working developers. GitHub Copilot has been available for almost three years. Cursor and similar editors embed AI at a deeper level than autocomplete. Reasoning-capable models can now engage in multi-step software design conversations. The question has shifted from “will this change programming?” to “how is it already changing it and what comes next?” Read more
Clean Code Psychopath
Published:
John Woods wrote something that has stayed with me: “Always code as if the guy who ends up maintaining your code will be a violent psychopath who knows where you live.” The quote is usually cited as a case for clean code. But there is a darker inversion of the same pressure that I have been thinking about — the developer who becomes so committed to an idea of clean code that the code stops serving the project. Read more
DeepSeek R1 Shakes the AI Model Market
Published:
DeepSeek released R1 on January 20, 2025, alongside a detailed technical report describing the model’s architecture and training methodology. DeepSeek is a Chinese AI research lab founded in 2023 as an offshoot of High-Flyer Capital Management, a quantitative hedge fund, with the stated goal of pursuing AGI research independent of product revenue pressures. The R1 model itself was a 671-billion-parameter Mixture-of-Experts architecture with approximately 37 billion parameters active per token during inference — a design that substantially reduced the computational cost of a forward pass relative to a dense model of equivalent total parameter count. The training pipeline centered on GRPO (Group Relative Policy Optimization), a reinforcement learning technique that rewarded correct answers on verifiable tasks (mathematics, competitive programming, formal logic) without requiring a separately trained reward model for each capability area. An intermediate model, R1-Zero, was trained using only RL with no supervised fine-tuning whatsoever; it spontaneously developed extended internal reasoning traces — long chains of intermediate steps visible in the model’s output before its final answer — a behavior the researchers described as emerging from the optimization objective rather than being explicitly trained. The final R1 model incorporated a small amount of supervised fine-tuning on high-quality examples followed by additional RL training, and performed comparably to OpenAI’s o1 model on benchmarks including AIME 2024 (mathematics olympiad problems), Codeforces (competitive programming), and MATH-500. Read more
Father Son Game Dev
Published:
Many children spend hours playing games. Some of them eventually ask a more interesting question: how are games made? That question is an opening. I have been exploring it with my son, and I want to share what the experience has actually taught me — about teaching, about programming, and about spending time together on something real. Read more
2024
OpenAI Releases the o3 Reasoning Model Preview
Published:
OpenAI previewed o3 on December 20, 2024 in a “12 Days of OpenAI” event, releasing benchmark results before the model was publicly available. The headline number was 87.7% on ARC-AGI (Abstraction and Reasoning Corpus for Artificial General Intelligence), a benchmark of visual pattern-recognition puzzles that had been specifically designed to resist language models through its emphasis on systematic generalization rather than pattern matching from training data. GPT-4o had scored 5% on the same benchmark; the previous state-of-the-art AI system had reached 53%. ARC-AGI creator François Chollet had explicitly stated that the benchmark could not be gamed by scale alone, making o3’s score the most discussed AI capability result of 2024. Read more
NVIDIA Becomes Central to the AI Infrastructure Boom
Published:
By November 2024, NVIDIA had become the indispensable infrastructure provider for large-scale AI training and inference, with its market capitalization briefly reaching $3.4 trillion in June 2024 — briefly making it the world’s most valuable company and reflecting the AI infrastructure spending wave triggered by ChatGPT’s November 2022 launch. The H100 SXM5 (Hopper architecture, TSMC N4, 80 GB HBM3, 3.35 TFLOPS BF16/FP16 dense training throughput, 700W TDP) had become the standard unit for LLM training: OpenAI trained GPT-4 on an estimated 25,000 A100s (Hopper’s predecessor); Anthropic, Google, and Meta deployed H100 clusters of similar scale. H200 SXM5 (announced December 2023, HBM3e 141 GB at 4.8 TB/s bandwidth vs H100’s 80 GB at 3.35 TB/s) offered approximately 1.9× inference throughput on memory-bound LLM serving workloads. NVIDIA’s GB200 NVL72 rack (announced March 2024, available late 2024 in Blackwell generation): 36 Grace Arm CPUs + 72 Blackwell B200 GPUs in a 72-GPU NVLink domain, delivering 1.44 ExaFLOPS FP4 for AI inference — targeting trillion-parameter model serving at economics impractical with H100. Read more
Remembering Atatürk on November 10, 2024
Published:
Every November 10, Turkiye pauses to remember Mustafa Kemal Ataturk. Read more
Intel Announces Core Ultra 200S
Published:
Intel announced Core Ultra 200S desktop processors (codenamed Arrow Lake) on October 24, 2024 for the LGA1851 socket. The flagship Core Ultra 9 285K featured 8 Lion Cove P-cores and 16 Skymont E-cores plus 4 low-power E-cores — 24 cores total — with no HyperThreading on any core: Intel removed simultaneous multithreading (SMT) from Lion Cove’s design entirely, arguing that the wider out-of-order execution engine in Lion Cove (larger instruction window, more execution units per core than Raptor Lake’s Golden Cove) made SMT less valuable than adding real cores. The 285K had a 125W TDP base and 250W max turbo power, with boost clocks to 5.7 GHz. The design was a disaggregated package: the CPU compute tile was manufactured on Intel 20A (Intel’s own process, roughly equivalent to TSMC N5 class), the GPU tile and SoC tile on TSMC N6, and the I/O tile on TSMC N6, all connected via Intel’s Foveros Direct die-to-die interconnect at 36 μm pitch — the same approach Intel used for Meteor Lake mobile (December 2023) now arriving at the desktop. Read more
Toyota The Company That Made Efficiency A Philosophy
Published:
Toyota is often described as one of the world’s largest automobile manufacturers. That description is correct. It is also incomplete. Toyota became important not only because of the cars it produced, but because of the way it learned to produce them. Read more
Welcome to Our Department
Published:
If you are beginning Computer Engineering in 2024, welcome to a field where tools are becoming more powerful and judgment is becoming more important. Read more
OpenAI Releases the o1 Reasoning Model
Published:
OpenAI released o1 on September 12, 2024 — initially called “o1-preview” for ChatGPT Plus and Team subscribers at $20/month, with API access at $15 per million input tokens and $60 per million output tokens. A smaller, faster variant called o1-mini launched the same day at $3/$12 per million tokens, designed for coding and STEM tasks on a tighter budget. The full o1 model (removing the “preview” label) launched December 5, 2024. Read more
Why Choose Computer Engineering in 2024?
Published:
In 2024, artificial intelligence is becoming a normal part of software products rather than a separate research topic. The pace of change this year has been striking even by recent standards. Read more
Google Introduces the Pixel 9 Generation Early
Published:
Google announced the Pixel 9 on August 13, 2024 — the earliest ever for a Pixel launch, moved from October to beat Apple’s iPhone 16 announcement in September. The Pixel 9 lineup included four models: Pixel 9 ($799, 6.3-inch), Pixel 9 Pro ($999, 6.3-inch), Pixel 9 Pro XL ($1,099, 6.8-inch), and Pixel 9 Pro Fold ($1,799, foldable). All used Google’s Tensor G4 processor (manufactured by Samsung on 4nm), with 12GB RAM on the standard Pixel 9 and 16GB on Pro models — increases over the Tensor G3’s 12GB configurations, explicitly motivated by on-device AI workload requirements. Read more
CrowdStrike Update Causes a Global Windows Outage
Published:
At approximately 04:09 UTC on July 19, 2024, CrowdStrike began deploying a content configuration update to its Falcon sensor — the kernel-level security agent running on Windows endpoints — that caused millions of machines to crash with a Blue Screen of Death (BSOD) and enter an unrecoverable boot loop. CrowdStrike Falcon operates as a kernel-mode driver (running at CPU Ring 0 privilege) that inspects process creation, network connections, file system operations, and memory allocations at a level below normal applications, giving it visibility into attacks that user-mode security tools cannot detect. The faulty update was not the Falcon sensor binary itself but a “rapid response content” channel file — a configuration update named C-00000291-*.sys that defined detection logic templates for a new IPC (Inter-Process Communication) Template Type. A bug in the validator for this new template type caused the Falcon sensor to attempt to access memory through a null pointer when loading the file during Windows boot, triggering a kernel STOP error before Windows could complete initialization. Machines that had already booted and were running were not immediately affected; machines that rebooted after 04:09 UTC during the rollout window entered the crash loop. Read more
The Gaming PC I Would Build in 2024
Published:
If I were building a gaming PC on July 1, 2024, this is the machine I would actually want. I am not simply choosing the most expensive part in every category. The priority is gaming performance, then reliability, cooling and enough headroom to keep the system useful for years. Read more
Apple Announces Apple Intelligence
Published:
Apple announced Apple Intelligence at WWDC on June 10, 2024, as a set of generative AI capabilities built into iOS 18, iPadOS 18, and macOS Sequoia. Initial features included Writing Tools (rewrite, proofread, and summarize text in any app), Image Playground (generating images from text prompts in three styles), Genmoji (custom emoji generated from descriptions), a redesigned Siri with conversational context and screen awareness, and a summarized notification inbox using on-device language models. Apple Intelligence launched in beta with iOS 18.1 in October 2024, initially for US English only, with language expansion planned through 2025. Read more
Microsoft Introduces Copilot Plus PCs
Published:
Microsoft announced Copilot+ PCs on May 20, 2024, defining a hardware category requiring a minimum 40 TOPS (tera-operations per second) NPU (Neural Processing Unit) — more than any existing Windows laptop NPU at the time. The first wave of Copilot+ PCs launched June 18, 2024, using Qualcomm’s Snapdragon X Elite and Snapdragon X Plus processors, which contained an Oryon CPU (derived from Nuvia’s acquisition), an Adreno GPU, and a Hexagon NPU rated at 45 TOPS. Devices from Surface, Samsung, Dell, HP, ASUS, Lenovo, and Acer launched simultaneously, priced from $999. Intel and AMD Copilot+ PC laptops (using Intel Lunar Lake and AMD Strix Point with their respective 47–50 TOPS NPUs) followed in fall 2024. Read more
Meta Releases Llama 3
Published:
Meta released Llama 3 on April 18, 2024, with 8-billion and 70-billion parameter models in both base and instruction-tuned variants, distributed through Hugging Face, llama.meta.com, and major cloud providers (AWS Bedrock, Google Vertex AI, Azure AI). The 8B and 70B Instruct variants outperformed comparable open-weight models (Mistral 7B, Gemma 7B) on standard benchmarks and approached GPT-3.5 Turbo performance on tasks like MMLU (79.5% for Llama 3 70B vs GPT-3.5’s ~70%) and coding. Meta announced that a Llama 3 model exceeding 400 billion parameters was in training. Read more
AI Coding Assistants Are Becoming Part Of The Editor
Published:
GitHub Copilot launched technical preview in June 2021 and became generally available in June 2022. In the roughly two years since, it has gone from curiosity to part of the daily development environment for many teams. By March 2024, GitHub reports over a million paying users and adoption at the majority of Fortune 500 companies. ChatGPT and Claude have expanded the use case beyond inline completion to conversational assistance — asking questions about code, requesting explanations, and generating complete functions in a chat interface. Cursor, a VS Code fork with deeper model integration, has emerged in 2023-2024 as a model for what tight AI-editor integration can look like. Read more
NVIDIA Announces the Blackwell GPU Architecture
Published:
NVIDIA announced the Blackwell GPU architecture at GTC on March 18, 2024 in a keynote where Jensen Huang said the company was “not in the semiconductor business — we’re in the infrastructure business.” The flagship B200 GPU used two connected dies (the GB202 silicon) totaling 208 billion transistors on TSMC’s 4NP custom process — more than double the H100’s 80 billion transistors on 4N. The B200 delivered 20 petaFLOPS at FP4, compared to the H100’s 4 petaFLOPS at FP8 (halving precision doubles throughput, so the increase was partly from precision and partly from transistor count). HBM3e memory reached 192 GB per B200 with 8 TB/s bandwidth — nearly 2.4× the H100’s 3.35 TB/s. Read more
Google Releases Gemini 1.5
Published:
Google announced Gemini 1.5 Pro on February 15, 2024 in a research preview that made it available to select developers through Google AI Studio and the Gemini API. The headline capability was a context window of up to one million tokens — roughly 750,000 words, equivalent to approximately 30,000 lines of code, 11 hours of audio, 1 hour of video, or several complete novels in a single request. For comparison, GPT-4 Turbo (announced November 6, 2023) offered a 128,000-token context, and Claude 2.1 (November 2023) offered 200,000 tokens; Gemini 1.5’s one-million-token window was five to eight times larger than any commercially available alternative at the time. Google demonstrated the capability with tasks including analyzing an entire 402-page technical manual to answer detailed questions, understanding a full-length movie provided as video frames, and performing few-shot in-context learning from hundreds of examples rather than the dozens that fit in previous context windows. The model also supported multimodal inputs — text, images, audio, and video — within the same context window, enabling queries like “at what timestamp in this hour-long lecture does the professor make an error in this equation?” Read more
Apple Vision Pro Reviews Bring Spatial Computing Into Focus
Published:
Apple opened Vision Pro preorders on January 19, 2024, selling out the first wave in minutes; early pre-order confirmations showed delivery dates in late February. Review embargoes lifted February 1, 2024, one day before the retail launch on February 2. The $3,499 starting price (with prescription lens inserts adding $149–$299) resulted in a device that journalists handled as a premium technology artifact rather than a consumer product, and reviewer consensus reflected that tension: extraordinary engineering at a price and form factor that limited its audience. Read more
2023
Google Releases Gemini
Published:
Google introduced Gemini on December 6, 2023, as a natively multimodal model family in three sizes: Nano (running on-device in Pixel 8 Pro phones), Pro (replacing Bard as Google’s conversational AI), and Ultra (the largest version, subject to further safety testing before release). Unlike earlier Google models that added multimodal capability by aligning separate text and vision encoders, Gemini was trained from the start on interleaved sequences of text, images, audio, and video, allowing reasoning that crossed modalities within a single pass. Read more
Java Superstars
Published:
Java 21 arrived in September 2023 as the latest long-term support release, and it is a significant one. Virtual threads — the main deliverable of Project Loom — are now production-ready. Structured concurrency and scoped values are in preview. Pattern matching for switch is finalized. Record patterns work in instanceof checks. After years of shorter improvement cycles since Java 9, the platform has changed considerably. The people shaping this evolution are worth understanding, but so is the community structure that makes Java’s development sustainable. Read more
OpenAI’s Leadership Crisis Shakes the AI Industry
Published:
On November 17, 2023, OpenAI’s board — composed of Ilya Sutskever (co-founder and Chief Scientist), Adam D’Angelo (Quora CEO), Tasha McCaulay (FTX Future Fund), and Helen Toner (Georgetown CSET) — fired CEO Sam Altman in a surprise announcement, stating he “was not consistently candid in his communications with the board.” Greg Brockman, President and co-founder, resigned in protest within hours. The board named CTO Mira Murati interim CEO, then replaced her with Emmett Shear (co-founder of Twitch) days later. The timing came 11 days after OpenAI’s first developer conference (DevDay, November 6), where GPT-4 Turbo and the GPT Store had been announced, with ChatGPT at approximately 100 million weekly active users. Read more
Remembering Atatürk on November 10, 2023
Published:
Every November 10, Turkiye remembers Mustafa Kemal Ataturk. Read more
Staying Engaged In Computer Science Classes
Published:
Computer science classes can become difficult for a reason that has nothing to do with intelligence: the subject moves from concrete to abstract unusually fast. A student learns to write a loop on Monday and hears about time complexity, memory models, and formal proofs before the month ends. It is easy to lose the thread between what you can see on the screen and what the teacher is now talking about. Read more
Ubuntu 23.10 Shows the New Desktop Installer Direction
Published:
Ubuntu 23.10 “Mantic Minotaur” was released on October 12, 2023 as a nine-month interim release between the Ubuntu 23.04 (April 2023) and Ubuntu 24.04 LTS (April 2024) releases. The most significant infrastructure change was the default deployment of Canonical’s new Flutter-based desktop installer, replacing the Ubiquity installer that had been the standard Ubuntu desktop installation tool since Ubuntu 5.10 “Breezy Badger” in October 2005. Ubiquity was built on Python and GTK, and while it had received updates throughout its 18-year life, its architecture made it difficult to share code between the desktop installer, the Ubuntu Server installer (Subiquity, introduced in Ubuntu 18.04), and the Ubuntu Core IoT/embedded installer. The new installer, written in Dart using the Flutter framework (the same cross-platform UI toolkit Google uses for mobile apps), shared the Subiquity backend used for server installations and presented a redesigned step-by-step interface with cleaner visuals, accessibility improvements, and a new partitioning UI. The choice of Flutter was notable: it was a cross-platform framework primarily associated with mobile and web development, not traditional Linux desktop software, reflecting Canonical’s assessment that Flutter’s widget consistency across platforms made it the most practical choice for sharing a single codebase across the desktop, server, and embedded installer variants. Other desktop changes included GNOME 45 (released September 20, 2023), which featured a new Quick Settings panel replacing the old system menu, an updated Activities overview with a workspace indicator in the top bar, and a breaking change to GNOME Shell extension APIs that required all existing extensions to be updated for GNOME 45 compatibility — a recurring pattern in GNOME releases that frustrated extension maintainers. Read more
Welcome to Our Department
Published:
Welcome to Computer Engineering in 2023. Read more
Apple Introduces the A17 Pro
Published:
Apple introduced the A17 Pro on September 12, 2023, exclusively in the iPhone 15 Pro ($999) and Pro Max ($1,199) — the first time Apple gave a chip a “Pro” designation to distinguish it from the A16 Bionic continuing in the iPhone 15 and 15 Plus. The A17 Pro was built on TSMC’s first-generation 3nm process (N3B, using EUV lithography), containing 19 billion transistors — up from the A16 Bionic’s 16 billion transistors on TSMC’s 4nm node. The A17 Pro’s 6-core CPU delivered 2 high-performance cores (at approximately 3.78 GHz) and 4 efficiency cores; Apple claimed 10% faster CPU performance than A16 while improving power efficiency at similar performance levels. Read more
Why Choose Computer Engineering in 2023?
Published:
In 2023, generative artificial intelligence has changed the public conversation about computing. Tools that can produce text, images, and code have made many people ask what software work will look like in the future. Read more
LK-99 Claims Trigger Intense Computational Analysis
Published:
On July 22, 2023, two preprint papers appeared on arXiv from Sukbae Lee and colleagues at the Quantum Energy Research Center in Seoul, claiming that a lead-apatite compound doped with copper — chemical formula Pb₉Cu(PO₄)₆O, dubbed LK-99 — exhibited superconductivity at room temperature and ambient pressure. A video showing a small pellet partially levitating above a magnet (a demonstration of the Meissner effect, in which superconductors expel magnetic fields) spread on social media within days. If true, the claim would be the most significant physics discovery in decades: room-temperature superconductors would transform power transmission, computing, and MRI machines by eliminating the need for liquid helium or liquid nitrogen cooling. Read more
Meta Releases Llama 2
Published:
Meta released Llama 2 on July 18, 2023 in partnership with Microsoft, distributing model weights through Hugging Face, llama.meta.com, and Microsoft Azure. The release included three sizes — 7B, 13B, and 70B parameters — in both base (pretrained) and Chat (RLHF-aligned) variants, under a custom commercial license that permitted use by organizations with fewer than 700 million monthly active users without additional licensing, and required a separate agreement from Meta for larger deployments. This was a significant change from the first LLaMA (February 2023), which had been released for research-only use and whose weights leaked via torrent within a week. Read more
The Gaming PC I Would Build in 2023
Published:
If I were building a gaming PC on July 1, 2023, this is the machine I would actually want. I am not simply choosing the most expensive part in every category. The priority is gaming performance, then reliability, cooling and enough headroom to keep the system useful for years. Read more
Apple Introduces the Vision Pro
Published:
Apple introduced Vision Pro at WWDC on June 5, 2023 as a “spatial computer” starting at $3,499. It went on sale February 2, 2024 in the US. The headset contained two micro-OLED displays (23 million pixels total across both eyes, higher than a 4K TV viewed from a typical distance), driven by an M2 chip for application processing and a new R1 chip for real-time sensor fusion. Twelve cameras, six microphones, and five sensors including LiDAR and a TrueDepth camera fed data continuously to the R1, which processed and delivered images to the displays within 12 milliseconds — a latency Apple stated was below human perceptual thresholds for lag between head movement and visual update. Read more
Google Announces the PaLM 2 Model
Published:
Google announced PaLM 2 at Google I/O on May 10, 2023, as the successor to PaLM (Pathways Language Model, April 2022, 540 billion parameters). PaLM 2 was offered in four sizes — Gecko (deployable on mobile devices), Otter, Bison, and Unicorn (the largest, with parameter count undisclosed) — allowing Google to serve the same model family from on-device inference to large cloud deployments. Google’s technical report described PaLM 2 as trained on a significantly broader multilingual corpus (100+ languages and programming content spanning 20+ programming languages) than PaLM, with a smaller model-to-data ratio (a “chinchilla-optimal” approach, citing Hoffmann et al. 2022’s finding that smaller models trained on more data outperform larger models on less data at the same compute budget). Read more
SpaceX Starship Completes Its First Integrated Flight Test
Published:
SpaceX launched Starship’s first integrated flight test (IFT-1) on April 20, 2023 at 8:33 AM CDT from Starbase in Boca Chica, Texas. The fully stacked vehicle — Super Heavy booster (B7) plus Starship upper stage (S24) — stood approximately 120 meters (394 feet) tall, making it the largest rocket ever flown. Super Heavy’s 33 Raptor 2 engines produced approximately 74.4 MN (16.7 million pounds-force) of liftoff thrust, surpassing the Saturn V (35.1 MN) and NASA’s Space Launch System Block 1 (39.1 MN) to make it the most powerful rocket ever successfully launched from the pad. The vehicle successfully cleared the launch tower and rose to approximately 38 km altitude over the Gulf of Mexico before the mission ended with an unplanned vehicle breakup. During the ascent, several Raptor engines failed to ignite or shut down early — approximately 6 of 33 engines were observed non-functional during flight — though the vehicle was designed to fly with 3–4 engine failures. Stage separation (planned as a “hot staging” maneuver, with the upper stage’s engines igniting before booster engine shutdown) did not execute as planned; the vehicle began rotating uncontrollably approximately 2:50 into the flight. The flight termination system (FTS) was triggered by SpaceX’s flight safety officer approximately 4 minutes into the flight as the vehicle descended uncontrolled over the Gulf. Read more
Gordon Moore Has Passed Away At The Age Of 94
Published:
Gordon Moore died on March 24, 2023, at the age of 94. His name is inseparable from one of the most consequential observations in the history of technology: Moore’s Law. The observation shaped not only the semiconductor industry but the software written to run on it, and the expectations of everyone who used computers for half a century. Read more
GPT-4 Is Released
Published:
OpenAI released GPT-4 on March 14, 2023, making it available to ChatGPT Plus subscribers ($20/month) and via the API to developers on a waitlist. Microsoft had already integrated GPT-4 into Bing Chat (launched February 7, 2023) before the formal public release — an early signal that enterprise integration was ahead of consumer access. OpenAI published a GPT-4 Technical Report the following day, disclosing extensive benchmark results but deliberately omitting parameter count, training compute, dataset composition, and architecture specifics, citing “competitive and safety reasons.” This represented a notable break from the research transparency of GPT-2 (2019) and GPT-3 (2020), both of which had technical papers with full methodology. On standard academic benchmarks, GPT-4 scored in the 90th percentile on the bar exam (GPT-3.5 had scored the 10th percentile), in the 93rd percentile on the SAT reading/writing section, and 88th percentile on SAT math. On MMLU (57 academic subjects, 5-shot), GPT-4 scored 86.4% versus GPT-3.5’s 70%. The model launched with 8,192 token context (a 32K context version was available to select API partners), compared to GPT-3.5’s 4,096 tokens. Read more
Meta Releases LLaMA
Published:
Meta released LLaMA (Large Language Model Meta AI) on February 24, 2023, as a family of decoder-only transformer models at four sizes: 7B, 13B, 33B, and 65B parameters. Training used 1 to 1.4 trillion tokens drawn from Common Crawl (67%), GitHub, Wikipedia, books, ArXiv papers, and Stack Exchange — a much larger dataset than GPT-3’s 300 billion tokens. The 13B model matched or exceeded GPT-3 (175B parameters) on most standard benchmarks, demonstrating that smaller models trained longer on cleaner data could match much larger models trained on fewer tokens, a finding that echoed the “Chinchilla scaling laws” paper DeepMind had published in March 2022. Read more
Leonardo Da Vinci: The Man Beyond The Mona Lisa
Published:
When people hear Leonardo da Vinci, the Mona Lisa usually appears first. That is understandable — it is one of the most recognized paintings in the world, displayed behind bullet-proof glass in the Louvre to crowds that number in the tens of thousands daily. But the painting is a narrow window into a much larger life. Leonardo spent more hours filling notebooks with observations and designs than he did holding a paintbrush. Read more
How Do Computers Make Decisions?
Published:
Computers do not make decisions in the human sense. They follow logical rules with perfect consistency and no understanding of why those rules exist. The interesting question is how that simple foundation — binary logic operating on electrical signals — scales up to the decision-making behavior we observe in real systems. Read more
Why Do Computers Use 1s And 0s?
Published:
People often ask why computers use 1s and 0s when we normally count with ten digits. The answer starts with the hardware — but the real reason runs deeper than transistors. Read more
Why We Need Operating Systems?
Published:
A computer without an operating system would still be a computer. The processor would still execute instructions. Memory would still store values. But using it would require detailed knowledge of the specific hardware — the exact memory addresses where data should be placed, the precise sequence of instructions to communicate with a storage device, the mechanism for switching between different tasks. Early computers operated this way. Human operators loaded programs manually, assigned resources directly, and managed the hardware as closely as the machine itself. Read more
Microsoft Expands Its Partnership With OpenAI
Published:
Microsoft announced on January 23, 2023 a new multi-year, multi-billion-dollar investment in OpenAI, reported as approximately $10 billion — the third major investment after $1 billion in 2019 and approximately $2 billion in 2021. The announcement came 54 days after ChatGPT’s November 30, 2022 public launch, during which ChatGPT reached 100 million monthly active users in 2 months — faster than any consumer internet product including TikTok (9 months) and Instagram (2.5 years) to the same milestone. The investment was structured partly in Azure cloud computing credits rather than fully in cash, deepening OpenAI’s existing exclusive reliance on Microsoft Azure for all training and inference compute. The financial structure included a “capped profit” arrangement: Microsoft and other investors receive returns up to a specific multiple of their investment, after which OpenAI’s nonprofit parent entity receives subsequent profits — a structure designed to allow investor returns while preserving OpenAI’s nonprofit mission orientation. Read more
2022
LastPass Discloses Password Vault Breach Details
Published:
LastPass disclosed on December 22, 2022 that an August 2022 security incident — in which an attacker had compromised a developer’s workstation and exfiltrated source code, technical documentation, and internal secrets — had been used as the first stage of a second, larger breach. In the second incident (September through October 2022), the attacker used credentials and keys stolen from a LastPass DevOps engineer’s home computer (compromised through a vulnerable third-party media player with an unpatched remote code execution vulnerability) to access LastPass’s cloud backup storage on Amazon S3. The attacker exfiltrated backups of LastPass customer vault data along with encryption keys and other infrastructure data. The disclosure came four months after the initial August 2022 breach notification, which had incorrectly characterized the damage as limited to source code and had not mentioned that customer vault data was at risk. The delayed and initially incomplete disclosure was widely criticized by security researchers and customers. Read more
To Be A Student Or To Graduate?
Published:
Students and graduates face different kinds of problems, and that difference is larger than it first appears. Understanding it early can make the transition significantly less disorienting. Read more
The Importance Of Active And Collaborative Education
Published:
A student can sit quietly through an entire lesson, nod at the right moments, and still learn very little. This is the central problem with passive education. Attention is not the same as understanding, and understanding is not the same as being able to use something. Read more
OpenAI Releases ChatGPT
Published:
OpenAI released ChatGPT on November 30, 2022 as a free research preview powered by GPT-3.5-turbo, an optimized version of the GPT-3.5 model family trained using InstructGPT techniques developed by OpenAI researchers in 2022. The training pipeline combined supervised fine-tuning (human contractors writing example conversations demonstrating helpful assistant behavior), reward model training (human raters comparing pairs of model responses and indicating which was better), and reinforcement learning from human feedback (RLHF) using Proximal Policy Optimization (PPO) to update the language model’s weights toward higher-scoring responses. The resulting model was dramatically better than raw GPT-3 at following instructions, maintaining topic coherence across a conversation, declining inappropriate requests, and producing well-structured output — behaviors that emerged from the RLHF process rather than raw pretraining. The web interface (chat.openai.com) required only a free account and a web browser, making it the first large language model experience accessible to a mainstream audience without any technical setup. OpenAI had previously offered GPT-3 access through an API requiring developer registration and payment, which limited its reach to technically sophisticated users. ChatGPT eliminated that barrier: within five days of launch it had one million registered users, and within two months it reached 100 million monthly active users — the fastest product in history to reach that milestone, surpassing TikTok (nine months) and Instagram (two and a half years) by a wide margin. Read more
Remembering Atatürk on November 10, 2022
Published:
Every November 10, Turkiye pauses to remember Mustafa Kemal Ataturk. Read more
Elon Musk Completes the Twitter Acquisition
Published:
Elon Musk completed the acquisition of Twitter in October 2022, ending months of legal, financial, and public drama around one of the world’s most visible social platforms. The deal mattered not only because of the price or the personalities involved, but because Twitter had become a strange kind of public infrastructure: part newswire, part social network, part professional identity system, and part political arena. Read more
How AI Tools Empower Developers To Write Better Code
Published:
GitHub Copilot went into technical preview in June 2021. By October 2022, it has been available to the general public for several months, and the reports from developers who use it daily are specific enough to form a clearer picture of what it actually changes. Read more
Welcome to Our Department
Published:
If you are starting Computer Engineering in 2022, welcome. Read more
NVIDIA Announces the RTX 40 Series
Published:
NVIDIA unveiled the GeForce RTX 40 series at a September 20, 2022 event, with the RTX 4090 and two versions of RTX 4080 (16GB and 12GB) announced. The RTX 4090 launched October 12 at $1,599; the 16GB RTX 4080 launched November 16 at $1,199. The 12GB “4080” was withdrawn after public backlash over the naming and re-released in January 2023 as the RTX 4070 Ti at $799. All used the Ada Lovelace architecture on TSMC’s 4N process node (a customized version of 5nm). The RTX 4090 contained 76.3 billion transistors on the AD102 die, nearly double the RTX 3090 Ti’s 40.4 billion on Samsung 8N. Read more
What Are The Latest Trends In Software Development?
Published:
Every year, lists of software development trends appear. Most of them include the same items at different stages of hype. In September 2022, I want to try something different: instead of simply listing what is popular, I want to say what I think is real, what is overstated, and what is worth investing in now. Read more
Why Choose Computer Engineering in 2022?
Published:
In 2022, the importance of computing can be seen from two very different directions: software and chips. Read more
Stable Diffusion Opens a New Era of Generative Images
Published:
Stability AI released Stable Diffusion 1.4 publicly on August 22, 2022, making the model weights downloadable under the CreativeML Open RAIL-M license (allowing commercial use with behavioral restrictions, unlike prior generative image models locked behind APIs). The model had been developed by CompVis (Ludwig Maximilian University of Munich, led by Robin Rombach) with funding from Stability AI and collaboration from Runway ML, based on the “Latent Diffusion Models” paper (Rombach et al., arXiv December 2021, CVPR 2022 Best Paper). The training dataset was LAION-5B (5.85 billion image-text pairs scraped from the public web), the largest openly published training dataset at the time, curated by the LAION non-profit organization. The SD 1.x models fit in approximately 4 GB (fp32 weights) or 2 GB (fp16), runnable on consumer NVIDIA GPUs with 6–8 GB VRAM or Apple Silicon M1/M2 (via CoreML or MPS backends, slower but functional without a discrete GPU) — a stark contrast to DALL-E 2 (OpenAI, April 2022, API-only at ~$0.02/image) and Midjourney (March 2022, Discord bot, subscription, closed source). Read more
James Webb Telescope Images Showcase Massive Scientific Computing
Published:
NASA publicly released the first full-color science images from the James Webb Space Telescope on July 12, 2022 — President Biden previewed one image (SMACS 0723, a gravitational lensing deep field) at the White House on July 11. JWST had launched December 25, 2021 on an Ariane 5 rocket, deployed its 18-segment primary mirror and sunshield over two weeks in orbit, and reached its operational position at the Sun-Earth Lagrange Point 2 (approximately 1.5 million kilometers from Earth in the direction away from the Sun) on January 24, 2022. Mirror alignment completed April 28, 2022, and commissioning of the four science instruments concluded in late June 2022 before the July release. The five inaugural images demonstrated capabilities across JWST’s instrument suite: the SMACS 0723 deep field showed thousands of galaxies as they existed 13 billion years ago through gravitational lensing; the Carina Nebula image revealed previously obscured star formation regions in infrared wavelengths that visible-light telescopes couldn’t penetrate; the WASP-96b spectrum showed the atmospheric transmission spectrum of an exoplanet with identifiable water vapor, carbon dioxide, and cloud signatures; the Southern Ring Nebula showed a dying star’s ejected shells in both NIRCam and MIRI imaging; and Stephan’s Quintet showed five interacting galaxies in an unprecedented wide-field infrared mosaic. Read more
The Gaming PC I Would Build in 2022
Published:
If I were building a gaming PC on July 1, 2022, this is the machine I would actually want. I am not simply choosing the most expensive part in every category. The priority is gaming performance, then reliability, cooling and enough headroom to keep the system useful for years. Read more
Apple Announces the M2 Processor
Published:
Apple announced the M2 processor at WWDC on June 6, 2022, first shipping inside a redesigned 13-inch MacBook Air that launched June 17 at $1,199. M2 used TSMC’s second-generation 5nm process and kept the same broad heterogeneous architecture as M1 — four performance cores, four efficiency cores, and up to ten GPU cores — while improving each block. Read more
Google Reveals Imagen
Published:
Google Research published the Imagen paper on May 23, 2022, alongside a website showing examples — but without releasing the model or a public API. Imagen combined a frozen 4.6-billion-parameter T5-XXL text encoder (the same language model used for NLP tasks) with a pipeline of three cascaded diffusion models: a base model generating 64×64 images from text, then two super-resolution diffusion models upscaling to 256×256 and 1024×1024. The T5-XXL encoder had been trained on text only, not image-text pairs, and Google’s finding was that using a large language model’s text representations as conditioning signals produced better text-image alignment than training a custom CLIP-style encoder specifically for image generation — suggesting that language understanding, not vision-specific training, drove Imagen’s prompt adherence. Read more
Generative Image Models Are Getting Hard To Ignore
Published:
Image generation from text has improved remarkably quickly. Systems such as DALL-E 2, announced by OpenAI this year, can create images from written descriptions with a level of coherence that would have looked much less convincing only a short time ago. Read more
Elon Musk Reaches an Agreement to Buy Twitter
Published:
Elon Musk disclosed a 9.2% stake in Twitter on April 4, 2022, making him the company’s largest shareholder, then reached an agreement to acquire Twitter for $54.20 per share — approximately $44 billion total — on April 25, 2022, after initially being offered a board seat. Musk financed the deal with $21 billion in equity (his own funds and co-investors including Oracle’s Larry Ellison, who committed $1 billion), a $12.5 billion margin loan against Tesla stock, and $13 billion in bank debt underwritten by Morgan Stanley, Bank of America, Barclays, and others. Read more
How Can I Stay Up-to-date With New Technologies And Best Practices?
Published:
Software development changes quickly enough to make almost everyone feel behind sometimes. In early 2022, there is Rust gaining serious traction for systems work, TypeScript becoming the default for large JavaScript projects, Kubernetes appearing in most deployment discussions, and a growing set of AI-assisted coding tools that did not exist two years ago. Meanwhile, fundamentals like data structures, networking, and operating systems are unchanged. Read more
NVIDIA Introduces the Hopper GPU Architecture
Published:
NVIDIA CEO Jensen Huang announced the Hopper GPU architecture at GTC (GPU Technology Conference) on March 22, 2022, during a keynote delivered in a chip fab-styled theatrical setting featuring a life-size H100 GPU prop. The flagship H100 (die GH100) was manufactured on TSMC N4 (4nm class) and contained 80 billion transistors — 2.6× the transistor count of the Ampere A100 (54 billion on Samsung 8nm). H100 SXM5 (the high-performance server module) shipped with 80 GB HBM3 at 3.35 TB/s memory bandwidth (vs A100’s 80 GB HBM2e at 2.0 TB/s), and delivered approximately 3.35 TFLOPS BF16/FP16 dense training throughput (vs 312 TFLOPS for A100) — roughly a 3× improvement in raw training arithmetic. The 4th-generation Tensor Cores added FP8 support, and the Transformer Engine (a hardware-software co-design) automatically selected between FP8 and FP16 precision on a per-layer, per-tensor basis during transformer training — using FP8 where precision loss was acceptable (typically activations and gradients after normalization) and FP16 where it was critical (typically weights and output projections), without requiring model developers to manually annotate precision choices. Read more
Valve Releases the Steam Deck
Published:
Valve began shipping Steam Deck units on February 25, 2022, fulfilling reservations placed starting July 2021 in order of the $5 deposit queue (over 100,000 units were pre-ordered in the first 48 hours). The hardware was built around AMD’s “Van Gogh” APU (internally codenamed “Aerith”): a Zen 2 quad-core/8-thread CPU (2.4–3.5 GHz) paired with an RDNA 2 GPU with 8 compute units (1.0–1.6 GHz, 1.6 TFLOPS FP32), 16 GB LPDDR5 RAM in a unified memory architecture, and a 40 Whr battery providing roughly 2–8 hours depending on the game’s GPU load. Three models shipped: 64 GB eMMC at $399, 256 GB NVMe PCIe 3.0 at $529, and 512 GB NVMe at $649. The 7-inch 1280×800 IPS display ran at 60 Hz with 400 nits brightness. All models included the same APU and memory — only storage and screen coating differed. Read more
Microsoft Announces Its Plan to Acquire Activision Blizzard
Published:
Microsoft announced the acquisition of Activision Blizzard on January 18, 2022 for $68.7 billion in cash — $95 per share, a 45% premium over Activision’s closing price on January 14. If completed, it would be the largest acquisition in Microsoft’s history and the largest in gaming history, surpassing Take-Two’s $12.7 billion acquisition of Zynga (also announced January 2022). Activision Blizzard’s portfolio included Call of Duty (the best-selling annual game franchise globally), World of Warcraft, Diablo, Overwatch, Candy Crush (King subsidiary), Hearthstone, and StarCraft. CEO Bobby Kotick, who was facing shareholder pressure over workplace misconduct allegations at Activision Blizzard, would eventually exit after closing. Read more
2021
FAQ Of Student Software Engineers
Published:
Students who want to become software engineers ask many of the same questions. Here are the ones I hear most often, answered as directly as I can. Read more
Log4Shell Exposes the Risk of Ubiquitous Dependencies
Published:
The Log4Shell vulnerability was disclosed in December 2021 in Log4j, a Java logging library embedded in an enormous number of applications. The flaw became critical because a specially crafted string could trigger JNDI lookups that, in vulnerable configurations, led to remote code execution. Read more
AWS Announces Graviton3
Published:
Amazon Web Services announced Graviton3 at re:Invent 2021 (December 1, 2021, Las Vegas) as its third-generation custom Arm server processor. The chip used 64 Arm Neoverse V1 cores (a high-performance variant of the Armv8.4 ISA, compared to the Neoverse N1 cores in Graviton2) manufactured on TSMC’s N5 (5nm) process versus Graviton2’s TSMC 7nm. AWS claimed 25% better compute performance, 2× better floating-point performance, 2× better cryptography performance, and 3× better machine learning performance compared to Graviton2 (the last comparison due to Graviton3’s support for BF16 and INT8 matrix multiply instructions via Arm’s Scalable Vector Extension). Graviton3 was also the first AWS-designed chip with DDR5-4800 memory support — a full year before DDR5 became widely available in consumer desktop platforms — yielding 50% higher memory bandwidth than Graviton2’s DDR4. AWS claimed 60% better energy efficiency versus comparable x86-based EC2 instances in the same workload category. Read more
Remembering Atatürk on November 10, 2021
Published:
Every November 10, Turkiye remembers Mustafa Kemal Ataturk. Read more
Windows 11 Is Released
Published:
Windows 11 released on October 5, 2021 as a free upgrade for eligible Windows 10 PCs, with a phased rollout that Microsoft controlled through hardware compatibility scoring — PCs meeting all requirements received a notification over weeks rather than all at once. The visual design replaced Windows 10’s left-anchored Start menu with a centered taskbar and Start button (repositionable to the left via settings), added rounded window corners, a redesigned Snap Layouts interface (Windows key + Z showing snap zone presets for 2–4 windows), and a new Widgets panel (news, weather, and calendar tiles). The Microsoft Store was rebuilt to accept Win32 applications (not only UWP apps), and Microsoft added DirectStorage API support — the same technology in Xbox Series X/S that allows the GPU to decompress NVMe SSD data directly, bypassing CPU and reducing game load times. Auto HDR automatically upgraded DirectX 11 and DirectX 12 games to use HDR color on supported displays. Read more
Welcome to Our Department
Published:
Welcome to Computer Engineering in 2021. Read more
Apple Introduces the A15 Bionic
Published:
Apple announced the A15 Bionic on September 14, 2021 with the iPhone 13 lineup: iPhone 13 ($799), 13 Mini ($699), 13 Pro ($999), and 13 Pro Max ($1,099). The A15 was manufactured on TSMC’s enhanced 5nm process (N5P), containing 15 billion transistors compared to the A14’s 11.8 billion. The CPU used 2 Avalanche high-performance cores and 4 Blizzard efficiency cores — Apple claimed 50% faster CPU performance than competitors at the time (targeting Qualcomm’s Snapdragon 888, which used Samsung 5nm and measured roughly 30–35% behind the A15 in single-thread benchmarks). The 16-core Neural Engine delivered 15.8 TOPS, up from the A14’s 11 TOPS, supporting the new Cinematic Mode video feature. Read more
Grady Booch, A Pioneer Of Computing Technology
Published:
Grady Booch is one of the most influential people in the history of software engineering, but his contributions are sometimes reduced to “he helped create UML,” which understates both what he built and what he argued for. I want to give his work more space. Read more
Why Choose Computer Engineering in 2021?
Published:
In 2021, it is difficult to find an important technology that does not involve computing. Cloud platforms run global services, artificial intelligence is entering more products, and the global semiconductor shortage has reminded everyone that software still depends on physical hardware. Read more
The Metaverse Is Still More Question Than Answer
Published:
The word “metaverse” comes from Neal Stephenson’s 1992 novel Snow Crash, where it describes a persistent virtual world accessible through goggles — a shared space where people can walk, talk, and conduct business as avatars. The term has been sitting in technology culture since then, invoked occasionally, not quite arriving. Read more
Apple Announces CSAM Detection Plans
Published:
Apple announced plans in 2021 for child-safety features that included on-device detection of known child sexual abuse material before images were uploaded to iCloud Photos. The announcement immediately became one of the most difficult privacy debates in consumer technology. Read more
Some Of Popular Java Developers
Published:
Java has existed since 1995, and in 2021 it is releasing a new version every six months, with Java 17 arriving later this year as the next long-term support release. The language has changed significantly over that period — records, sealed classes, pattern matching, text blocks, and a new approach to concurrency through Project Loom are all reshaping what Java code looks like. Behind that evolution are people who made specific technical decisions that influenced how millions of developers think and write code. Read more
Steam Deck Is Announced
Published:
Valve announced the Steam Deck on July 15, 2021 and opened reservations the following day. The handheld PC came in three storage configurations: $399 (64 GB eMMC), $529 (256 GB NVMe), and $649 (512 GB NVMe). The device used a custom AMD APU — codenamed Aerith — with four Zen 2 CPU cores at up to 3.5 GHz, RDNA 2 graphics with 8 compute units, and 16 GB of unified LPDDR5 memory. The 7-inch 1280×800 touchscreen was rated at 60 Hz. First units shipped to customers beginning February 25, 2022. Read more
The Gaming PC I Would Build in 2021
Published:
If I were building a gaming PC on July 1, 2021, this is the machine I would actually want. I am not simply choosing the most expensive part in every category. The priority is gaming performance, then reliability, cooling and enough headroom to keep the system useful for years. Read more
Windows 11 Is Announced
Published:
Microsoft announced Windows 11 at a dedicated event on June 24, 2021, led by Chief Product Officer Panos Panay. The announcement followed weeks of leaks — a build of Windows 11 had circulated online on June 15, 2021, confirming the visual redesign before the official event. The user interface changes included a centered taskbar and Start button (repositionable to the left in Settings), a new Start menu showing pinned apps and recent documents (but removing live tiles entirely), redesigned context menus and dialog boxes with rounded corners, a Widgets panel (news, weather, sports, calendar pulled from Microsoft services), and Snap Layouts (hovering over a window’s maximize button revealed 2–4 zone layout presets). The Microsoft Store was redesigned to allow Win32 application submission — developers could submit traditional desktop software, not only UWP/MSIX apps — addressing one of developers’ long-standing grievances with the previous Store. DirectStorage (a Windows API allowing the GPU to decompress NVMe data directly, bypassing CPU and system RAM for asset loading) was announced as a Windows 11 feature, though it was later backported to Windows 10 as well. Read more
Revolutionizing The Artificial Intelligence Landscape
Published:
Artificial intelligence in 2021 is surrounded by excitement that sometimes makes it difficult to see clearly. The same conversation that includes genuine technical milestones also includes cryptocurrency speculation, NFT marketplaces, and metaverse announcements. The noise makes it harder to identify what is real and durable. Read more
Google Announces LaMDA
Published:
Google introduced LaMDA (Language Model for Dialogue Applications) at Google I/O on May 18, 2021 as a language model trained specifically for multi-turn, open-ended conversation. The model had 137 billion parameters and was trained on 1.56 trillion words of public web documents and dialogue data. Unlike Google’s earlier neural conversation research, LaMDA was designed to maintain coherent context across many conversational turns rather than responding to isolated queries. Read more
Apple Introduces the M1 iMac and New iPad Pro
Published:
Apple introduced a redesigned 24-inch iMac and a new iPad Pro in April 2021, both using the M1 processor announced at the Mac event in November 2020. The iMac, priced from $1,299, replaced the 21.5-inch and 27-inch Intel models with a single M1 design just 11.5 mm thick — thinner than many smartphones — in seven colors. It included a 24-inch 4.5K Retina display, a 1080p front camera, and a six-speaker sound system. The M1 Mac mini and MacBook Air had already launched; adding the iMac and iPad Pro showed the chip could span from compact desktops to tablets with the same architecture. Read more
Microsoft Introduces Mesh for Mixed Reality
Published:
Microsoft announced Mesh at its Ignite conference on March 2, 2021 as a cloud platform for shared, real-time mixed-reality experiences. The initial demonstration ran on HoloLens 2 ($3,500), showing multiple people in different locations appearing as holographic representations — either realistic avatars built from photogrammetry or simplified cartoon avatars — in a shared three-dimensional workspace where they could manipulate 3D models, annotate CAD files, and communicate spatially. The Azure backend handled spatial anchoring (tying virtual content to real-world coordinates), identity and access control, real-time state synchronization, and cross-device rendering adaptation. Read more
NVIDIA Launches the GeForce RTX 3060
Published:
NVIDIA released the GeForce RTX 3060 on February 25, 2021 at an MSRP of $329. It used the GA106 die on Samsung’s 8nm process with 3,584 CUDA cores, 13 Shader TFLOPS, and — unusually for its price tier — 12 GB of GDDR6 memory on a 192-bit bus. The RTX 3060 Ti ($399, December 2020) used a cut-down GA104 die with 8 GB; the standard RTX 3060’s larger memory allocation was a deliberate decision to make it attractive for content creation workloads where VRAM capacity matters as much as throughput. Read more
Quotes From Famous Computer Scientists
Published:
Computer science has produced many memorable sentences. The best of them are useful not because they settle arguments, but because they reframe questions in ways that make a problem easier to think about. Here are several I return to often, with what I actually think they mean. Read more
Apple Silicon And The Return Of Processor Diversity
Published:
Apple announced its transition from Intel processors to its own Apple Silicon at WWDC in June 2020 and shipped the first M1 Macs — the MacBook Air, 13-inch MacBook Pro, and Mac mini — in November 2020. By January 2021, reviews and benchmarks have had time to accumulate, and the results are unusual enough to warrant serious attention. Read more
WhatsApp Privacy Changes Trigger a Messaging-App Migration
Published:
WhatsApp began notifying its 2 billion users on January 4, 2021 about an updated privacy policy requiring acceptance by February 8, 2021 to continue using the app (a deadline later extended to May 15, 2021 after widespread backlash). The notification was poorly worded and generated mass confusion: many users interpreted it as WhatsApp announcing that it would begin sharing the content of private messages with Facebook for advertising purposes. Social media posts warning friends to switch to Signal or Telegram spread rapidly, often citing this misunderstanding. WhatsApp published blog posts attempting to clarify that message content remained end-to-end encrypted and invisible to WhatsApp’s servers — true since WhatsApp had implemented the Signal Protocol end-to-end encryption for all personal chats in April 2016. WhatsApp had been acquired by Facebook for approximately $19 billion in February 2014, and co-founders Jan Koum and Brian Acton (who left Facebook in 2018; Acton donated $50 million to found the Signal Foundation in January 2018 alongside Moxie Marlinspike) had originally committed to maintaining WhatsApp’s privacy properties. The January 2021 policy update was primarily about WhatsApp’s Business API: companies using WhatsApp to communicate with customers (for order confirmations, customer service, etc.) would be able to store those business-customer conversations on Facebook’s infrastructure and potentially use that data for Facebook advertising targeting. Read more
2020
SolarWinds Reveals a Major Software Supply Chain Attack
Published:
FireEye disclosed on December 8, 2020 that it had been breached and its Red Team tools stolen; investigating that intrusion led to the discovery of a much larger campaign. On December 13, 2020, SolarWinds and the U.S. government simultaneously disclosed that SolarWinds Orion — a widely deployed IT monitoring platform used by the U.S. government, Fortune 500 companies, and critical infrastructure — had been compromised at its build stage. Attackers (attributed to APT29, Russia’s SVR intelligence service) had modified SolarWinds’ Orion software build process between March and June 2020, inserting a backdoor component called SUNBURST into legitimate, digitally signed Orion updates (versions 2019.4 through 2020.2.1). Approximately 18,000 organizations downloaded the malicious updates; among SolarWinds’ roughly 33,000 customers at the time, the installed-base included the U.S. Treasury, Department of Commerce, Department of Homeland Security, State Department, NIH, and parts of the Department of Defense. Read more
What Is Quantum Computing?
Published:
Quantum computing is often described as a technology that will make computers unbelievably fast. That description is too simple and somewhat misleading. A quantum computer is not a faster version of an ordinary computer for every task. It is a different model of computation that can offer major advantages for specific kinds of problems — and very little advantage for most of the things ordinary computers do every day. Read more
Remote Work Is Changing How Software Teams Communicate
Published:
The COVID-19 pandemic has forced many software teams to work remotely for much longer than they expected. Software development was already more remote-friendly than many professions, but working from home occasionally is very different from building an entire working culture around distance. Read more
Apple Releases the First M1 Macs
Published:
Apple released the MacBook Air, 13-inch MacBook Pro, and Mac mini with M1 on November 17, 2020. The MacBook Air started at $999, the MacBook Pro at $1,299, and the Mac mini at $699 — unchanged prices from their Intel predecessors. Apple had announced the transition from Intel to Apple silicon at WWDC in June 2020, promising a two-year transition window; the first M1 Macs arrived only five months later. Read more
Remembering Atatürk on November 10, 2020
Published:
Every November 10, Turkiye pauses to remember Mustafa Kemal Ataturk. Read more
AMD Announces the Zen 3 Architecture
Published:
AMD announced Zen 3 on October 8, 2020 with the Ryzen 5000 desktop processor family, going on sale November 5, 2020: Ryzen 9 5950X (16 cores/32 threads, $799), Ryzen 9 5900X (12c/24t, $549), Ryzen 7 5800X (8c/16t, $449), and Ryzen 5 5600X (6c/12t, $299). The foundational architectural change was the redesign of the Core Complex (CCX): Zen 2 divided its cores into two 4-core CCXes each with their own 16 MB L3 cache, so a thread on one CCX accessing data cached on the other CCX faced ~40 ns of cross-CCX latency. Zen 3 merged all 8 cores of a CCD (Core Chiplet Die) into one unified CCX sharing a single 32 MB L3 cache, cutting cross-core L3 latency from ~40 ns to ~10 ns. The impact was largest in gaming workloads where the game engine, physics, audio, and rendering threads frequently accessed shared game state: Ryzen 5000 regularly outperformed the competing Ryzen 3000 by 15–20% in gaming despite using the same TSMC 7nm (N7P slightly refined) manufacturing process, demonstrating that the architectural bottleneck had been the CCX boundary latency rather than clock speed or manufacturing node. Read more
Smart Antennas
Published:
A traditional antenna has a radiation pattern largely fixed by its physical shape and orientation. If you want to direct signal in a particular direction, you physically point the antenna that way. Smart antennas change this. By combining multiple antenna elements with signal processing, a system can steer its effective beam electronically — no moving parts required. In 2020, with 5G networks rolling out and Wi-Fi 6 becoming standard in new hardware, smart antenna techniques are no longer only a research topic. They are inside the devices people are using today. Read more
Welcome to Our Department
Published:
If you are beginning Computer Engineering in 2020, your first year is starting in an unusual time. Read more
NVIDIA Announces the RTX 30 Series
Published:
NVIDIA CEO Jensen Huang announced the GeForce RTX 30 series on September 1, 2020 in an online event, with the RTX 3080 ($699), RTX 3090 ($1,499), and later RTX 3070 ($499, October 29) going on sale September 17, October 24, and October 29 respectively. The Ampere architecture (GA102/GA104 dies manufactured on Samsung 8nm, not TSMC as NVIDIA had used previously) reorganized the streaming multiprocessor design so that each SM contained twice the FP32 CUDA core count of Turing by adding a second FP32 data path that could also execute INT32 simultaneously — effectively doubling FP32 throughput without doubling the SM count, at the cost of the INT32-FP32 overlap being an architecture-specific optimization rather than general-purpose parallelism. The RTX 3080 shipped with 8,704 CUDA cores and 10 GB GDDR6X at 320 GB/s bandwidth; the RTX 3090 shipped with 10,496 CUDA cores and 24 GB GDDR6X at 936 GB/s. The GDDR6X memory (developed by Micron) used PAM4 (Pulse-Amplitude Modulation 4) signaling at 19.5 Gbps per pin versus GDDR6’s 16 Gbps, adding approximately 21% bandwidth at equivalent pin count and frequency. Compared to the RTX 2080 Ti ($999, Turing’s flagship, 4,352 CUDA cores, 11 GB GDDR6, 14.2 TFLOPS FP32), the RTX 3080 offered approximately 29.8 TFLOPS FP32 — more than 2× the shader throughput — at $699. Read more
Why Choose Computer Engineering in 2020?
Published:
The year 2020 has made the importance of digital infrastructure unusually visible. Work, education, communication, shopping, entertainment, and many public services have moved online at a scale few people expected, and in many cases at a speed that revealed which systems were ready and which were not. Read more
Epic Games Challenges Mobile App Store Policies
Published:
On August 13, 2020, Epic Games updated Fortnite — one of the world’s most widely played games, with approximately 350 million registered accounts — to offer a direct payment option that bypassed Apple’s in-app purchase system. Epic’s “Epic Direct Payment” offered Fortnite’s in-game currency (V-Bucks) at a 20 percent discount compared to the price charged through Apple’s IAP system ($7.99 versus $9.99 for 1,000 V-Bucks), where Apple’s 30 percent commission had been built into the price. Apple’s App Store Review Guidelines required all digital content purchases in iOS apps to use Apple’s in-app purchase system, with Apple receiving 30 percent (15 percent for subscriptions after the first year, for developers earning under $1M through Apple’s Small Business Program). Apple removed Fortnite from the App Store within hours of detecting the policy violation; Google removed it from Google Play the same day (though Fortnite had already been available via direct download/sideload on Android). Epic had prepared for the removal: simultaneously with the update, Epic filed a lawsuit against Apple in the US District Court for the Northern District of California, and a separate lawsuit against Google, both alleging antitrust violations. Epic also published a parody video titled “Nineteen Eighty-Fortnite” — a direct parody of Apple’s famous “1984” Super Bowl advertisement — with Fortnite characters fighting an “Apple Corporation” controlling citizens, reversing the iconic imagery to cast Apple as the monopoly rather than the liberator. Read more
GPT-3 Demonstrates the Scale of Large Language Models
Published:
OpenAI published the GPT-3 paper (“Language Models are Few-Shot Learners”) on arXiv May 28, 2020, and opened a private API beta in June 2020. The model contained 175 billion parameters — 117 times larger than GPT-2’s 1.5 billion — trained on approximately 300 billion tokens drawn from a filtered version of Common Crawl (60%), WebText2 (the web corpus from GPT-2, 22%), two book datasets (16%), and English Wikipedia (3%). The transformer architecture used 96 layers, 96 attention heads, and 12,288-dimensional embeddings, requiring approximately 800 GB to store in 16-bit floating-point precision. Estimated training cost was roughly $4.6 million at then-current V100 cloud GPU prices, requiring several thousand GPUs for months. Read more
The Gaming PC I Would Build in 2020
Published:
If I were building a gaming PC on July 1, 2020, this is the machine I would actually want. I am not simply choosing the most expensive part in every category. The priority is gaming performance, then reliability, cooling and enough headroom to keep the system useful for years. Read more
Apple Announces the Move From Intel to Apple Silicon
Published:
Apple announced at WWDC 2020 (June 22, 2020) that Macs would transition from Intel x86-64 processors to Apple-designed ARM-based silicon over a two-year period — the Mac’s third major CPU architecture change after Motorola 68k to PowerPC (1994) and PowerPC to Intel (2005–2006). Apple CEO Tim Cook announced the transition alongside a Developer Transition Kit (a Mac mini with the A12Z Bionic chip from the iPad Pro, loaned to registered developers for $500) to allow software porting before consumer hardware arrived. Apple emphasized two compatibility layers: Universal 2 binaries (a fat binary format containing both ARM64 and x86-64 code in one application package, allowing one download to run natively on both architectures) and Rosetta 2 (a successor to the original Rosetta used during the PowerPC-to-Intel transition, now performing ahead-of-time translation of x86-64 code to ARM64 at app install time rather than interpreting at runtime, giving dramatically faster performance than runtime JIT translation). Virtualization of ARM Linux and Windows for ARM was also supported. Read more
Software Bug Of Therac-25
Published:
The Therac-25 is one of the most studied cases in software-engineering history because the consequences of the bugs it contained were not degraded performance or corrupted data. Patients received radiation doses hundreds of times above therapeutic levels. At least six people were seriously injured. Three died. Read more
Software Bug Of Pathfinder
Published:
Mars Pathfinder landed successfully on July 4, 1997 — the first Mars lander since the Viking missions of 1976. The entry, descent, and landing were flawless. Within days, the spacecraft and its small rover Sojourner were returning data and images. Then the resets began. Read more
Microsoft Releases Windows Terminal 1.0
Published:
Microsoft released Windows Terminal 1.0 in May 2020, roughly a year after its first public preview at Microsoft Build 2019. The application replaced decades-old infrastructure: Command Prompt ran on conhost.exe, a host that dated to Windows NT and had no support for modern text rendering, multiple tabs, or proper Unicode. Read more
Apple and Google Announce Exposure Notification Technology
Published:
Apple and Google announced the Exposure Notification API on April 10, 2020, seventeen days after the World Health Organization declared COVID-19 a pandemic, in an unprecedented collaboration between the two companies whose competing mobile operating systems controlled over 99 percent of the global smartphone market. The announcement came as governments worldwide were evaluating contact-tracing applications that varied widely in their privacy properties — Singapore’s TraceTogether (March 20, 2020) used a centralized Bluetooth proximity log architecture where the health ministry held contact graphs, raising concerns about government surveillance. Several European countries were building centralized systems as well. The Apple-Google design deliberately chose a decentralized architecture: each phone generated a daily Temporary Exposure Key (TEK) using a cryptographically secure random source, then derived short-lived Rolling Proximity Identifiers (RPIs) from the TEK every 10-20 minutes using HMAC-SHA256. Phones broadcast their current RPI over Bluetooth Low Energy advertising packets while simultaneously logging the RPIs they observed from nearby devices, along with signal strength (RSSI) as a proxy for distance. No names, phone numbers, GPS locations, or account identifiers were broadcast or logged — only the cryptographic identifiers, which were unlinkable to individuals without knowledge of the originating TEK. Read more
Blogging For Your Career And Life
Published:
Writing in public is different from writing for yourself, and that difference is where the value comes from. Read more
Remote Work Tools Become Critical Infrastructure
Published:
The World Health Organization declared COVID-19 a pandemic on March 11, 2020, and within days, shelter-in-place orders in major cities (San Francisco Bay Area March 16, New York City March 22) forced millions of knowledge workers, students, and teachers to begin working and learning from home simultaneously. The speed and scale of the transition placed unprecedented demand on digital collaboration infrastructure. Zoom Video Communications had approximately 10 million daily meeting participants in December 2019; by April 2020 that figure had reached 300 million — a 30× increase in four months — representing a growth trajectory that would normally take 5-7 years at expected rates. Zoom’s architecture (WebRTC-based media with their own media encoding, meeting rooms hosted on Zoom’s servers) had to scale media processing, recording storage, and global anycast network routing in weeks rather than years. Microsoft Teams grew from 32 million daily active users in March 2020 to 75 million in April 2020. Slack, which had been primarily a text-based channel tool, saw concurrent user counts surge to 12.5 million. These platforms faced not only scale challenges but reliability requirements that consumer-grade services had never faced: a dropped video call in a business meeting, a school lesson that couldn’t connect, or a hospital telemedicine consultation that failed had real costs that casual use did not. Read more
Samsung Unveils the Galaxy Z Flip
Published:
Samsung unveiled the Galaxy Z Flip at Samsung Unpacked on February 11, 2020, pricing it at $1,380 — significantly less than the Galaxy Fold ($1,980) and Motorola Razr 2019 ($1,499), its direct competitors in the emerging foldable clamshell category. The Z Flip’s 6.7-inch Infinity Flex Dynamic AMOLED display folded vertically to a compact 73.6×87.4 mm square, compared to the Motorola Razr’s similar form factor. The critical materials difference was Samsung’s use of Ultra-Thin Glass (UTG) — approximately 30 μm thick glass from Schott AG with a polymer backing — as the display substrate, replacing the plastic film (polyimide) used in the Galaxy Fold and Motorola Razr. The UTG provided a harder surface that resisted fingernail scratching and felt more like traditional glass to the touch, though it was still flexible enough to fold. A 1.1-inch 300×116 pixel cover display showed time, notifications, and music controls when closed. The SoC was the Snapdragon 855+ (not the then-current Snapdragon 865), with 8 GB RAM and 256 GB UFS 3.0 storage, running Android 10 with One UI 2.1. Read more
The Story Of Computing
Published:
The story of computing is also a story about human limits. For thousands of years, people have built tools to extend what the mind alone can do — to count faster, to record more, to calculate at scales impossible by hand. An abacus, a mechanical calculator, a punched card, and a modern processor look completely different but serve the same basic impulse: move some mental work outside the human mind. Read more
Microsoft Ends Support for Windows 7
Published:
Microsoft ended extended support for Windows 7 on January 14, 2020 — a milestone that the company had announced and counted down for years. Windows 7 had shipped October 22, 2009, giving it a 10-year-3-month support lifecycle: mainstream support ended January 13, 2015 (after which no new features or non-security fixes were added), and extended support (security patches only) ran for five additional years until January 14, 2020. At end-of-life, Windows 7 still accounted for approximately 26–33% of all active Windows PCs globally depending on the survey methodology — tens of millions of machines ranging from home desktops to industrial controllers to hospital workstations — making it one of the largest simultaneous end-of-support events in consumer software history. The healthcare sector illustrated the most severe risks: WannaCry (May 12, 2017) had spread via the EternalBlue exploit (a leaked NSA tool targeting the SMBv1 protocol), affecting over 300,000 computers running Windows 7 and Windows XP across 150 countries, shutting down approximately one-third of NHS England trusts and forcing cancellation of 19,000 medical appointments. The SMB vulnerability had been patched (MS17-010, March 14, 2017) but hospitals running Windows 7 systems hadn’t applied the patch due to lengthy medical device recertification cycles. Read more
2019
Wi-Fi 6 Hardware Becomes Widely Available
Published:
The Wi-Fi Alliance certified the 802.11ax standard as “Wi-Fi 6” in September 2019, after three years of IEEE standardization work beginning with the 802.11ax task group formation in 2014. The standard’s peak theoretical throughput reached 9.6 Gbps (vs 3.5 Gbps for Wi-Fi 5/802.11ac) using a combination of 1024-QAM modulation (25% more data per symbol than Wi-Fi 5’s 256-QAM) and wider 160 MHz channels, but the more consequential improvements were in multi-device efficiency. Intel shipped the Wi-Fi 6-certified AX200 M.2 adapter in July 2019 (the first mainstream PC Wi-Fi 6 card), Samsung’s Galaxy S10 (March 2019) was the first Wi-Fi 6-certified smartphone, and Apple’s iPhone 11 (September 2019) added Wi-Fi 6 — establishing it in flagship hardware before Christmas 2019. Router availability was wider: the Netgear Nighthawk AX12 (RAX200) launched at ~$500, Asus ROG Rapture GT-AX11000 at ~$450, and TP-Link Archer AX6000 at ~$300, all in 2019. Read more
GitHub Announces Actions for General Availability
Published:
GitHub Actions reached general availability on November 13, 2019, extending a beta period that began in August 2018. The feature added a built-in CI/CD system to GitHub repositories, letting developers define automated workflows in YAML files stored at .github/workflows/. A workflow specified triggers (push to a branch, pull request opened, release created, scheduled cron, or manual dispatch), jobs (each running in an isolated virtual machine), and steps (shell commands or reusable “actions” from the GitHub Marketplace). GitHub provided hosted runners on Ubuntu (2-core, 7GB RAM), Windows, and macOS at no cost for public repositories, with paid minutes for private repositories starting at $0.008 per minute for Linux. Read more
Remembering Atatürk on November 10, 2019
Published:
Every November 10, Turkiye remembers Mustafa Kemal Ataturk. Read more
Google Claims Quantum Supremacy With Sycamore
Published:
Google published a paper in Nature on October 23, 2019 claiming that its 53-qubit Sycamore processor had achieved “quantum supremacy” — completing a specific computational task that would take the world’s fastest classical supercomputers an astronomically long time to match. The paper had leaked two days early when a draft appeared on NASA’s website (NASA collaborates with Google on quantum computing research) on September 20, 2019, generating widespread coverage before the official publication. The task chosen was Random Circuit Sampling (RCS): apply a random sequence of quantum logic gates to the 53 qubits (one of Sycamore’s 54 qubits was non-functional), then measure all qubits to obtain a 53-bit output string. Repeat this millions of times. The output distribution of a random quantum circuit is theoretically very hard to classically simulate because the quantum state evolves through a superposition of 2^53 (approximately 9 quadrillion) simultaneous values during the computation. Google estimated that the Summit supercomputer at Oak Ridge National Laboratory — the world’s fastest at 200 petaflops at the time — would require approximately 10,000 years to sample from the same distribution that Sycamore sampled in 200 seconds. Sycamore operated at approximately 15 millikelvin (colder than the cosmic microwave background at ~2.7 K) using superconducting transmon qubits, achieving a two-qubit gate error rate of approximately 0.6 percent and a single-qubit gate error rate below 0.1 percent — among the lowest error rates achieved in a superconducting quantum processor at that scale. Read more
Welcome to Our Department
Published:
If you are starting Computer Engineering in 2019, you are entering a discipline that touches almost every industry. Read more
Apple Introduces the A13 Bionic
Published:
Apple introduced the A13 Bionic with the iPhone 11, 11 Pro, and 11 Pro Max in September 2019. The chip used TSMC’s second-generation 7nm process (N7P) with around 8.5 billion transistors, up from 6.9 billion in the A12. Two high-performance CPU cores were rated at 20% faster than the A12’s, and four efficiency cores used 40% less power for the same workload. Read more
Why Choose Computer Engineering in 2019?
Published:
In 2019, software is becoming an invisible layer inside almost everything we use. Phones, vehicles, payment systems, factories, streaming services, medical devices, and public infrastructure all depend on computing. Studying Computer Engineering means learning how this layer is designed. Read more
OpenAI Releases GPT-2 in Stages
Published:
OpenAI published its research paper on GPT-2 in February 2019 and released the model in stages across the year: the 117M parameter version in February, the 345M version in May, the 774M version in August, and the full 1.558 billion parameter model in November 2019. The staged release was explicitly justified by OpenAI’s concerns about potential misuse for generating disinformation and spam at scale — the first time a major AI lab publicly withheld model weights for safety rather than commercial reasons. Read more
Daily Life Of A Software Engineer
Published:
The daily life of a software engineer is usually less dramatic than movies make programming look. There are no montages of cascading green characters or last-minute keystrokes that save cities. Most days are a mixture of reading, testing, discussing, debugging, and — somewhere in the middle — writing code. Read more
Book Suggestions For Software Engineers
Published:
Software engineers spend a lot of time learning from screens — documentation, source code, issue trackers, tutorials, and online discussions. Books offer a different pace. They allow one author to develop an idea fully, without competing with notifications and search results, and without the pressure to be immediately practical that tutorials often carry. Read more
Facebook Announces More Details of Libra’s Technical Design
Published:
Facebook published the Libra whitepaper on June 18, 2019, through its Calibra subsidiary, proposing a global digital currency backed by a reserve basket of low-volatility assets — approximately 70% USD, 26% EUR, and smaller allocations to JPY and GBP government securities and money market funds. The Libra Association, incorporated in Geneva, Switzerland, launched with 28 founding members (later expanded to 100 targeted) including Visa, Mastercard, PayPal, Stripe, Uber, Lyft, and Spotify, each contributing $10 million. Facebook’s own subsidiary Calibra (later Novi) would build the consumer wallet, while the Association would govern the protocol. Read more
The Gaming PC I Would Build in 2019
Published:
If I were building a gaming PC on July 1, 2019, this is the machine I would actually want. I am not simply choosing the most expensive part in every category. The priority is gaming performance, then reliability, cooling and enough headroom to keep the system useful for years. Read more
Raspberry Pi 4 Is Released
Published:
The Raspberry Pi 4 Model B was released on June 24, 2019, in three RAM variants: 1GB ($35), 2GB ($45), and 4GB ($55) LPDDR4 — replacing the Raspberry Pi 3 B+ (launched March 2018) as the current flagship. The BCM2711 SoC used a quad-core ARM Cortex-A72 at 1.5 GHz, an out-of-order execution microarchitecture significantly more capable than the Pi 3’s in-order Cortex-A53 at 1.4 GHz. Sysbench CPU benchmarks showed roughly 2–3× improvement in integer throughput; the out-of-order pipeline allowed the Cortex-A72 to extract instruction-level parallelism that the A53 executed strictly in sequence. Read more
Google Introduces the Pixel 3a and On-Device AI Features
Published:
Google announced the Pixel 3a and Pixel 3a XL at Google I/O on May 7, 2019. The Pixel 3a launched at $399 (5.6-inch) and the 3a XL at $479 (6-inch) — less than half the Pixel 3’s $799 starting price. The phones used a Snapdragon 670 (Pixel 3a) rather than the Pixel 3’s Snapdragon 845, with a plastic back instead of glass, no wireless charging, and no front stereo speakers. They retained the same 12.2MP rear camera sensor and f/1.8 lens as the Pixel 3, the same pixel binning algorithm, and Google’s full HDR+ and Night Sight computational photography pipeline running in software on the Snapdragon 670’s DSP. Read more
6TiSCH Protocol
Published:
Low-power wireless devices face a fundamental tension. They need reliable communication, but their batteries may need to last months or years in places where replacing them is difficult or expensive. Keeping a radio active continuously is not an option. 6TiSCH is a protocol architecture designed to resolve this tension — allowing constrained devices to participate in Internet-style networks while managing energy carefully. Read more
Microsoft Releases Chromium-Based Edge Preview Builds
Published:
Microsoft announced the new Chromium-based Edge on December 6, 2018, and released the first Canary and Dev channel builds on January 20, 2019, followed by the Beta channel on April 8, 2019. The decision meant abandoning EdgeHTML — the proprietary rendering engine built from the remnants of Trident (Internet Explorer’s engine) and introduced with Windows 10 in 2015 — along with the Chakra JavaScript engine. At the time of the announcement, Edge held approximately 4% global browser market share versus Chrome’s ~63%, Internet Explorer’s ~7%, and Firefox’s ~10% (StatCounter data, Q4 2018). The compatibility situation was the primary driver: websites routinely displayed “Works best in Chrome” banners when accessed from Edge, and developer tools in Chromium’s DevTools were far more capable than Edge’s F12 tools. The extension ecosystem was another gap — the Chrome Web Store had tens of thousands of extensions while Edge’s own extension library had a few hundred. Read more
Google Introduces Stadia
Published:
Google announced Stadia at GDC in March 2019, promising a cloud gaming service that would stream games running on data-center hardware directly to Chrome browsers, Chromecast Ultras, and Pixel phones. The platform launched November 19, 2019 in 14 countries with a Founder’s Edition bundle ($129) containing a Stadia Controller, a Chromecast Ultra, and three months of Stadia Pro ($9.99/month), which offered 4K/60fps streaming and a rotating selection of free games. A free tier at 1080p/60fps was announced but not released until April 2020. Read more
Google Announces the End of Google Plus for Consumers
Published:
Google shut down the consumer version of Google+ on April 2, 2019. The shutdown had been announced October 8, 2018, when Google disclosed that a software bug introduced in March 2015 and patched in March 2018 had exposed the private profile fields of up to 500,000 Google+ accounts to third-party app developers through the People API — specifically, apps that were granted access to public profile data could also read private profile fields (name, email address, occupation, gender, age) that users had intended to share only with specific people. Google had not publicly disclosed this bug when it was discovered in March 2018 — citing that no evidence of misuse had been found — a decision criticized after The Wall Street Journal reported it in October 2018, suggesting Google’s non-disclosure was influenced by concerns about regulatory and reputational consequences following the Cambridge Analytica/Facebook scandal (March 2018). A second API bug was found in November 2018, affecting approximately 52.5 million users by exposing additional profile fields (name, email address, occupation, age) to apps that users had not explicitly authorized, accelerating the shutdown timeline from August 2019 to April 2, 2019. Read more
NVIDIA Introduces RTX Graphics to More Laptop Systems
Published:
NVIDIA announced GeForce RTX 20-series laptop GPUs at CES in January 2019, bringing the Turing architecture’s dedicated RT cores for ray tracing and Tensor cores for DLSS into mobile systems. The mobile lineup included the RTX 2060, 2070, and 2080, with most high-end configurations also available in Max-Q variants designed for thinner chassis. Read more
2018
Kubernetes 1.13 Strengthens Cloud-Native Infrastructure
Published:
Kubernetes 1.13 released on December 3, 2018, marking kubeadm — the official Kubernetes cluster bootstrapping tool — as generally available (GA) after nearly two years in alpha/beta. kubeadm init (initialize a control plane node) and kubeadm join (join a worker node to an existing cluster) provided a consistent, supported procedure for bringing up a production-grade Kubernetes cluster from bare OS without hiding the cluster behind a managed service abstraction. Prior to kubeadm GA, bootstrapping a cluster involved either running a vendor-managed installer (kops, Rancher, OpenShift) that made different decisions about network configuration and storage, or manually reading multi-page documentation to run etcd and the kube-apiserver, kube-scheduler, and kube-controller-manager as system services. kubeadm’s GA status meant distributions and tooling could declare a standard bootstrapping interface, enabling reproducible cluster setup across on-premises hardware, VMs, and bare metal. Read more
Raspberry Pi 3 Model A+ Is Released
Published:
The Raspberry Pi Foundation released the Raspberry Pi 3 Model A+ in November 2018 at $25, ten dollars less than the Model B+. It brought the processor and wireless capabilities of the Pi 3B+ — the fastest model in the third generation — into the smaller Model A form factor, which had always prioritized size and cost over maximum connectivity. Read more
Remembering Atatürk on November 10, 2018
Published:
Every November 10, Turkiye pauses at 09:05 to remember Mustafa Kemal Ataturk. Read more
Machine Learning Is Becoming A Normal Software Tool
Published:
Machine learning is no longer something I hear about only in research papers. TensorFlow was open-sourced in November 2015. Keras made it more approachable — and Google officially adopted Keras as TensorFlow’s high-level API in 2017. PyTorch, released by Facebook’s research lab in September 2016, gained significant traction in academic research through 2017 and 2018 because its dynamic computation graph made debugging more natural. scikit-learn has made classical machine-learning algorithms available in Python with a consistent API since 2011. Cloud platforms — AWS SageMaker (launched November 2017), Google Cloud ML Engine, Azure Machine Learning — now offer managed environments for training and deploying models without running your own infrastructure. Read more
IBM Announces Its Acquisition of Red Hat
Published:
IBM announced on October 28, 2018 that it would acquire Red Hat for $34 billion in cash — $190 per share, representing a 63 percent premium over Red Hat’s closing stock price on October 26 — in the largest software acquisition in history at that point, surpassing Microsoft’s $26 billion LinkedIn acquisition in 2016. Red Hat had been founded in 1993 by Marc Ewing and Bob Young and had built its business on a commercial support and subscription model for open-source software: customers paid not for the software itself (which remained freely available) but for Red Hat’s enterprise support lifecycle, certification testing, and legal indemnification. By fiscal year 2018, Red Hat reported approximately $3 billion in annual revenue, growing around 20 percent per year. IBM CEO Ginni Rometty framed the acquisition as the cornerstone of IBM’s pivot to hybrid cloud — the model in which enterprise workloads run partly on on-premises servers, partly on private clouds, and partly on public clouds (AWS, Azure, GCP), with Red Hat’s technology providing a consistent operating environment across all three. Red Hat Enterprise Linux (RHEL) had become the dominant enterprise Linux distribution, running the majority of Fortune 500 production workloads. Equally important was OpenShift, Red Hat’s Kubernetes distribution, which Red Hat had built up through its acquisition of CoreOS (February 2018, ~$250 million) — giving IBM a leading Kubernetes management platform at a moment when container orchestration was becoming the standard for cloud-native application deployment. Read more
Welcome to Our Department
Published:
Welcome to Computer Engineering in 2018. Read more
Apple Introduces the A12 Bionic
Published:
Apple announced the A12 Bionic on September 12, 2018 alongside the iPhone XS ($999), XS Max ($1099), and XR ($749 — releasing October 26). Manufactured on TSMC’s first-generation 7nm process (N7, not yet EUV-based) with approximately 6.9 billion transistors, the A12 was one of the first mass-production chips built at 7nm — Qualcomm’s Snapdragon 855 remained on 7nm TSMC but didn’t ship until December 2018, and Samsung’s 7nm processes weren’t available until 2019. The A12’s CPU used two Vortex high-performance cores (2.49 GHz) and four Tempest efficiency cores — Apple’s first chip with a heterogeneous big.LITTLE-style CPU arrangement, following ARM’s own design pattern — plus a four-core GPU that Apple claimed was 50% faster than the A11’s GPU. Read more
How Computer Computes?
Published:
A computer looks complicated from the outside, but its basic operation can be described with a simple cycle repeated billions of times per second. Understanding that cycle helps explain why software behaves the way it does — why some operations are fast, others slow, and what actually happens when code runs. Read more
Why Choose Computer Engineering in 2018?
Published:
Technology in 2018 is increasingly built around data. Companies collect it, applications process it, machine-learning systems learn from it, and users are becoming more aware — sometimes more alarmed — about how it is stored and protected. That makes Computer Engineering a particularly relevant field to study. Read more
NVIDIA Announces the Turing GPU Architecture
Published:
NVIDIA announced the Turing GPU architecture at SIGGRAPH on August 13, 2018, with the GeForce RTX 2080 Ti ($999/$1,199), RTX 2080 ($699/$799), and RTX 2070 ($499) going on sale September 20, 2018. Turing (TU102 die for RTX 2080 Ti, TU104 for RTX 2080) was manufactured on TSMC’s 12nm FinFET process and contained 18.6 billion transistors in the flagship TU102 — 54% more than Pascal’s GP102 (12 billion). The two defining new hardware blocks were RT Cores (first ever in a consumer GPU, one per SM) and 2nd-generation Tensor Cores (inherited from the Volta data-center GPU, brought to consumer hardware for the first time). RT Cores accelerated the most computationally expensive parts of ray tracing: BVH (Bounding Volume Hierarchy) traversal (the tree search to find which scene geometry might intersect a ray) and ray-primitive intersection testing (the geometric calculation to determine exact intersection point and surface normal). Without RT Cores, ray tracing required thousands of CUDA cores to evaluate these tests in software, consuming the majority of the compute budget for even one bounce of reflected light per pixel at 1080p resolution. With RT Cores offloading these two operations to fixed-function hardware, the CUDA cores could focus on shading and color calculation, allowing real-time ray-traced shadows, reflections, and ambient occlusion at playable frame rates for the first time. Read more
Popular Questions About Software
Published:
People who are not software developers often have remarkably good questions about how software works. Here are some of the ones I hear most often, along with honest answers. Read more
Google Chrome Begins Stronger HTTPS Warnings
Published:
Chrome 68 shipped on July 24, 2018 with a single widely noticed change: every HTTP page displayed a “Not secure” label in the address bar’s security chip, to the left of the URL. Previously, Chrome and other browsers had used a positive indicator model — HTTPS pages showed a green padlock icon with or without the domain name, while HTTP pages showed nothing or a neutral icon, creating the impression that HTTP was a safe baseline rather than an actively insecure choice. Google had been phasing in the change across multiple Chrome releases: Chrome 56 (January 2017) first added the “Not secure” label on HTTP pages containing password or credit card input fields; Chrome 62 (October 2017) extended the label to all HTTP pages with any form input and all HTTP pages in Incognito mode; Chrome 68 expanded it to all HTTP pages unconditionally. Chrome 70 (October 2018) made the label red-colored when users began typing into any HTTP page, adding visual urgency at the moment of risk. Google had announced this incremental plan in September 2016 as part of its public commitment to treating HTTPS as the baseline security assumption for the web. Read more
The Gaming PC I Would Build in 2018
Published:
If I were building a gaming PC on July 1, 2018, this is the machine I would actually want. I am not simply choosing the most expensive part in every category. The priority is gaming performance, then reliability, cooling and enough headroom to keep the system useful for years. Read more
GitHub Is Acquired by Microsoft
Published:
Microsoft announced the acquisition of GitHub on June 4, 2018 for $7.5 billion in Microsoft stock — the largest acquisition in Microsoft’s history at the time, surpassing the $2.5 billion paid for Minecraft’s developer Mojang in 2014. The deal closed October 26, 2018. GitHub CEO Chris Wanstrath, who had co-founded the company with Tom Preston-Werner and P.J. Hyett in 2008, stepped down; Nat Friedman (previously CEO of Xamarin, itself acquired by Microsoft in 2016) was appointed GitHub CEO. Read more
About Computer Vision Technology In Food Science
Published:
Computer vision is often associated with faces, autonomous vehicles, or robotics. Food science is a less obvious application, but it is one of the areas where camera-based inspection has been deployed commercially for decades — and where recent advances in deep learning have significantly expanded what is practical. Read more
JavaScript Modules Are Finally Becoming Normal
Published:
For most of JavaScript’s history, the language had no built-in way to split code across files and have those files depend on each other explicitly. This is a basic feature that Python, Java, and Ruby all had from early on. JavaScript’s absence of a module system was a genuine hole, and the ecosystem filled it with workarounds that accumulated into something genuinely complicated. Read more
GDPR Changes How Software Handles Personal Data
Published:
The EU’s General Data Protection Regulation (GDPR) became enforceable on May 25, 2018, two years after its adoption in April 2016. It applied to any organization processing personal data of EU residents, regardless of where the organization was headquartered — making it effectively a global regulation for any company with EU users. Maximum fines were €20 million or 4% of global annual turnover, whichever was higher. The first major fine came in January 2019: France’s CNIL fined Google €50 million for insufficient transparency in its Android onboarding consent flow. British Airways received a £183 million fine (later reduced to £20 million) in 2020 for a 2018 data breach affecting 500,000 customers. Read more
Cloudflare Launches 1.1.1.1 Public DNS
Published:
Cloudflare launched its public DNS resolver at the IP address 1.1.1.1 on April 1, 2018 — deliberately choosing April Fool’s Day for a memorable launch date — in partnership with APNIC (Asia-Pacific Network Information Centre). APNIC had owned the 1.1.1.1 address for years but had been unable to use it productively because router firmware worldwide had hardcoded it as a test address, flooding it with junk traffic; the partnership gave Cloudflare access to the memorable address while APNIC gained research data on global DNS query patterns. Cloudflare’s anycast network (servers in 150+ cities at that time) delivered median global response times of ~14 ms according to independent tracking by DNSPerf.com, compared to ~20 ms for Google’s 8.8.8.8 (launched December 2009) and ~20 ms for OpenDNS’s 208.67.222.222 (launched 2005) — making Cloudflare’s claim of “the world’s fastest DNS resolver” defensible at launch. The secondary resolver address was 1.0.0.1. Read more
GDPR And The Question Of Who Owns Personal Data
Published:
The timing of this post is not accidental. This month — March 2018 — the Cambridge Analytica story broke in the New York Times and the Observer, describing how the company had harvested data from up to 87 million Facebook users without their explicit consent, using an academic researcher’s app as the collection mechanism. The story arrived two months before the GDPR’s enforcement date of May 25, 2018. If there was ever a moment to think carefully about who owns personal data and what obligations come with collecting it, this is it. Read more
Google Announces Kubernetes Is Graduating at the CNCF
Published:
Kubernetes achieved graduated project status in the Cloud Native Computing Foundation on March 6, 2018, becoming the first project to complete the CNCF’s three-level maturity progression (Sandbox → Incubating → Graduated). The CNCF had been founded in July 2015 when Google donated Kubernetes to the foundation at the same event where Kubernetes 1.0 was announced (July 21, 2015, at OSCON in Portland) — a governance move designed to prevent Kubernetes from being perceived as Google’s proprietary project and to enable contributions and adoption across the industry without concerns that Google could change the direction unilaterally. CNCF graduation required a Technical Oversight Committee vote with a two-thirds supermajority; prior to the vote, the project had to demonstrate documented governance processes, committer diversity (no single company dominating more than 50 percent of commits), completion of a third-party security audit, and substantial production adoption by organizations outside the originating company. By March 2018, Kubernetes had over 1,400 contributors from 200+ organizations across its repositories, with Google’s share of commits having fallen well below 50 percent as Red Hat, CoreOS (acquired by Red Hat February 2018), Huawei, Microsoft, and IBM made substantial contributions. Simultaneously with Kubernetes’s graduation, Prometheus (the monitoring system originally developed at SoundCloud) also graduated, making the two foundational pieces of the cloud-native stack — orchestration and observability — both vendor-neutral CNCF graduated projects. Read more
SpaceX Launches Falcon Heavy With Advanced Computing Onboard
Published:
SpaceX launched Falcon Heavy for the first time on February 6, 2018 from Kennedy Space Center’s Launch Complex 39A — the same pad that launched Apollo 11 in 1969. The rocket combined three Falcon 9 first-stage cores (27 Merlin 1D engines total) and could deliver 63,800 kg to low Earth orbit, 26,700 kg to geostationary transfer orbit, and 16,800 kg to Mars — more than any operational rocket since NASA’s Saturn V flew its last mission in 1973. The test payload was Elon Musk’s personal 2010 Tesla Roadster, with a “Starman” mannequin in the driver’s seat and a copy of The Hitchhiker’s Guide to the Galaxy in the glovebox. The car was placed into a heliocentric solar orbit intended to reach Mars’s orbit; it slightly overshot, entering an orbit with aphelion approximately 257 million kilometers from the Sun. Read more
Meltdown Spectre And The Cost Of Going Fast
Published:
On January 3, 2018, Google Project Zero published details of two classes of vulnerabilities affecting almost every modern processor: Meltdown and Spectre. The coordinated disclosure involved Intel, AMD, ARM, the major operating system vendors, and cloud providers — a coordination effort that had been running secretly since June 2017 while patches were developed. Read more
Meltdown and Spectre Expose CPU Speculation Risks
Published:
Meltdown and Spectre became public in January 2018 and changed how many engineers thought about processor performance. The vulnerabilities were disturbing because they did not look like ordinary software bugs. They came from the techniques modern CPUs use to run programs faster, especially speculative execution and caching. Read more
2017
Apple Confirms It Slows Some Older iPhones to Protect Batteries
Published:
Apple confirmed on December 20, 2017 that iOS implemented “performance management” — dynamic CPU and GPU frequency limiting — on older iPhones with chemically degraded batteries, following a Reddit post by user “TeckFire” in October 2017 claiming that replacing an iPhone 6S battery had restored its performance, and a detailed benchmark analysis published December 9, 2017 by John Poole of Primate Labs (makers of Geekbench). Poole compared Geekbench 4 single-core scores from thousands of iPhone 6S devices running different iOS versions and found a distinctive bimodal distribution in iOS 10.2.1 and later: devices clustered at either the expected ~2500 score range or a lower ~1400-1600 range, with the lower cluster growing larger in iOS versions released in 2017. When Poole tested the same 6S with a fresh battery versus a degraded one, the degraded battery device scored in the lower cluster. Apple’s statement confirmed that iOS 10.2.1 (released January 23, 2017) had introduced performance management for iPhone 6, 6 Plus, 6s, and 6s Plus, and that iOS 11.2 extended it to iPhone 7 and iPhone 7 Plus. The feature had not been disclosed in any iOS release notes, developer documentation, or user-facing settings. Read more
The Rise Of Smart Speakers
Published:
Amazon introduced the Echo in late 2014, and it spent a year available only by invitation before opening broadly. Google Home launched in November 2016. Apple announced HomePod at WWDC in June and it will ship next year. In three years, voice-activated speakers have moved from a product few people had heard of to the fastest-growing consumer electronics category. Read more
Mozilla Releases Firefox Quantum
Published:
Mozilla released Firefox 57 — branded Firefox Quantum — on November 14, 2017, describing it as the biggest improvement to Firefox in over a decade. The release incorporated Stylo, a parallel CSS engine written in Rust and derived from Mozilla’s experimental Servo browser project. Stylo processed style recalculation across multiple CPU cores simultaneously, taking advantage of modern multi-core CPUs that the original Firefox engine (Gecko) had been written before were common. Mozilla reported Firefox 57 used 30% less memory than Firefox 52 and was twice as fast on the Speedometer browser benchmark, while internal testing showed it catching up to Chrome on most common web page rendering tasks. Read more
Remembering Atatürk on November 10, 2017
Published:
Every November 10, Turkiye remembers Mustafa Kemal Ataturk. Read more
Google Releases the Pixel 2
Published:
Google announced the Pixel 2 and Pixel 2 XL on October 4, 2017. The standard Pixel 2 had a 5-inch 1080p OLED display and sold for $649; the Pixel 2 XL had a 6-inch 2880×1440 pOLED display and sold for $849. Both used the Snapdragon 835, 4GB RAM, and a single 12.2 MP rear camera with an f/1.8 aperture — no telephoto lens, no wide-angle, just one sensor with a dual-pixel autofocus array. The iPhone 8 Plus released the same month had a dual 12MP setup (wide + telephoto), and Samsung’s Galaxy Note 8 also shipped with two rear cameras. Read more
Welcome to Our Department
Published:
If you are beginning Computer Engineering in 2017, welcome to a field full of attractive buzzwords. Read more
Apple Introduces Face ID with the iPhone X
Published:
Apple announced the iPhone X (pronounced “ten”) on September 12, 2017 at the Steve Jobs Theater — the first event at the new Apple Park campus in Cupertino — pricing it at $999 for 64 GB and $1,149 for 256 GB. The iPhone X replaced the physical Home button (present on every iPhone since 2007) entirely, enabling a 5.8-inch OLED Super Retina display (2436×1125, 458 ppi) that covered almost the entire front face of the device with a notch at the top for the TrueDepth camera system. The TrueDepth camera array comprised seven components: a dot projector (projecting 30,000 invisible infrared dots in a structured pattern onto the user’s face), an infrared camera (capturing the distortion of the dot pattern to build a precise 3D depth map), a flood illuminator (illuminating the face with IR light to enable recognition in the dark), a 7MP visible-light front camera, an ambient light sensor, a proximity sensor, and a speaker. The system worked in darkness where no 2D camera could function, and Apple quoted a false-acceptance rate of 1 in 1,000,000 for Face ID versus 1 in 50,000 for Touch ID — a 20× improvement in the probability of a random person’s face unlocking the phone. Read more
Why Choose Computer Engineering in 2017?
Published:
In 2017, choosing Computer Engineering means entering a field that is expanding far beyond traditional software development. Read more
AMD Launches the Threadripper Platform
Published:
AMD launched Ryzen Threadripper on August 10, 2017, with the Threadripper 1950X (16 cores, 32 threads) at $999 and the Threadripper 1920X (12 cores, 24 threads) at $799. Both used AMD’s TR4 socket, a physically massive 4094-pin LGA package on the X399 chipset. The 1950X was priced directly against Intel’s new Core i9-7900X (10 cores, $999 launched two months earlier) and offered 60% more cores for the same price — a competitive statement rather than a niche positioning. Read more
How To Study To Be A Good Software Developer
Published:
Learning software development can easily become overwhelming because there is always another topic to study. A programming language leads to a framework. The framework leads to databases, networking, testing, operating systems, design, and security. Each topic reveals three more. Read more
Intel Launches the Skylake-X Platform
Published:
Intel launched Skylake-X in June 2017 for the LGA 2066 socket (Basin Falls platform), replacing Broadwell-E. The lineup opened with the 10-core Core i9-7900X at $999 and the 8-core i7-7820X at $599, then extended in August to 12-, 14-, 16-, and 18-core models topped by the i9-7980XE at $1,999. All chips supported quad-channel DDR4, 44 PCIe 3.0 lanes (on the top SKUs), and Intel’s X299 chipset. Core i9 was itself a new brand tier Intel created with this launch because i7 no longer distinguished the highest-end parts. Read more
The Gaming PC I Would Build in 2017
Published:
If I were building a gaming PC on July 1, 2017, this is the machine I would actually want. I am not simply choosing the most expensive part in every category. The priority is gaming performance, then reliability, cooling and enough headroom to keep the system useful for years. Read more
Apple Introduces the Core ML Framework
Published:
Apple introduced Core ML at WWDC on June 5, 2017, shipping with iOS 11 (September 19, 2017), macOS High Sierra (September 25, 2017), watchOS 4, and tvOS 11. Core ML provided a unified on-device inference framework: developers converted pre-trained models from common training frameworks (Keras, Caffe 1, scikit-learn, XGBoost, libsvm) into Apple’s .mlmodel format using the coremltools Python package, then added the .mlmodel file to an Xcode project where Xcode automatically generated a Swift/Objective-C class wrapping the model’s inputs and outputs. At runtime, Core ML automatically routed computation to the most appropriate hardware: the CPU (using BNNS — Basic Neural Network Subroutines — a low-level optimized library), GPU (via Metal Performance Shaders, for models with GPU-friendly layer types), or the Neural Engine on devices that had one. The A11 Bionic (iPhone X, September 2017) shipped the first dedicated Neural Engine — a 2-core block running at approximately 600 GOPS, used exclusively for Core ML inference and not exposed to general Metal computation — enabling face recognition for Face ID and Animoji at low power. The A12 Bionic (iPhone XS, September 2018) expanded this to an 8-core Neural Engine capable of 5 TOPS, an 8× improvement, and established the Neural Engine as a permanent dedicated accelerator on all subsequent Apple SoCs. Read more
Google Announces Kotlin Support for Android
Published:
Google announced first-class Kotlin support for Android at Google I/O on May 17, 2017, endorsing Kotlin as an officially supported language for Android development alongside Java. Kotlin was created by JetBrains (the company behind IntelliJ IDEA, on which Android Studio was based) and had reached version 1.0 in February 2016 after years of development that started in 2010. The announcement included immediate support in Android Studio 3.0 (released October 2017) — Kotlin tooling was built into the IDE at the same level as Java, with full debugging support, code completion, refactoring, and a Java-to-Kotlin converter that could automatically translate existing Java files. Kotlin compiled to JVM bytecode (like Java), meaning Kotlin and Java source files in the same project compiled to interchangeable .class files, and Kotlin code could directly call Java APIs and vice versa — including the entire Android SDK (android.app.Activity, android.widget., android.content.), all Jetpack libraries, and any Java library in Maven Central. Read more
Kotlin Arrives On Android
Published:
Java has been the main language of Android development since the first Android SDK in 2008. It has worked, but it has also accumulated frustrations that Android developers have complained about for years. Kotlin — designed by JetBrains and reaching version 1.0 in February 2016 — has been gaining serious traction as an alternative, and the community momentum suggests this is no longer a niche experiment. Read more
Docker Introduces the Moby Project
Published:
Docker announced the Moby Project at DockerCon 2017 in Austin, Texas on April 18, 2017, as a structural reorganization of Docker’s open-source components. Solomon Hykes, Docker’s founder and CTO, introduced Moby as an open framework of modular components from which container systems — including Docker itself — could be assembled. Moby encompassed the building blocks Docker had developed over four years: containerd (container runtime, donated by Docker to the CNCF in March 2017), runc (the OCI-compliant low-level runtime, originally submitted as the reference implementation when Docker co-founded the Open Container Initiative in June 2015), LinuxKit (a toolkit for building minimal, immutable Linux distributions optimized for container workloads), BuildKit (a next-generation image build backend with parallel layer execution and improved caching), SwarmKit (orchestration primitives), and Notary (content trust and signing). Docker simultaneously rebranded its products: what had been “Docker” (free) became “Docker CE” (Community Edition), and “Docker CS” (commercially supported) became “Docker EE” (Enterprise Edition), clarifying the commercial/open-source boundary. The docker/docker GitHub repository was renamed to moby/moby, which caused immediate confusion among the large number of contributors who had contributed to that repository expecting to contribute to the Docker product. Read more
Serverless Computing And The Idea Of Less Maintenance
Published:
The word “serverless” is a little funny because the servers do not actually disappear. They are still there — running in AWS data centers in Virginia or Oregon or wherever the platform routes your traffic. The servers become less visible to the developer, which is different from not existing. Read more
AMD Launches Ryzen
Published:
AMD launched Ryzen on March 2, 2017, with three models: Ryzen 7 1800X (8 cores/16 threads, 3.6 GHz base/4.0 GHz boost, $499), Ryzen 7 1700X ($399), and Ryzen 7 1700 ($329). All used AMD’s new AM4 socket and the X370 or B350 chipset. The Ryzen 7 1800X was positioned directly against Intel’s Core i7-6900K (8 cores, $1,089) — AMD offered the same core count at less than half the price. Geekbench multi-core scores placed the 1800X within 5% of the i7-6900K while costing $590 less. Read more
Cloudflare Discloses the Cloudbleed Bug
Published:
Cloudflare disclosed Cloudbleed on February 23, 2017, after Google Project Zero researcher Tavis Ormandy reported the bug to Cloudflare five days earlier on February 17 and Cloudflare deployed a fix within hours. The vulnerability had existed in production since September 22, 2016 — 5 months before disclosure — and the worst-case period of highest data leakage ran from February 13 to 18, 2017. Cloudflare estimated it had leaked roughly 1 in every 3,300,000 HTTP responses over the worst period. Read more
GitLab Faces a Major Database Incident and Shares the Recovery Publicly
Published:
On January 31, 2017 at approximately 5:20 PM UTC, a GitLab database administrator accidentally deleted 300 GB of production PostgreSQL data from db1.cluster.gitlab.com while attempting to remove a directory on a staging server. The error occurred while manually remediating a replication lag incident under time pressure: the administrator ran rm -rf targeting a directory path but was connected to the production server rather than staging. GitLab’s CEO Sid Sijbrandij and the engineering team made an unusual decision: they live-streamed the recovery effort on YouTube, opened a public Google Doc tracking the incident state, and posted updates in real time to Twitter — narrating the disaster as it unfolded rather than waiting until service was restored. Read more
About Semantic Web Technologies
Published:
Tim Berners-Lee proposed the Semantic Web in a 2001 Scientific American article written with James Hendler and Ora Lassila. The vision was compelling: an extension of the existing web in which information would have explicit meaning, allowing software agents to reason across data from different sources, make connections automatically, and perform tasks that currently require human judgment to navigate. Read more
2016
Amazon Opens Its First Checkout-Free Store Concept
Published:
Amazon announced Amazon Go on December 5, 2016 via a public video, describing a 1,800 square-foot convenience store at 2131 7th Avenue in Seattle’s South Lake Union neighborhood (Amazon’s corporate campus) where shoppers could scan a QR code in the Amazon Go mobile app at an entry gate, take items from shelves, and walk out — receiving an automatic receipt and charge to their Amazon account minutes later, with no checkout lines or cashiers required. The store opened to Amazon employees in December 2016 for internal testing; the public launch was delayed over a year to January 22, 2018, reportedly due to technical difficulties tracking large crowds (more than 20 shoppers at once in early tests created tracking accuracy problems) and disambiguating who had taken items in dense crowd scenarios. The underlying technology — which Amazon branded “Just Walk Out” — combined ceiling-mounted depth cameras (hundreds per store), weight sensors embedded in store shelves, computer vision deep learning models, and the Amazon app entry ID to build a real-time “virtual cart” for each shopper. Contrary to widespread speculation, the system did not use RFID tags on individual products; all tracking was camera and sensor fusion based. Read more
Why Two Factor Authentication Is Worth The Extra Step
Published:
Passwords are structurally weak as the sole mechanism for account security, and the events of the past few years have made this concrete rather than theoretical. In 2012, LinkedIn suffered a breach that exposed 6.5 million password hashes; in 2016, it became clear that the actual number was 117 million. Yahoo’s 2013 breach, disclosed in 2016, exposed credentials for 500 million accounts. Adobe, Dropbox, and dozens of other services have had major breaches. The stolen credentials from these breaches are traded and combined into large credential lists that attackers use in automated “credential stuffing” attacks — testing username and password combinations across many services simultaneously. Read more
The Value Of Walking Away From The Screen
Published:
Programmers have a particular relationship with difficult problems. When something does not work, the instinct is to stay. One more test. One more search. One more change to the configuration. An hour later the problem is unchanged, and the mind is worse equipped to solve it than it was when the session started. Read more
Google Home and Modern Voice Assistants Expand
Published:
Google Home went on sale November 4, 2016 at $129, announced at Google I/O in May 2016 and launching four months after the Pixel phones introduced Google Assistant (October 4, 2016). The device competed directly with Amazon Echo (which had launched June 23, 2015 at $179.99 for Prime members) after Amazon had built a 1.5-year head start in the smart speaker market with Alexa. Google Home’s hardware included a 2-inch tweeter and passive bass radiator for audio output, a 7-microphone far-field array for voice pickup, and a 360° sound design with a fabric mesh exterior (replacing Amazon’s cylindrical black plastic aesthetic). The far-field microphone array performed beamforming — electronically steering the sensitive direction of the combined microphone pickup toward the detected voice source and suppressing the direction of the device’s own speaker output — to enable voice detection while music played at moderate volume from the same device. Always-on keyword detection (“Ok Google”) ran on-device using a small neural network model; the audio before and after keyword detection was streamed to Google’s servers for the full speech recognition and natural language processing to occur. Read more
About Social Network Analysis
Published:
Social network analysis studies relationships rather than only individuals. That small shift in perspective — from nodes to edges, from people to the structure between them — can reveal things that individual-level data cannot. Read more
Remembering Atatürk on November 10, 2016
Published:
Every November 10, Turkiye pauses to remember Mustafa Kemal Ataturk. Read more
Google Announces the First Pixel Phones
Published:
Google announced the Pixel and Pixel XL on October 4, 2016, with the Pixel (5-inch 1080p AMOLED) starting at $649 and the Pixel XL (5.5-inch 1440p AMOLED) starting at $769. Both ran on the Snapdragon 821 with 4GB RAM and a 12.3MP rear camera — the same sensor spec as Sony’s Xperia phones, since Sony manufactured the sensor. HTC manufactured the hardware under contract with Google owning the design. Unlike Nexus devices, Google retained the software experience entirely: no HTC UI overlay, no carrier bloatware from US carriers, and a commitment to three years of OS updates and security patches directly from Google. Read more
Welcome to Our Department
Published:
If you are starting Computer Engineering in 2016, welcome. Read more
Chatbots Before The Hype
Published:
In April 2016, Facebook opened the Messenger Platform to developers, allowing companies to build bots that could receive and send messages through Messenger’s API. In the weeks that followed, thousands of bots appeared — bots to order pizza, check weather, browse CNN headlines, get customer support, and more. Microsoft launched the Bot Framework in March 2016 with connectors for Skype, Slack, and other platforms. Amazon Alexa, which had launched with the Echo in November 2014, was drawing attention from developers building voice-driven skills for the device. Read more
Apple Removes the Headphone Jack from the iPhone
Published:
Apple announced the iPhone 7 and 7 Plus on September 7, 2016, with both models conspicuously missing the 3.5mm headphone jack that had been on every iPhone since the original in 2007. Apple marketing chief Phil Schiller justified the removal on stage with the word “courage,” which became a widely mocked summary of the decision. In the box: a Lightning-to-3.5mm adapter (so existing wired headphones still worked with the included adapter plugged in, though simultaneously charging was then impossible) and EarPods with a Lightning connector. Separately, Apple introduced AirPods at $159 — its own wireless earbuds, not shipping until December 2016. Read more
Why Choose Computer Engineering in 2016?
Published:
Choosing a university department means thinking about both the present and the future. In 2016, Computer Engineering is one of the fields where those two perspectives align better than in most others. Read more
Intel Announces More Details of Kaby Lake
Published:
Intel officially announced Kaby Lake (7th generation Intel Core) at Intel Developer Forum (IDF) in San Francisco in August 2016, with mobile SKUs available in September 2016 and desktop processors launching in January 2017. The Kaby Lake announcement formalized Intel’s departure from the two-decade “tick-tock” strategy — alternating between process node shrinks (“tick”) and new microarchitecture designs (“tock”) every year — replacing it with a three-phase “Process-Architecture-Optimization” (PAO) model. Skylake (6th gen, 2015) was the architecture phase on 14nm; Kaby Lake was the optimization phase, also on 14nm (labeled “14nm+” internally, with improved fin geometry and higher drive current enabling higher clock frequencies at similar voltage). IPC (instructions per clock) was essentially unchanged from Skylake — the same Skylake microarchitecture, core design, L1/L2/L3 cache hierarchy, and ring bus topology — but clock speed improvements were meaningful: the desktop Core i7-7700K reached 4.2 GHz base / 4.5 GHz single-core turbo versus the Core i7-6700K’s 4.0 GHz / 4.2 GHz on the same 14nm process. The 14nm+ improvements that enabled higher clocks also reduced leakage current, contributing to slightly better power efficiency in mobile SKUs. Read more
How To Be A Good Software Developer
Published:
Becoming a good software developer is not only about learning more programming languages. Languages matter. Tools matter. Frameworks matter. But after spending enough time around software, I think the deeper skills are more stable and harder to replace. Read more
Pokemon Go Brings Augmented Reality to the Mainstream
Published:
Pokémon Go launched on July 6, 2016 in the United States, Australia, and New Zealand, developed by Niantic (which had spun out from Google’s X division in 2015) in partnership with The Pokémon Company and Nintendo. Within two weeks it had expanded to Europe, Canada, and most major markets. The game used GPS location data and real-world map data to place 151 original-generation Pokémon at geographic coordinates in the physical world, requiring players to walk to those locations to encounter and catch them. PokéStops (where players collected in-game items) and Gyms (where players battled) were placed at real-world landmarks — museums, public art installations, parks, notable architecture, historic sites — derived from Niantic’s earlier game Ingress (2012), which had built a global database of “portals” at user-submitted real-world locations. The AR mode that became the game’s visual signature overlaid a rendered Pokémon on the smartphone’s live camera feed using the device’s gyroscope and accelerometer to track orientation, placing the creature in a fixed world-space position that the camera revealed as the player moved the phone. Most players turned off AR mode for the actual catching interaction because of battery drain and targeting difficulty; the GPS-driven exploration of the physical world was the game’s actual core mechanic. Read more
The Gaming PC I Would Build in 2016
Published:
If I were building a gaming PC on July 1, 2016, this is the machine I would actually want. I am not simply choosing the most expensive part in every category. The priority is gaming performance, then reliability, cooling and enough headroom to keep the system useful for years. Read more
Docker Announces Built-In Swarm Mode
Published:
Docker announced built-in Swarm mode at DockerCon Seattle (June 19–21, 2016), integrating cluster orchestration directly into Docker Engine 1.12 rather than requiring the separate Docker Swarm standalone tool. The previous Docker Swarm (released November 2014) required a separate daemon, an external key-value store (etcd, Consul, or ZooKeeper) for cluster state, and manual TLS certificate management — significant operational complexity for teams that wanted basic multi-host container deployment. The new Swarm mode embedded all of this into the standard docker swarm init and docker service create commands. Read more
Containers Are Changing The Way Software Is Delivered
Published:
Deployment problems are often environment problems. A program works correctly on a developer’s machine, then fails on a staging server because a library is a different version, a configuration file has a different path, or a system dependency is missing. Debugging these discrepancies is time-consuming and tedious because the mismatch can be anywhere in a large stack. Read more
Google Introduces the Tensor Processing Unit
Published:
Google announced the Tensor Processing Unit (TPU) at Google I/O on May 18, 2016, with Jeff Dean revealing that TPUs had already been deployed in Google’s data centers since 2013 — three years of silent production use. The TPU v1 was an inference-only ASIC built on a 28nm process, consuming 40 watts (roughly a quarter of an equivalent GPU), and connected to a host CPU via PCIe. It targeted deployed inference rather than training, focusing entirely on the matrix operations that dominate neural network forward passes during serving. Read more
Vim A Powerful Text Editor
Published:
Vim is one of those tools that looks unfriendly before it starts to make sense. The first time someone opens it, even quitting the editor requires knowledge that is not obvious — :q to quit, :wq to save and quit, or :q! to quit without saving. That has become a standing joke among programmers, but it also explains something important about the editor. Vim does not try to behave like a normal graphical text editor. It is a modal editor with a grammar. Read more
Deep Learning Is Changing Computer Vision
Published:
Computer vision has always fascinated me because the task sounds simple when we describe it in human language: “Find the object in this image.” “Recognize this face.” “Tell me what is happening in this picture.” Humans do these things before they can write. For computers, the same tasks have been among the hardest problems in artificial intelligence for decades. Read more
Microsoft Releases Ubuntu on Windows Preview
Published:
Microsoft announced the Windows Subsystem for Linux (WSL) at BUILD 2016 in San Francisco on March 30, 2016, and released the first public preview on April 6, 2016 as part of Windows 10 Insider Build 14316. The feature was developed in partnership with Canonical — Ubuntu’s creator — and debuted as “Bash on Ubuntu on Windows,” reflecting that the initial user-space environment was Ubuntu 14.04 LTS. The announcement was notable in context: Microsoft had declared “Linux is a cancer” through then-CEO Steve Ballmer in 2001, and the company’s relationship with open-source software had been defined by opposition and IP disputes through much of the 2000s. Under CEO Satya Nadella (appointed February 2014) and developer tools head Scott Guthrie, Microsoft’s position had reversed: the company had acquired developer tooling companies, open-sourced .NET Core, published Visual Studio Code, and introduced extensive Azure Linux support. WSL was the direct expression of that shift for developer tooling on the desktop. The architecture of WSL 1 was engineered by a small team within Microsoft’s kernel group: it used “pico processes” — a lightweight process model introduced in Windows 10 — combined with an NT kernel driver (lxss.sys and lxcore.sys) that intercepted Linux system calls (syscalls) made by Linux ELF binaries and translated them into equivalent NT kernel operations. This allowed unmodified Linux binaries compiled for x86-64 Ubuntu to execute within Windows without a virtual machine or kernel swap: bash, ssh, python, gcc, apt-get, grep, awk, sed, and most standard Unix command-line tools ran as native processes in the Windows process model. Read more
Raspberry Pi 3 And The Small Computer That Keeps Growing
Published:
The Raspberry Pi 3 launched last month, and the specifications represent a meaningful jump from the Pi 2 — but the change I find most interesting is not the processor. Read more
AlphaGo Defeats Lee Sedol
Published:
AlphaGo, developed by DeepMind (acquired by Google in January 2014), defeated Lee Sedol — a 9-dan professional Go player widely considered the best active player of the 2010s — in a five-game match held March 9–15, 2016 in Seoul, South Korea, by a score of 4–1. The result shocked the professional Go community: Go’s enormous complexity (roughly 10^170 possible board positions versus chess’s estimated 10^44) had led experts to estimate that computers would not reach top-human level for another decade. AlphaGo had already defeated European Go champion Fan Hui 5–0 in October 2015, but Fan Hui was ranked approximately 700th in the world at the time; Lee Sedol was a fundamentally different caliber of opponent. The match was broadcast live on YouTube and watched by an estimated 200 million viewers in China and internationally. AlphaGo’s architecture combined two deep convolutional neural networks: a policy network (trained first by supervised learning on approximately 30 million positions from the KGS Go server, then refined by reinforcement learning through self-play) that predicted the probability distribution over next moves, and a value network (trained by self-play) that estimated the probability of winning from a given board position. These networks guided a Monte Carlo Tree Search that explored the game tree selectively, simulating future positions along the most promising branches rather than searching exhaustively. Read more
Google Announces Accelerated Mobile Pages
Published:
Google announced the Accelerated Mobile Pages (AMP) project on October 7, 2015 and began rolling out AMP results in Google Search in February 2016. AMP was conceived as a response to the mobile web’s performance crisis: news and publishing sites serving rich HTML pages with large JavaScript bundles, advertisement scripts, and tracking pixels were taking 8-22 seconds to load on 3G connections — slow enough that users frequently abandoned pages before they finished loading, and that competing closed platforms (Facebook Instant Articles, announced May 2015; Apple News, announced June 2015) offered significantly faster reading experiences by serving content inside native apps rather than the mobile browser. AMP’s technical approach was a restricted HTML subset — AMP HTML — that banned arbitrary JavaScript entirely (replacing it with an AMP component library), required explicit size declarations on all external resources (images, iframes, ads) to prevent layout shifts as content loaded, and mandated asynchronous resource loading to ensure the page’s content was never blocked by an external script or stylesheet. AMP JS was the library that managed asynchronous resource loading, extension components (<amp-img>, <amp-video>, <amp-ad>, <amp-analytics>), and rendering optimization. The AMP specification was open-sourced on GitHub and positioned as a community project rather than a Google-proprietary format, though Google remained the primary contributor. Read more
Why Progressive Web Apps Are Interesting
Published:
The term “Progressive Web App” was coined by Google engineer Alex Russell in a blog post in June 2015. The concept describes a web application that adopts a set of features progressively — it works for every user regardless of browser, but adds capabilities like offline access, installation, and push notifications for users whose browsers support them. Read more
AlphaGo Prepares for Its Match Against Lee Sedol
Published:
On January 27, 2016, DeepMind published the Nature paper “Mastering the Game of Go with Deep Neural Networks and Tree Search” describing the AlphaGo system and simultaneously announcing that AlphaGo would face Lee Sedol — 9-dan professional Go player, 18-time world champion, and widely considered the strongest player of the 2010s — in a five-game match in Seoul in March 2016. The paper disclosed that AlphaGo had already defeated Fan Hui, the European Go champion, 5–0 in a formal match held secretly at DeepMind’s London offices on October 5–9, 2015 — the first time a computer program had beaten a professional Go player in a formal match on a standard 19×19 board without handicap. Previous Go AIs, including Zen and Crazy Stone, used Monte Carlo Tree Search (MCTS) with handcrafted evaluation functions and required 4-stone or larger handicaps to compete with professional players. Fan Hui, a 2-dan professional and European champion, was ranked approximately 700th in the world — a professional level, but far below top-10 world ranking where Lee Sedol operated. Go’s complexity — a branching factor of approximately 250 legal moves per turn (versus chess’s 35), a game tree estimated at 10^170 positions, and position evaluation that defied the heuristic functions effective in chess — had led the AI community to expect computational professional-level Go was decades away. Read more
2015
Let’s Encrypt Enters Public Beta
Published:
Let’s Encrypt entered public beta on December 3, 2015, removing the invite requirement from its limited beta (started November 2015) and opening free TLS certificate issuance to any website operator. The certificate authority was founded in April 2014 by the Internet Security Research Group (ISRG), backed by the Electronic Frontier Foundation, Mozilla, Cisco, Akamai, and IdenTrust — with IdenTrust cross-signing Let’s Encrypt’s intermediate certificate, which allowed browsers to trust Let’s Encrypt-issued certificates immediately without adding a new root CA to browser trust stores (a process taking years). The ACME protocol (Automatic Certificate Management Environment, finalized as RFC 8555 in 2019) allowed a server daemon to prove control of a domain through either an HTTP-01 challenge (placing a token at a specific URL path on the domain) or a DNS-01 challenge (adding a TXT record to the domain’s DNS zone), then receive a signed 90-day DV (Domain Validation) certificate. The 90-day lifetime was deliberately shorter than the 1–3 year certificates sold by commercial CAs: short-lived certificates reduced the window of exposure from key compromise, and the automation model made renewal every 60–90 days practical. Certbot, the EFF-developed command-line client, automated the challenge-response and certificate installation process for Apache and Nginx on Linux in a single command. Read more
HTTPS Is Becoming Everyones Problem
Published:
For years, HTTPS was associated mainly with banks, shopping carts, and login pages — places where passwords and credit card numbers obviously needed protection. A simple personal website or blog routinely stayed on plain HTTP. Certificates cost money, expired annually, required configuration, and for a blog about recipes or software notes, the effort seemed disproportionate. Read more
Google Open-Sources TensorFlow
Published:
Google open-sourced TensorFlow on November 9, 2015 under the Apache 2.0 license. The project emerged from DistBelief, Google’s internal first-generation distributed deep-learning system used to train the 2012 ImageNet model that sparked the deep-learning surge. TensorFlow was a complete rewrite, designed to be more flexible, faster on single machines, and easier to use outside Google. The initial release supported Python and C++ on Linux and Mac, with CUDA GPU acceleration available from day one. Read more
Remembering Atatürk on November 10, 2015
Published:
Every November 10, Turkiye remembers Mustafa Kemal Ataturk. Read more
Microsoft Introduces the Surface Book
Published:
Microsoft announced the Surface Book at a press event on October 6, 2015 — Panos Panay introduced it as “the ultimate laptop” — pricing it at $1,499 (Core i5, 8GB RAM, 128GB SSD, integrated GPU only) up to $2,699 (Core i7, 16GB RAM, 512GB SSD, discrete GPU). It was the first Microsoft Surface product with a clamshell laptop form factor (previous Surfaces were tablets with optional keyboard accessories). The Surface Book’s 13.5-inch display (3000×2000 resolution, 267 ppi, 3:2 aspect ratio) was mounted on a proprietary Dynamic Fulcrum Hinge — a curved hinge composed of rotating cylinders that kept the display and keyboard close together without touching the display when open, and which left a visible gap between the closed display and keyboard base. The hinge connected the display and keyboard base with four proprietary electrical contacts that carried USB 3.0 and a PCIe-like signal path, allowing the discrete GPU in the keyboard base to be accessed by the CPU and GPU in the display unit when the laptop was in clamshell mode. Read more
Apple Introduces the A9 and 3D Touch
Published:
Apple announced the iPhone 6s and 6s Plus on September 9, 2015, introducing the A9 processor and 3D Touch. The A9 was co-manufactured by both TSMC (16nm FinFET) and Samsung (14nm FinFET) — an unusual dual-source arrangement that caused public concern about performance differences between the two variants, though Apple stated the real-world gap was within a few percent. The A9 carried approximately 2 billion transistors, two high-performance CPU cores clocked at 1.85 GHz, and a six-core PowerVR GT7600 GPU. Compared to the A8, Apple claimed 70% faster CPU performance and 90% faster GPU performance while using the same battery size. Read more
Edsger Wybe Dijkstra
Published:
Edsger Wybe Dijkstra was one of the people who helped turn programming from a practical craft into a serious intellectual discipline. He was born in Rotterdam in 1930. His parents were both highly educated, and Dijkstra initially studied theoretical physics at Leiden University. Read more
Google Becomes Part of Alphabet
Published:
Larry Page published a blog post on August 10, 2015 announcing that Google was restructuring under a new holding company named Alphabet Inc. — the parent company of Google and a collection of separately managed subsidiaries. The restructuring took effect immediately: Sundar Pichai (who had been Google’s SVP of Products overseeing Search, Maps, Android, Chrome, and YouTube) became CEO of Google (the subsidiary), while Larry Page became CEO of Alphabet and Sergey Brin became President of Alphabet. Page explained the rationale in the announcement: Google had become “a bit uncomfortable” as a single entity encompassing both its highly profitable core businesses and long-term research projects that required different management cultures, time horizons, and capital allocation approaches. Keeping Waymo’s (then Google Self-Driving Car Project’s) multi-decade R&D investment lumped inside Google’s advertising revenue reporting made it difficult to assess the performance of either business on its own terms. Read more
Windows 10 Is Released
Published:
Microsoft released Windows 10 on July 29, 2015 and offered it as a free upgrade to Windows 7 and 8.1 users for one year. The offer addressed Windows 8’s poor reception: launched in October 2012, Windows 8 removed the Start menu in favor of a touch-first tiled interface that laptop and desktop users largely rejected. Windows 8.1 in October 2013 partially reversed course. Windows 10 went further by restoring the Start menu and making the modern interface optional. Read more
The Gaming PC I Would Build in 2015
Published:
If I were building a gaming PC on July 1, 2015, this is the machine I would actually want. I am not simply choosing the most expensive part in every category. The priority is gaming performance, then reliability, cooling and enough headroom to keep the system useful for years. Read more
Microsoft Releases Visual Studio Code
Published:
Microsoft announced Visual Studio Code at Build 2015 on April 29, 2015 and released it as a free public preview the same day, with VS Code 1.0 (the first stable release) shipping April 14, 2016. The project was led by Erich Gamma — one of the Gang of Four (co-author of Design Patterns: Elements of Reusable Object-Oriented Software, 1994) and the original creator of JUnit — who had joined Microsoft in 2011 specifically to build a new developer tool for the web era. VS Code was built on GitHub’s Electron framework (then called “Atom Shell,” renamed Electron in 2015), which wrapped Chromium’s rendering engine and Node.js in a single bundle, enabling a cross-platform desktop application built in TypeScript, HTML, and CSS — running on Windows, macOS, and Linux from the same codebase. The Monaco editor component (VS Code’s text editing core) was separately open-sourced and became the basis for browser-based code editing in Azure DevOps, GitHub’s web editor (., now github.dev), CodeSandbox, StackBlitz, and many other web IDEs. VS Code itself was published as open source under the MIT License on GitHub (github.com/microsoft/vscode) at launch. Read more
Google Photos Launches
Published:
Google launched Google Photos at Google I/O on May 28, 2015, spinning it out from Google+ Photos as a standalone product with a simple and powerful offer: free unlimited photo storage for images up to 16 megapixels and videos up to 1080p, with no storage quota against the user’s Google Drive or Gmail space. Google+ had been launched in June 2011 partly as a photo-sharing platform, and over four years it had accumulated enormous numbers of photos; by separating Google Photos from the declining Google+ social network, Google preserved the photo infrastructure and its 200+ million existing users without requiring them to maintain Google+ engagement. The pricing and capacity offer was immediately competitive: Apple’s iCloud offered 5 GB free (shared across photos, mail, and backups), Dropbox offered 2 GB free, and Amazon Prime Photos (November 2014) offered unlimited photo storage to Prime subscribers but only 5 GB for videos. Free unlimited storage for full-resolution photos — at a scale Google could afford because photo storage costs had declined and because Google’s use of the photo data for machine learning purposes provided indirect commercial value — positioned Google Photos as the primary cloud photo backup destination for Android users and a compelling alternative for iOS users. Read more
Apple Watch Reaches Consumers
Published:
The first Apple Watch went on sale on April 24, 2015 in two sizes — 38mm and 42mm — and three collections: Sport starting at $349, Watch starting at $549, and Edition in 18-karat gold starting at $10,000. It used Apple’s S1 system-in-package, a small OLED display with Force Touch, heart-rate sensing, accelerometers, Taptic Engine haptic feedback, and a Digital Crown for scrolling. Watch required an iPhone 5 or later as a paired device and communication partner. Read more
USB Type-C Reaches New Consumer Devices
Published:
USB Type-C arrived on major consumer devices in early 2015: the 12-inch MacBook (announced March 9, released April 10, $1,299) shipped with a single USB-C port as its only connector, and the second-generation Chromebook Pixel (announced March 11, $999) added two USB-C ports alongside standard USB-A. Both used USB 3.1 Gen 1 (5 Gbps) over the new connector. The MacBook’s single-port approach drew significant criticism because users needed a $79 USB-C to USB Adapter or a $79 USB-C to MagSafe adapter to connect legacy accessories or charge via existing cables. Read more
Docker And The Idea Of Portable Software
Published:
One sentence has caused programmers pain for a very long time: “It works on my machine.” Docker, which released version 1.0 last June and has grown quickly since then, is the most practical answer I have seen to this problem. Read more
Raspberry Pi 2 Is Released
Published:
The Raspberry Pi 2 Model B was released on February 2, 2015 at the same $35 price as the Pi 1 Model B+, replacing the BCM2835 SoC (single ARM11 core at 700 MHz) with the BCM2836 (quad-core ARM Cortex-A7 at 900 MHz). The Cortex-A7 delivered approximately 6× the raw compute performance of the ARM11 in multithreaded workloads, and the upgrade to 1 GB LPDDR2 RAM (from 512 MB) made concurrent processes, browser tabs, and compiled code workloads materially faster. Microsoft announced Windows 10 IoT Core support for the Pi 2 the same day as the hardware release — the first Raspberry Pi to support any version of Windows, enabled by the ARMv7-A instruction set that Windows requires (the Pi 1’s ARMv6 was incompatible with Windows Phone and Windows RT binary distributions). Read more
Microsoft Reveals More Details About Windows 10
Published:
Microsoft held a “Windows 10: The Next Chapter” event on January 21, 2015 in Redmond, revealing major features of Windows 10 in detail for the first time since the Technical Preview had launched in October 2014. The most visible reversal of Windows 8’s decisions was the return of the Start menu — not the Windows 7 style menu (a list of programs), but a hybrid: a left panel listing recently used and all apps alphabetically, and a right panel with the Windows 8 live tiles resizable in a grid. Live tiles remained optional — users could remove them entirely or populate the right panel with a traditional app grid, giving both the visual dashboard of Windows 8 and the familiar Start menu feel of Windows 7. The Start button on the taskbar was restored as a permanent fixture (Windows 8 had removed it; Windows 8.1 had restored the button but sent it to the full-screen Start Screen). Read more
2014
Docker Containers Move Deeper Into Production
Published:
Docker had released version 1.0 in June 2014 — its first production-ready release — and by December 2014 adoption was accelerating past developer experimentation into actual production deployments. Docker Hub, the public registry, hosted over 60,000 images by the end of 2014. Major cloud providers were integrating Docker support: Amazon ECS (Elastic Container Service) launched in November 2014 as Amazon’s first managed container service, and Google Container Engine (later renamed GKE) launched in beta in November 2014 built on Kubernetes, which Google had open-sourced in June. Read more
Amazon Introduces the Echo
Published:
Amazon announced the Echo in November 2014 as an invitation-only product, priced at $179 ($99 for Amazon Prime members). It went on general sale in June 2015. The device was a black cylinder 9.25 inches tall containing two speakers — a 2-inch tweeter and a 2.5-inch woofer — and a seven-microphone array with 360° sound pickup. The microphone array used beamforming to isolate the direction of the speaker’s voice and suppress background noise, allowing Alexa to hear commands from across a room while music was playing. Read more
Remembering Atatürk on November 10, 2014
Published:
Every November 10, Turkiye pauses to remember Mustafa Kemal Ataturk. Read more
HTML5 Becomes a W3C Recommendation
Published:
HTML5 became an official W3C Recommendation on October 28, 2014, eight years after the WHATWG began drafting it in 2006 with Apple, Mozilla, and Opera as founding members after the W3C had chosen to pursue XHTML 2.0 instead. By 2014, Chrome, Firefox, Safari, and Internet Explorer 11 had all shipped most of the specification, so the Recommendation ratified practice already in production rather than introducing new features. Read more
Apple Introduces the First Apple Watch
Published:
Apple announced the Apple Watch at a September 9, 2014 event alongside the iPhone 6, with Tim Cook calling it “the most personal device we’ve ever created.” The watch went on sale April 24, 2015 in three collections — Apple Watch Sport (aluminum, starting at $349), Apple Watch (stainless steel, starting at $549), and Apple Watch Edition (18-karat gold, starting at $10,000) — in 38mm and 42mm case sizes with interchangeable bands across nine styles. Read more
IBM Announces a Major Investment in Cloud Computing
Published:
IBM announced in January 2014 a $1.2 billion global cloud infrastructure investment, the largest single capital commitment to cloud expansion IBM had made, building on its June 2013 acquisition of SoftLayer Technologies for approximately $2 billion. SoftLayer, founded in 2005 and operating 13 data centers with approximately 25,000 customers at the time of acquisition, had built its platform around a hybrid bare-metal and virtual machine model connected through a private network backbone — a design that differentiated it from AWS, which offered only virtualized instances. IBM’s 2014 investment expanded SoftLayer’s data center footprint to include new locations in London, Hong Kong, Singapore, Amsterdam, Dallas, Washington D.C., Toronto, San Jose, Sydney, and Mexico City, bringing the total to over 40 data centers across six continents. IBM also launched IBM Bluemix in February 2014 (now IBM Cloud), a Platform-as-a-Service offering based on the open-source Cloud Foundry framework, providing hosted runtimes for Java, Node.js, Ruby, and Python applications alongside a growing catalog of IBM Watson APIs, IBM Cloudant (a CouchDB-compatible database service IBM acquired in February 2014 for approximately $150 million), and integrations with IBM’s existing enterprise software portfolio including DB2, WebSphere application server, and IBM MQ message queuing. IBM claimed a cloud revenue run rate exceeding $7 billion for 2014, though the definition included hybrid cloud managed services and software licenses hosted by customers, not only public cloud infrastructure. Read more
Google Announces Kubernetes
Published:
Google announced Kubernetes as an open-source project on June 6, 2014 at DockerCon, and released version 0.1 in July 2014 under the Apache 2.0 license. The project was created by Joe Beda, Brendan Burns, and Craig McLuckie, who drew on Google’s decade of experience running its internal Borg cluster manager — which by 2014 was scheduling hundreds of thousands of jobs across Google’s data centers. Docker had popularized Linux containers in 2013, but it addressed packaging and running single containers, not orchestrating thousands of them across many machines. Read more
The Gaming PC I Would Build in 2014
Published:
If I were building a gaming PC on July 1, 2014, this is the machine I would actually want. I am not simply choosing the most expensive part in every category. The priority is gaming performance, then reliability, cooling and enough headroom to keep the system useful for years. Read more
Apple Announces Swift
Published:
Apple announced Swift at WWDC on June 2, 2014 — a complete surprise to the developer community, as the project had been kept entirely secret. Chris Lattner, who had joined Apple in 2005 and created the LLVM compiler infrastructure and Clang C/C++ frontend, had been developing Swift in secret since July 2010 as a modern replacement for Objective-C. Objective-C had been Apple’s primary development language for iOS and macOS since NeXT’s acquisition in 1997, but it carried decades of C compatibility baggage: pointer arithmetic could produce undefined behavior, nil pointer dereferences silently returned 0 or nil (a common source of logic errors), string handling required manual bridging between NSString and C strings, and the square-bracket message-sending syntax ([object method:argument]) was unfamiliar to developers coming from any other modern language. Xcode 6 Beta (released the same day) included full Swift support and Playgrounds — an interactive REPL-like environment where code executed incrementally and results appeared inline alongside the code, enabling exploration of Swift syntax and API behavior without the compile-run-debug cycle. Read more
Microsoft Introduces the Surface Pro 3
Published:
Microsoft announced the Surface Pro 3 on May 20, 2014, with retail availability beginning June 20, 2014 at prices from $799 (Intel Core i3, 64GB) through $1,949 (Core i7, 512GB). The device used a 12-inch display at 2160×1440 resolution (216 PPI, 3:2 aspect ratio) — a change from the 10.6-inch 16:9 display used on Surface Pro 1 and 2, motivated by the observation that 16:9 screens waste vertical space when displaying documents, making them awkward for productivity work while they’re optimized for video. The Surface Pen (256 pressure levels, AAA battery, magnetic attachment to the right edge) and the redesigned kickstand (adjustable to any angle from 22° to 150°, versus the Surface Pro 2’s two fixed positions) were positioned to make the device genuinely viable for handwriting, annotation, and note-taking workflows. Read more
Heartbleed Reveals a Serious OpenSSL Weakness
Published:
Heartbleed (CVE-2014-0160) was publicly disclosed on April 7, 2014, simultaneously by Neel Mehta of Google Security and the Codenomicon research team. The vulnerability had existed in OpenSSL versions 1.0.1 through 1.0.1f since the Heartbeat extension was added on March 14, 2012 — meaning the flaw had been present for over two years before discovery. OpenSSL powered roughly 17 percent of all HTTPS-secured web servers at the time, running on Apache and nginx installations worldwide. The Heartbeat extension (RFC 6520) was designed to keep TLS sessions alive without a full handshake: a client sends a payload and declares its length, and the server echoes the same data back. OpenSSL’s implementation in tls1_process_heartbeat() failed to validate that the declared length matched the actual payload length provided. An attacker could send a 1-byte payload while claiming a length of 65,535 bytes, causing OpenSSL to copy 64 KB of process memory — whatever happened to reside adjacent to the buffer — directly into the heartbeat response. The attack required no authentication and left no log traces in standard server configurations. Read more
Heartbleed And The Fragile Trust Behind The Lock Icon
Published:
On April 7, 2014, a security bug called Heartbleed was publicly disclosed, making a library most people had never heard of suddenly important to millions of people. The bug — CVE-2014-0160 — affected OpenSSL, the open-source cryptographic library used by the majority of servers to provide HTTPS connections. Read more
Responsive Web Design Is Becoming Necessary
Published:
A few years ago, designing a website meant designing for a computer screen. The assumption was safe enough when the percentage of web traffic from phones was in single digits. In 2014, that assumption is becoming dangerous. Read more
Java 8 Is Released
Published:
Java 8 (JDK 8) was released on March 18, 2014, under Oracle’s stewardship after completing development as Project Lambda (JSR 335), led primarily by Brian Goetz. The central addition was lambda expressions: anonymous functions passed as values, enabling code like list.stream().filter(x -> x.size() > 5).map(String::toUpperCase).collect(toList()) instead of a loop with anonymous Comparator or Runnable inner classes. Lambdas required a complementary language feature — functional interfaces (interfaces with exactly one abstract method, annotated optionally with @FunctionalInterface) — and a new syntax for method references (String::length, System.out::println, ArrayList::new), which allowed passing existing named methods where lambdas would otherwise be used. Read more
Wearable Computing Is Looking For A Purpose
Published:
Computers have moved from desks to laps and then into pockets. The next suggested destination is the body, and in February 2014 several companies are in the middle of that experiment with mixed results. Read more
Facebook Acquires WhatsApp
Published:
Facebook announced the acquisition of WhatsApp on February 19, 2014 for approximately $19 billion — $4 billion in cash, $12 billion in Facebook stock, and $3 billion in restricted stock units for WhatsApp employees. The price was remarkable for a company founded in 2009 by Jan Koum and Brian Acton, two former Yahoo engineers, with around 450 million active users and only about 55 employees at the time of the deal. Read more
Google Acquires DeepMind
Published:
Google’s acquisition of DeepMind closed in January 2014 (announced January 26, 2014) for a reported price of approximately $500 million — the largest UK tech acquisition to that point and one of the largest AI company acquisitions ever. DeepMind had been founded in 2010 in London by Demis Hassabis (a former Cambridge University chess prodigy and game designer who had co-designed Theme Hospital at Bullfrog Productions in the 1990s, then completed a PhD in cognitive neuroscience at UCL), Shane Legg (a machine learning researcher who had studied under Marcus Hutter on universal AI theory), and Mustafa Suleyman (a co-founder without a technical research background but with policy and startup experience). The company had been working on combining deep neural networks with reinforcement learning — training agents through trial-and-error reward signals — a combination that had been theoretically understood for decades but was now computationally practical with GPUs and large datasets. Facebook had also attempted to acquire DeepMind; Google prevailed, and DeepMind’s acquisition agreement included an unusual condition: an ethics board would be established to oversee the use of DeepMind’s technology, a clause that reflected early concerns about AI safety at the company. Read more
2013
Docker Reaches Its First Major Wave of Adoption
Published:
Docker was released as an open-source project on March 20, 2013 by Solomon Hykes in a five-minute lightning talk at PyCon US, presented as an internal tool from dotCloud (a platform-as-a-service company). The GitHub repository went from 0 to 6,700 stars within six months, an unusually fast adoption curve for infrastructure tooling. By December 2013, Docker had attracted commitments from Red Hat (including Docker in RHEL), Google (integrating Docker with GCE), IBM, Rackspace, and Amazon Web Services (announcing Docker support on EC2), establishing it as the emerging standard for Linux container packaging even before version 1.0. Read more
JavaScript Frameworks And The Search For Order
Published:
JavaScript applications have grown large enough that writing them as a collection of event handlers is no longer comfortable. Gmail, Google Maps, Trello, and similar applications live entirely in the browser and behave like desktop software: state changes without page loads, data synchronizes with a server in the background, and user interactions update multiple parts of the interface simultaneously. Managing this complexity in unstructured JavaScript produces code that becomes difficult to understand and modify after a certain size. Read more
The PlayStation 4 and Xbox One Launch
Published:
Sony’s PlayStation 4 launched November 15, 2013 in North America at $399, followed by Microsoft’s Xbox One on November 22, 2013 at $499. Both used AMD’s “Jaguar” x86-64 CPU core architecture (8 cores at 1.6 GHz in PS4, 1.75 GHz in Xbox One) and AMD GCN (Graphics Core Next) GPU architectures, representing a sharp architectural departure from the previous generation: PS3 had used IBM’s Cell processor (a heterogeneous multi-core with 7 Synergistic Processing Elements running at 3.2 GHz) and PS4 replaced it entirely with x86-64 that any developer with PC experience could program directly. Xbox 360 had used a PowerPC-based IBM Xenon CPU. The x86-64 transition eliminated the platform-specific optimization work that had defined PS3 development (where Cell’s SPE parallelism required explicit task allocation), making initial multi-platform games significantly cheaper to develop and port. Read more
Remembering Atatürk on November 10, 2013
Published:
Every November 10, Turkiye remembers Mustafa Kemal Ataturk. Read more
Windows 8.1 Is Released
Published:
Windows 8.1 went on sale on October 17, 2013 as a free update from Windows 8, delivered through the Windows Store — the first time a major Windows upgrade was distributed this way. The update came less than a year after Windows 8 launched on October 26, 2012, an unusually fast response driven by poor reception: Windows 8 had been criticized for replacing the Start menu with a full-screen tile interface that made no accommodation for mouse-and-keyboard users on non-touch PCs. Read more
Online Courses And The Changing Classroom
Published:
The New York Times called 2012 “The Year of the MOOC.” Massive open online courses arrived with enormous energy: Coursera launched in April 2012, founded by Stanford professors Daphne Koller and Andrew Ng; edX launched two months later as a nonprofit from MIT and Harvard; Udacity, founded by Sebastian Thrun after his Stanford AI course enrolled 160,000 students online in 2011, was already operating. By September 2013, millions of students worldwide have enrolled in courses on these platforms, often from institutions they could not have attended in person. Read more
Apple Introduces the 64-Bit A7 Processor
Published:
Apple announced the iPhone 5s on September 10, 2013, introducing the A7 — which Apple described as the first 64-bit processor in a consumer smartphone. The A7 used Apple’s Cyclone CPU cores, designed in-house around the ARMv8 architecture, and contained 1 billion transistors manufactured on Samsung’s 28nm process. Cyclone’s dual high-performance cores ran at up to 1.3 GHz; Apple claimed 2× CPU and 2× GPU performance compared to the A6. The chip paired with 1GB of LPDDR3 RAM — the same as the iPhone 5, and nowhere near the 4GB minimum where a 64-bit address space provides a practical RAM benefit. Read more
Microsoft Finalizes Windows 8.1
Published:
Microsoft released Windows 8.1 to manufacturing (RTM) on August 27, 2013, completing the build that would go to device makers and enterprise customers before the October 17 public release. RTM meant the code was final and ready for hardware certification — OEMs needed this lead time to preinstall 8.1 on new PCs shipped around the holiday season. The August milestone arrived less than ten months after Windows 8 launched on October 26, 2012, signaling how seriously Microsoft was treating user complaints. Read more
Google Chromecast Is Released
Published:
Google announced and began selling the Chromecast on July 24, 2013 for $35, alongside a limited-time offer of three months of Netflix free — which effectively made the device free for existing Netflix subscribers. Google sold out of the first production run within hours. The dongle was 72mm long, plugged directly into a TV’s HDMI port, and drew power from a USB port (on the TV or via the included wall adapter), with no remote control, no standalone interface, and no local application storage beyond a small bootloader. Read more
The Gaming PC I Would Build in 2013
Published:
If I were building a gaming PC on July 1, 2013, this is the machine I would actually want. I am not simply choosing the most expensive part in every category. The priority is gaming performance, then reliability, cooling and enough headroom to keep the system useful for years. Read more
Apple Reveals iOS 7 and the New Mac Pro
Published:
Apple’s WWDC 2013 keynote on June 10, 2013 featured two redesigns that signaled different directions: iOS 7 replaced the skeuomorphic visual language that had defined iPhone software since 2007, and the new Mac Pro abandoned the aluminum tower form factor that had served professional desktops since the Power Mac era. iOS 7 was led by Jony Ive following the departure of Scott Forstall in October 2012 — Forstall had championed the leather-and-wood textures, torn paper, and felt-covered stitching that Apple called skeuomorphic design. iOS 7 replaced all of it with flat, translucent surfaces, a system-wide blur effect (processing layers beneath the current view as frosted glass), a new thinner system font (Helvetica Neue Light), redesigned icons with a consistent rounded-rectangle grid, and physics-based animations using UIKit Dynamics. New features included Control Center (swipe up from any screen to reach brightness, Wi-Fi, Bluetooth, and media controls without leaving the current app), AirDrop for peer-to-peer file transfer over Wi-Fi Direct and Bluetooth LE, and iTunes Radio as a streaming music service competing with Pandora. iOS 7 shipped September 18, 2013 alongside the iPhone 5s and iPhone 5c, and within four days Apple reported 74 percent of active devices had updated — the fastest adoption rate for any iOS version to that point. Read more
A Genius And Visionary Nicola Tesla
Published:
Nikola Tesla is one of those historical figures whose real achievements are impressive enough without adding legends. He was born in 1856 in Smiljan, in what is now Croatia, and moved to the United States in 1884. His work in electrical engineering became especially important during the development of alternating-current power systems. Read more
Google Introduces Android Studio
Published:
Google announced Android Studio at Google I/O on May 15, 2013 as a replacement for the Eclipse + ADT (Android Development Tools) plugin combination that had been the official Android development environment since Android’s first SDK release in 2008. The preview (version 0.1) was available immediately for download; Android Studio 1.0 stable shipped on December 8, 2014. The choice of JetBrains’ IntelliJ IDEA Community Edition as the base was a departure from Eclipse, which Google had promoted since 2008 but which suffered well-known performance problems under heavy Android project loads — long startup times, memory pressure on projects with many modules, and frequent crashes under refactoring operations. Read more
Google Releases Blink as a New Browser Engine
Published:
Google announced Blink on April 3, 2013 as a fork of the WebCore component of WebKit — the rendering engine Google had used in Chrome since its initial release on September 2, 2008. WebKit itself had origins in KHTML, the rendering engine developed by the KDE project for the Konqueror browser, which Apple forked in January 2003 to create Safari. Chrome had adopted WebKit because it was already mature, standards-compliant, and open-source, but Chrome immediately diverged from Safari’s architecture by using V8 (Google’s own JavaScript engine, developed separately) rather than WebKit’s JavaScriptCore, and by implementing a multi-process model in which each browser tab ran in a separate sandboxed renderer process. This multi-process architecture — fundamental to Chrome’s stability and security properties — created permanent architectural tension with WebKit, which had been designed for a single-process model. Chrome engineers working in the shared WebKit repository found themselves maintaining complex compatibility layers and workarounds to support Chrome’s multi-process design in a codebase whose primary consumers (Apple Safari) used a different architecture. Opera simultaneously announced it was also abandoning its long-running Presto rendering engine in favor of Blink, becoming the third major browser to share a Chromium-based engine alongside Chrome and the various Chromium-derived browsers already in existence. Read more
Raspberry Pi And The Joy Of Small Computers
Published:
The Raspberry Pi is one year old. It launched in February 2012, and in the time since, about a million boards have been sold — far more than the Raspberry Pi Foundation expected when they planned the initial production run of ten thousand. Read more
Samsung Launches the Galaxy S4
Published:
Samsung announced the Galaxy S4 on March 14, 2013, at a theatrical event at Radio City Music Hall in New York — 3,000 people attended — and launched it globally in April 2013, priced around $649 unlocked ($199–$249 on two-year contracts in the US). The S4 had a 5-inch Full HD (1920×1080) Super AMOLED display at 441 PPI, up from the Galaxy S3’s 4.8-inch 306 PPI display. Samsung shipped two different chipsets by region: a Qualcomm Snapdragon 600 (quad-core 1.9 GHz) in North America, Europe with LTE, and several Asian markets, and a Samsung Exynos 5410 (octa-core, but only 4 active at once using big.LITTLE) in other markets. Both had 2GB RAM. Read more
Ubuntu Announces Its Phone Interface
Published:
Canonical announced Ubuntu for Phones on January 2, 2013, with a promotional video and detailed specification document. The interface relied entirely on edge gestures: swiping right from the left edge revealed a launcher bar, swiping left from the right edge exposed a notifications panel, swiping up from the bottom edge revealed app controls, and swiping down from the top edge showed settings and quick toggles. There was no physical home button in the design; all navigation was gestural. Canonical intended to ship Ubuntu Phone on hardware in 2013 but the first commercially available devices — the BQ Aquaris E4.5 Ubuntu Edition and Meizu MX4 Ubuntu Edition — did not reach the market until February and November 2015 respectively. Read more
3D Printing And The Idea Of Downloading An Object
Published:
For most of my life, a digital file and a physical object have belonged to different worlds. A document can become a printout. A photograph can become a print. But a bracket, a gear, a custom case, or a replacement part for something discontinued — these required a factory, a workshop, or a person with specialized tools. Read more
Facebook Introduces Graph Search
Published:
Mark Zuckerberg announced Facebook Graph Search on January 15, 2013 at an event at Facebook’s Menlo Park headquarters, releasing it in a limited US English beta. Graph Search was led by Tom Stocky and Lars Rasmussen — Rasmussen was the co-founder and original engineer of Google Maps (sold to Google 2004) and creator of Google Wave (launched 2009, discontinued 2010), who joined Facebook in 2010. The system accepted natural-language queries structured around Facebook’s Open Graph data model: “Photos of my friends taken in Paris,” “Friends of friends who work at Google who went to Harvard,” “Restaurants in New York my friends have visited,” and “People who like The Dark Knight and Star Wars.” The queries were parsed into structured graph traversals over Facebook’s social graph — nodes (people, pages, photos, places, events, groups) connected by typed edges (likes, tagged-in, lives-in, works-at, attended, friend-of) — rather than keyword-based text retrieval. Bing web search was integrated as a fallback when queries lacked sufficient social data to produce useful results. Read more
2012
Google Compute Engine Expands Cloud Competition
Published:
Google announced Google Compute Engine in preview at Google I/O in June 2012 and expanded it to a limited general availability in December 2012, removing the invitation requirement and allowing any Google account to create virtual machines. Compute Engine launched with standard Linux VM types (n1-standard-1 through n1-standard-16, up to 16 vCPUs and 60 GB RAM), persistent disks, and snapshots for backup and clone — directly comparable to Amazon EC2 instance types that Amazon had been selling since 2006. Google priced standard instances approximately 10% below Amazon’s equivalent pricing at launch. Read more
The First ARM-Based Windows RT Devices Arrive
Published:
Windows RT devices went on sale October 26, 2012 alongside Windows 8, led by Microsoft’s Surface RT ($499 for 32GB, $599 for 64GB, with the Type Cover keyboard sold separately at $129). ASUS, Dell, Lenovo, and Samsung also launched Windows RT tablets the same day. All ran on ARM processors — primarily the NVIDIA Tegra 3 quad-core at 1.3 GHz in most devices including Surface RT — rather than the x86 CPUs that had defined Windows PCs since the early 1980s. The ARM chips offered better power efficiency: Surface RT quoted ten hours of battery life vs. roughly five to six hours on contemporary x86 Windows 8 ultrabooks. Read more
Remembering Atatürk on November 10, 2012
Published:
Every November 10, Turkiye pauses to remember Mustafa Kemal Ataturk. Read more
Windows 8 Is Released
Published:
Microsoft released Windows 8 on October 26, 2012 at $119.99 (Windows 8) and $199.99 (Windows 8 Pro) retail, with a $39.99 upgrade promotion for Windows XP/Vista/7 users available through January 31, 2013. The launch was the most radical Windows interface change since Windows 95: the Start menu was removed entirely and replaced by the Start Screen — a full-screen grid of live tiles that updated dynamically with app content (news headlines, weather conditions, email previews, calendar appointments). The Start Screen and all “Metro” (later “Modern UI”) apps ran in a full-screen tiled environment designed for touch input, contrasting with the traditional desktop environment that remained present as a separate mode. Switching between them required either a hot corner gesture, keyboard shortcut, or Start button click — on a non-touch laptop or desktop PC, this context switch felt unnatural, as the Start Screen’s large tile targets and swiping gestures were sized for finger input but controlled with a mouse cursor. Read more
The iPhone 5 Brings LTE to Apple Phones
Published:
Apple announced the iPhone 5 on September 12, 2012 at the Yerba Buena Center for the Arts in San Francisco, with units available for purchase September 21. It was the first iPhone to support LTE (4G Long Term Evolution) cellular connectivity, the first with a 4-inch display, and the first to use the Lightning connector. The hardware design was notably thin at 7.6 mm (versus iPhone 4S at 9.3 mm) and 112 grams — Apple’s marketing characterized it as “the thinnest and lightest iPhone ever.” The 4-inch Retina display (1136×640 at 326 ppi) maintained the same 326 ppi pixel density as the iPhone 4/4S while adding 176 rows of pixels, switching from a 3:2 to a 16:9 aspect ratio that matched the aspect ratio of widescreen video — meaning iPhone 5 could display 1080p HD video content without letterboxing on the sides. The Apple A6 SoC was Apple’s first processor using custom-designed CPU cores (codenamed Swift) rather than standard ARM Cortex implementations licensed from ARM Holdings, clocked at 1.3 GHz in a dual-core configuration, delivering approximately 2× CPU and 2× GPU performance versus the A5 in iPhone 4S. The A6 packaged 1 GB LPDDR2 RAM and a PowerVR SGX543MP3 GPU. LTE was supported through a Qualcomm MDM9615 modem, which for the first time combined GSM/UMTS and CDMA radio support in a single chip, enabling Apple to consolidate what had previously been separate hardware variants for AT&T and Verizon/Sprint into a smaller number of antenna configurations. Read more
Windows 8 Reaches RTM
Published:
Windows 8 reached RTM (release-to-manufacturing) on August 1, 2012, and Microsoft made the final build available to MSDN and TechNet subscribers on August 15. OEM PC manufacturers received the RTM build on the same date to begin factory installation, with the public sale date set for October 26, 2012 — nearly three years after Windows 7 launched in October 2009. The RTM announcement at a BUILD developer conference session marked the last point developers could verify their applications before retail availability. Read more
OS X Mountain Lion Is Released
Published:
Apple released OS X Mountain Lion (version 10.8) on July 25, 2012 at $19.99 — the first time a Mac OS X release had been priced below $29, and the lowest price Apple had charged for a Mac OS release. Mountain Lion was distributed exclusively through the Mac App Store, requiring OS X Snow Leopard (10.6.8) or later and an Apple ID, continuing the App Store-only OS distribution model Apple had introduced with Lion (10.7) in 2011. The release brought more than 200 features from iOS to the Mac, completing a deliberate convergence of the two operating system families that Apple had been pursuing since Lion. Messages replaced iChat: the new app combined the older AIM/Yahoo/Jabber protocols (iChat’s capabilities) with iMessage — Apple’s end-to-end encrypted messaging system introduced in iOS 5 (October 2011) — allowing Mac users to send messages to iPhone, iPad, and iPod touch users over data rather than SMS, with message delivery and read receipts. Notification Center (pulled directly from iOS 5’s design) arrived in the top-right corner of the Mac desktop: applications registered notifications that appeared as banners or alerts, collected in a notification sidebar revealed by swiping from the right edge of the trackpad. Reminders and Notes (both iOS 5 additions) arrived as separate Mac applications, syncing via iCloud. Game Center, Airplay Mirroring (mirror the Mac screen wirelessly to an Apple TV 2nd generation or later), Twitter integration in system-wide share sheets, and Facebook integration followed. Read more
The Day Windows Cleaned Up After Me
Published:
For years, I had been trained to respect Linux. At university, I had professors who loved it. Later, at work, I spent a great deal of time on embedded systems. So Linux was not some exotic operating system I occasionally installed out of curiosity — it was part of the technical world I knew. Windows development, on the other hand, was something I had done much less of. Then, in 2012, I had to use it, and because engineers are apparently incapable of simply using a new environment normally, I decided to experiment. Read more
What Should a Professional Chamber Actually Do?
Published:
In July 2012, I wrote about the newly formed Chamber of Computer Engineers. My first reaction was not philosophical — it was financial. I wrote, roughly: I will love this chamber as long as nobody comes to my door saying, “hands in your pockets.” In other words: please do not ask me for dues. That was the joke. But underneath it was a more serious question: what exactly should a professional chamber do for a computer engineer? Read more
The Gaming PC I Would Build in 2012
Published:
If I were building a gaming PC on July 1, 2012, this is the machine I would actually want. I am not simply choosing the most expensive part in every category. The priority is gaming performance, then reliability, cooling and enough headroom to keep the system useful for years. Read more
Microsoft Announces the Surface Tablet
Published:
Microsoft announced the first Surface tablets at a secret event in Los Angeles on June 18, 2012, with Steve Ballmer personally unveiling two models. Surface RT carried an NVIDIA Tegra 3 ARM processor and ran Windows RT — a version of Windows 8 that could not install legacy x86 desktop applications. Surface Pro was planned around an Intel Core i5 Ivy Bridge processor and full Windows 8, capable of running any existing Windows software. Both featured a 10.6-inch ClearType display at 1366×768, a built-in kickstand, and a magnetic cover attachment. The Touch Cover — a 3mm fabric slab with a touch-sensitive keyboard — and the Type Cover — a 5mm physical keyboard — were announced alongside the tablets, priced at $119 and $129 respectively. Read more
Google Introduces the Knowledge Graph
Published:
Google introduced the Knowledge Graph on May 16, 2012 in a blog post by CEO Larry Page titled “Introducing the Knowledge Graph: things, not strings.” The system launched initially for US English searches, with international rollout following over subsequent months. At launch, the Knowledge Graph contained approximately 500 million objects and 3.5 billion facts connecting them — people, places, books, movies, music, companies, and scientific concepts sourced from Freebase (acquired by Google in 2010), Wikipedia, the CIA World Factbook, and other structured databases. Read more
Google Drive Launches
Published:
Google launched Drive on April 24, 2012, offering 5 GB of free storage with integration into Google Docs, Sheets, and Slides. The service entered a cloud-storage market that Dropbox had pioneered in 2008 — Dropbox had approximately 50 million users by 2012 and offered 2 GB free — but Drive’s advantage was converting Google’s existing Docs user base and tying storage directly to a Google account that many people already owned for Gmail. Read more
NVIDIA Launches the Kepler GPU Architecture
Published:
NVIDIA launched the Kepler architecture on March 22, 2012 with the GeForce GTX 680 at $499. The GK104 die at the heart of the GTX 680 contained 3.54 billion transistors on a 28nm TSMC process, delivering 3.09 TFLOPS of single-precision compute. The GTX 680 matched or outperformed AMD’s Radeon HD 7970 (launched January 2012 at $549) in most games while drawing significantly less power — 195W vs. 250W under load. NVIDIA followed with the GTX 670 in May 2012 at $399, using a cut-down GK104 that offered most of the GTX 680’s performance at lower cost. Read more
The Unexpected Hanging Paradox
Published:
The Unexpected Hanging Paradox appears to prove that a judge’s announcement cannot be fulfilled — and then the announcement is immediately fulfilled. The contradiction is not in the judge’s statement. It is in the prisoner’s reasoning about knowledge. Read more
The Zenos Paradox
Published:
Zeno of Elea, a Greek philosopher from the fifth century BCE, created several paradoxes about motion that troubled thinkers for two thousand years and ultimately demanded new mathematics to resolve. Read more
The Berrys Paradox
Published:
Berry’s Paradox lives in the uncomfortable space between language and mathematics — the space where ordinary words meet formal definitions and produce results neither can handle alone. Read more
Raspberry Pi Goes on Sale
Published:
The Raspberry Pi went on sale on February 29, 2012 (deliberately chosen as a leap-day launch) through two licensed distributors, RS Components and Farnell/Element14 — both UK-based electronics distributors that had agreed to handle manufacturing orders. Within minutes of the 6 AM UK launch, traffic overwhelmed both distributor websites, knocking them offline as tens of thousands of people tried to place orders simultaneously. Pre-registration lists ahead of the launch had exceeded 700,000 names, massively outstripping the initial production run of 10,000 boards. The Model B priced at $35 (£22) included a Broadcom BCM2835 SoC (700 MHz ARM1176JZF-S ARM11 core with 256 MB RAM co-packaged on the same SoC die — later upgraded to 512 MB in October 2012), VideoCore IV GPU (capable of 1080p30 H.264 hardware decode and OpenGL ES 2.0), HDMI output, composite video jack, 3.5mm audio, two USB 2.0 ports, 100 Mbps Ethernet, an SD card slot for the boot OS, and a 26-pin GPIO header. The cheaper Model A ($25) had one USB port and no Ethernet. The board was assembled at the Sony UK Technology Centre in Pencoed, Wales — a deliberate choice to keep manufacturing in the UK despite lower-cost options in China. Read more
The Hilberts Paradox
Published:
Hilbert’s Hotel is one of my favourite ways to think about infinity. David Hilbert described the idea in a 1924 lecture in Göttingen; it was not published formally but was later popularised by the physicist George Gamow in his 1947 book One Two Three… Infinity. It remains the clearest illustration I know of why infinity does not behave like a very large finite number. Read more
The Birthday Paradox
Published:
The Birthday Paradox is not a real contradiction. It is a collision between probability and intuition — and the gap between them is large enough that the result consistently surprises people even after they have heard the explanation before. Read more
The Banach-tarski Paradox
Published:
The Banach-Tarski Paradox is a theorem in mathematics stating that a solid ball in three-dimensional space can be decomposed into a finite number of disjoint subsets, which can then be reassembled — using only rigid motions: rotations and translations — into two solid balls each identical in size to the original. One ball becomes two. No stretching, no scaling, no adding material. Read more
The Liar Paradox
Published:
Consider the sentence: “This statement is false.” It is only four words, but it creates a problem that has troubled philosophers and logicians for more than two thousand years. Read more
The Barber Paradox
Published:
The Barber Paradox is a simple story that carries the same logical problem as Russell’s Paradox. Imagine a village with one barber. The rule is this: The barber shaves all men in the village who do not shave themselves, and only those men. At first, the rule seems clear. Read more
Raspberry Pi Production Begins
Published:
Raspberry Pi production began in January 2012 at the Sony UK Technology Centre in Pencoed, Wales, with the first batch of Model B boards going through SMT (surface-mount technology) assembly and automated optical inspection before shipment to distributors RS Components and Farnell/Element14. The Raspberry Pi Foundation had spent 2011 refining the design to hit the $35 price point: an early prototype used an Atmel ATmega644 microcontroller, but the decision shifted to the Broadcom BCM2835 SoC (a multimedia processor Broadcom designed for IPTV set-top boxes and mobile phones) after Eben Upton, working at Broadcom, arranged access to the chip at educational pricing. The BCM2835 integrated a 700 MHz ARM1176JZF-S core (ARM11 architecture, ARMv6 ISA), a VideoCore IV GPU (supporting 1080p30 H.264 hardware decode, OpenGL ES 2.0, and OpenVG), and 256 MB of LPDDR2 RAM in a package-on-package stack above the SoC — a single component rather than separate CPU, GPU, and memory chips. The GPU’s hardware H.264 decoder was critical for education use cases like media centers (XBMC ran acceptably on the Pi using GPU-accelerated video) and for making a $35 device feel useful. Read more
Russells Paradox
Published:
Russell’s Paradox is one of those problems that looks harmless until one sentence turns mathematics against itself. Bertrand Russell discovered it around 1901 while reading Gottlob Frege’s Grundgesetze der Arithmetik — an attempt to derive all of arithmetic from pure logic. Russell noticed that Frege’s axiom allowing unrestricted set formation contained a contradiction, and he wrote a letter to Frege in June 1902 explaining it. Read more
The Invention Of Imaginary Numbers: Separating Math From Reality
Published:
Mathematics began very close to the physical world. People counted objects, measured land, followed the movement of stars, and solved practical problems. Read more
Imaginary Numbers
Published:
The name “imaginary number” sounds almost like a joke — how can a number be imaginary? The name reflects a history of suspicion. These numbers were useful before anyone was comfortable saying they were real. Read more
2011
HTML5 Becomes the Center of Modern Web Development
Published:
By December 2011, “HTML5” had become both a technical specification and a marketing term that redefined expectations for what the web could do. The W3C had published the first public working draft of the HTML5 specification in January 2008, and the final W3C Recommendation would not arrive until October 28, 2014; but browser vendors were implementing features incrementally without waiting for the specification to freeze. The practical trigger for accelerated HTML5 adoption had been Steve Jobs’s April 2010 “Thoughts on Flash” open letter, which explained why the iPhone and iPad would never support Adobe Flash and positioned HTML5 as the appropriate open standard for web video and interactive content. Flash had dominated rich internet applications since the late 1990s: at its peak in 2010, approximately 97 percent of web browsers had Flash installed and most video sites (including YouTube and Vimeo) delivered video through Flash players. HTML5’s <video> and <audio> elements offered native browser-level media playback, but the codec debates complicated adoption: Apple supported H.264/MPEG-4 AVC (proprietary, royalty-bearing via MPEG-LA, but hardware-accelerated in Apple devices); Mozilla and Opera refused to ship H.264 support and backed Ogg Theora (royalty-free); Google’s acquisition of On2 Technologies produced VP8, which Google open-sourced as WebM in May 2010. YouTube began offering an HTML5 video experiment in January 2010, but codec inconsistency meant developers still needed fallback Flash players for most of the following two years. Read more
Raspberry Pi Hardware Design Nears Production
Published:
By November 2011, the Raspberry Pi Foundation — founded by Eben Upton, Rob Mullins, Jack Lang, and Alan Mycroft at the University of Cambridge — had finalized the hardware design of the Raspberry Pi Model B and contracted manufacturing. The project had grown from a 2006 idea about a cheap programmable computer to teach Python to children. The final spec: Broadcom BCM2835 SoC (ARM11 core at 700 MHz plus VideoCore IV GPU capable of 1080p H.264 decode), 256MB of LPDDR2 RAM, HDMI output, composite video, 3.5mm audio, two USB 2.0 ports, 100Mbps Ethernet, SD card for storage, and a 26-pin GPIO header — all for a target retail price of $35 (Model B) or $25 (Model A, without Ethernet, one USB port). Read more
Remembering Atatürk on November 10, 2011
Published:
Every November 10, Turkiye remembers Mustafa Kemal Ataturk. Read more
Apple Introduces Siri on the iPhone 4S
Published:
Apple introduced Siri as a headline feature of the iPhone 4S at an event on October 4, 2011 — the first Apple event presided over by Tim Cook as CEO, and the day before Steve Jobs died on October 5. Apple had acquired Siri Inc. on April 28, 2010 for approximately $200 million; the original Siri app had been released on the App Store on February 4, 2010 as a standalone application that used natural language to book restaurants, find movies, and access web services through third-party API integrations with OpenTable, Fandango, and others before Apple acquired the company and removed the app from the App Store four weeks later. The Siri built into the iPhone 4S was substantially redesigned for deeper iOS integration and marketed with the label “Beta” — the first time Apple had shipped a product feature explicitly marked as unfinished, acknowledging that the system would improve over time rather than being presented as complete at launch. Siri ran on the iPhone 4S’s Apple A5 SoC (the dual-core ARM Cortex-A9 processor also used in iPad 2), with audio processing performed on-device but natural language understanding and response generation performed on Apple’s servers, requiring a data connection for most requests. Supported capabilities at launch included setting alarms and reminders, sending messages and emails, making calls, querying the weather, looking up restaurant information via Yelp, performing calculations, and answering factual questions through a Wolfram Alpha integration. Siri responded in synthesized speech and displayed text on screen simultaneously. Read more
When AltaVista Disappeared
Published:
Some technologies disappear loudly: a company announces the end, people write long farewell articles, old users share screenshots, and someone makes a documentary. Then there are the technologies that simply stop answering. Read more
When Mounting a Disk Became a Linux Lesson
Published:
There are technical problems that begin with a sentence so innocent that you already know the evening is lost. “I just need to write something to this disk.” That was more or less the situation. The disk was NTFS. The system was Linux. The tool involved was ntfs-3g. Nothing about this sounded dramatic — mount the disk, copy the files, continue with life. Instead, the computer gradually transformed into something that felt less like a workstation and more like a geological formation: slow, heavy, unresponsive. Every operation seemed to require negotiation. Read more
A Friday Night at IKEA Ankara
Published:
Some places begin as stores and somehow become part of a city’s social life. That was the strange thing about IKEA Ankara. In September 2011, I wrote about it with the kind of confidence I often had then when observing other people’s habits. I had tried the famous meatballs, and my conclusion was that anyone going there only to eat them needed to reconsider their life choices. I had also heard — and partly observed — that Mamak’s nightlife seemed to be happening there; the place was crowded with people who did not fit any single category. And when people said it was the biggest IKEA in Turkey, I dismissed that too: of course it was big, I thought, half of it was pillows and duvets. That was the entry — short, mocking, very certain. Read more
Windows 8 Is Shown to Developers
Published:
Microsoft unveiled Windows 8 at the BUILD developer conference September 13–16, 2011 in Anaheim, California, and gave all 5,000 attendees a Samsung Series 7 Slate tablet preloaded with a Windows 8 Developer Preview build. The conference was led by Steven Sinofsky (president of Windows division) and introduced the Metro interface publicly for the first time: a full-screen, tile-based Start screen designed for touch, replacing the Start menu that had been in Windows since Windows 95 in 1995. The tiles were live — updating dynamically with news, weather, email counts, and photos without opening the full application. Read more
Google Agrees to Acquire Motorola Mobility
Published:
Google announced an agreement to acquire Motorola Mobility on August 15, 2011 for $12.5 billion in cash — $40 per share, a 63% premium to Motorola Mobility’s closing price the previous Friday. Motorola Mobility had been spun off from Motorola Solutions (the enterprise and government radio business) only seven months earlier, in January 2011. The deal’s stated rationale from Google CEO Larry Page was protecting Android from patent litigation: Motorola Mobility held approximately 17,000 granted patents and 7,500 pending applications covering cellular communications, wireless protocols, and mobile device design, accumulated over Motorola’s 83-year history as a pioneer in wireless technology. Read more
The Bar Hidden Inside an Office Building
Published:
Some places feel strange before you even enter them — not because they are dangerous, not because they are beautiful, but because they are in the wrong building. In August 2011, I wrote a short entry about SSK İşhanı in Ankara. Years earlier, it had contained the second bar I had ever gone to, called Baraka. What I remembered most was not the music, not the drinks, not even the people — it was the fact that the bar was inside an office building. Read more
GitHub And The Social Side Of Code
Published:
Source-code hosting used to feel like infrastructure — a necessary utility, not a social space. Sourceforge (launched 1999) and Google Code (launched 2006) provided repositories, bug trackers, and download pages. They were functional. You submitted a project, it hosted the files, other developers found it through search. Read more
OS X Lion Moves Software Distribution Toward App Stores
Published:
Apple released OS X 10.7 Lion on July 20, 2011 through the Mac App Store for $29.99 — the first Mac operating system sold as a digital download only, with no DVD option at retail. Snow Leopard (2009) had been available as a $29 disc, itself a reduction from Leopard’s $129; Lion went further by eliminating physical media entirely for consumer upgrades. (A USB flash drive option was made available for machines without Mac App Store access, priced at $69, via Apple.com.) The Mac App Store had launched only seven months earlier, in January 2011. Read more
The Gaming PC I Would Build in 2011
Published:
If I were building a gaming PC on July 1, 2011, this is the machine I would actually want. I am not simply choosing the most expensive part in every category. The priority is gaming performance, then reliability, cooling and enough headroom to keep the system useful for years. Read more
Apple Announces iCloud
Published:
Apple CEO Steve Jobs announced iCloud at WWDC on June 6, 2011, describing it as a replacement for MobileMe — Apple’s $99/year subscription sync service that had launched in 2008 to widespread criticism for reliability failures and data loss. Jobs called MobileMe’s execution “not our finest hour” and announced that iCloud’s core services would be free with 5 GB of storage. iCloud launched alongside iOS 5 (October 12, 2011) and OS X Lion (July 20, 2011), which were prerequisites. The service replaced the Mac-as-hub model that had defined Apple’s digital hub strategy since 2001 (the original iTunes philosophy where the Mac was the master copy of music, photos, and calendars, with devices syncing via USB cable). In iCloud’s model, the cloud was the hub: contacts, calendars, email, and purchased content would propagate to every device automatically, with no cable required. MobileMe transitioned to iCloud and was shut down on June 30, 2012. Read more
Google Introduces Chromebooks
Published:
Google announced the first commercial Chromebooks at Google I/O on May 11, 2011, with two models going on sale June 15: the Samsung Series 5 Chromebook ($499 Wi-Fi / $549 3G) and the Acer Chromebook AC700 ($349). Both ran Chrome OS, a Linux-kernel operating system built around the Chrome browser as the exclusive application runtime, with no support for installing native Windows or Mac software. User accounts, bookmarks, apps, and settings synchronized to Google’s servers, meaning any Chromebook could be logged into and return the same environment within seconds. Read more
Ubuntu Introduces the Unity Desktop by Default
Published:
Ubuntu introduced the Unity desktop by default in 2011, making one of the boldest user-interface changes in the distribution’s history. For years, Ubuntu had been associated with a relatively familiar GNOME 2 desktop. Unity moved the experience toward a custom shell with a launcher, global menu ideas, search-centered navigation, and a layout influenced by smaller screens as well as traditional PCs. Read more
Firefox Moves to a Faster Release Model
Published:
Firefox’s move to a faster release model in 2011 was a major cultural change for one of the web’s most important browsers. Earlier browser releases often felt like large events separated by long waits. The web, however, was moving too quickly for that rhythm. New standards, security fixes, performance improvements, and developer needs demanded a shorter feedback loop. Read more
IBM Watson Defeats Jeopardy Champions
Published:
IBM’s Watson competed against Ken Jennings (winner of a 74-game consecutive win streak in 2004) and Brad Rutter (highest all-time Jeopardy prize earnings) in a three-game match taped in January 2011 and broadcast February 14–16, 2011. Watson won with $77,147 in combined scores versus Rutter’s $21,600 and Jennings’ $24,000. The $1 million first prize was donated to charity; IBM had spent approximately four years and $900 million on the project. Watson’s hardware comprised 90 IBM POWER7 servers with 2,880 processor cores and 16 TB of RAM, occupying 10 racks. The system had no internet access during the match; all 200 million pages of content (Wikipedia, Britannica, literary works, newspapers, and databases) were stored locally. Watson could analyze 200 million pages per second. Read more
Intel Introduces the Sandy Bridge Architecture
Published:
Intel introduced the Sandy Bridge architecture in 2011, and it quickly became one of the defining CPU generations of the early 2010s. For many desktop and laptop users, Sandy Bridge delivered the kind of improvement that could actually be felt: strong general performance, better efficiency, and integrated graphics that were becoming less of an afterthought. Read more
2010
The Quiet Value Of Making Backups
Published:
Backups are one of the least exciting parts of using a computer. Nothing new appears when a backup works. There is no faster program, no new feature, no visual change at all. We only create another copy of something we already have. Read more
Google Announces the Chrome Web Store
Published:
Google opened the Chrome Web Store on December 7, 2010, simultaneously with the launch of Chrome OS and the Cr-48, a matte-black developer pilot Chromebook shipped to invited developers. The store distributed three categories of Chrome products: browser extensions (scripts and UI overlays that modified Chrome’s behavior, already available since Chrome 4 in January 2010), browser themes (visual customizations), and web applications (a new category that came in two variants). “Hosted apps” were essentially metadata packages pointing to an existing web URL — an icon, name, description, and a link — that placed a shortcut in Chrome’s New Tab page app launcher and could request additional permissions like offline storage or notification APIs. “Packaged apps” were downloaded to the user’s local Chrome profile and could use Chrome’s local storage APIs, background page capabilities, and hardware access APIs without requiring a network connection, running through Chrome’s own execution environment rather than a traditional OS process. The distinction between hosted and packaged apps was frequently confusing to users: a hosted app for Gmail looked identical in the store to a packaged app for a game, but their offline behavior, storage access, and execution models were completely different. Read more
The Kinect Sensor Reaches Consumers
Published:
Microsoft launched Kinect for Xbox 360 on November 4, 2010 in North America, priced at $149.99 standalone or $299 bundled with an Xbox 360 and Kinect Adventures. It was the fastest-selling consumer electronics device at the time, selling 8 million units in 60 days according to Microsoft — faster than the iPhone or iPad had sold in comparable launch windows. The device sat below or beside a TV and required approximately 6 feet of clearance between the player and the sensor. Read more
Why I Keep Notes While Programming
Published:
When I am deep inside a programming problem, I often believe I will remember everything. The reason for this strange function is obvious. I am certain I will recall why I changed that setting. The error message will surely stay in my mind — I spent three hours fighting it. Read more
Remembering Atatürk on November 10, 2010
Published:
Every November 10, Turkiye pauses at 09:05 to remember Mustafa Kemal Ataturk. Read more
Microsoft Launches Windows Phone 7
Published:
Microsoft launched Windows Phone 7 on October 21, 2010 in the United States and Europe, introducing hardware from HTC (HD7, Surround), Samsung (Focus, Omnia 7), LG (Optimus 7), and Dell (Venue Pro). The launch was a complete architectural break from Windows Mobile 6.5 (released October 2009), which had attempted to put a desktop-like interface onto a touchscreen — using stylus-oriented menus, scroll bars, and desktop Windows conventions that were poorly suited to finger input. Windows Phone 7 instead adopted the Metro design language (later renamed Modern UI by Microsoft): a home screen of resizable Live Tiles that updated dynamically with current information (new emails, weather, news), horizontal panorama-scrolling UI sections within apps, large Helvetica Neue typography without frames or borders, and a consistent navigation model using three hardware buttons (Back, Start, Search) rather than on-screen navigation. The visual identity was derived partly from transit signage design and represented a deliberate departure from both iOS’s skeuomorphic icon grids and Android’s fragmented manufacturer skins. The user experience was praised by reviewers as distinctive and fluid; Windows Phone 7 was widely considered a genuinely good mobile operating system design at launch. Read more
Google Introduces Instant Search
Published:
Google launched Instant Search on September 8, 2010, announced by Marissa Mayer (VP of Search Products and User Experience) at a San Francisco press event. The feature updated search results dynamically as users typed each character, with results appearing typically after 2–3 characters without requiring the user to press Enter or click Search. The visual effect was a continuously changing results page driven by AJAX (XMLHttpRequest calls returning JSON/HTML partial updates) — the same technique that had enabled Gmail’s interface in 2004, but applied to search at much greater scale. Read more
Oracle Completes Major Changes Around Java and Sun
Published:
Oracle completed its $7.4 billion acquisition of Sun Microsystems on January 27, 2010, nine months after announcing the deal in April 2009 (delayed by EU antitrust review). Sun’s portfolio included Java (language, JVM, and OpenJDK), MySQL (then the world’s most widely used open-source database), Solaris OS, ZFS, VirtualBox, and SPARC server hardware. Oracle CEO Larry Ellison framed the acquisition as adding strategic technology assets to Oracle’s enterprise software and database stack; the most commercially sensitive asset was Java, given its ubiquity in enterprise applications, Android’s toolchain, and university education — and Oracle’s pending legal strategy. Read more
JavaScript Is Growing Up
Published:
JavaScript once felt like the language used for small tricks inside a web page — validate a form, open a popup, change an image on hover. The serious programming happened on the server. JavaScript was often treated as a detail that backend developers delegated to someone else, or worked around entirely. Read more
Node.js Gains Attention as Server-Side JavaScript
Published:
Ryan Dahl created Node.js and first presented it at JSConf EU on November 8, 2009, releasing version 0.1.100 that evening. By mid-2010, it had attracted significant developer attention following the release of Express.js (the first major web framework for Node, June 2010) and the creation of npm (Node Package Manager, January 2010) by Isaac Schlueter. Node 0.2 shipped in August 2010, stabilizing the API enough for production experiments. Microsoft, Joyent (which hired Dahl and was paying for the project), and Heroku were all actively supporting or evaluating Node deployments by late 2010. Read more
Before Likes Became Likes
Published:
Long before every platform put a heart, thumb, star, flame, clap, or reaction button under everything we wrote, people were already doing the same emotional mathematics: who liked this, who agreed, who noticed, who said something, and who said something but did not press the button. Read more
C Is a Manual Transmission
Published:
Years ago, I tried to explain the difference between C and Java with cars. C, I wrote, was like driving a manual transmission. Java was like driving an automatic. The comparison was not academically complete. It was not supposed to be. It was the kind of explanation you give when you have spent enough time with both languages to stop thinking only about syntax and start thinking about how they feel. Read more
The Day Everyone Became a Manager
Published:
There are many ways to become a manager: promotion, experience, performance, politics, seniority, a carefully planned career path. And then there is the method we discovered one day at a small company — everyone above you disappears. Read more
The Gaming PC I Would Build in 2010
Published:
If I were building a gaming PC on July 1, 2010, this is the machine I would actually want. I am not simply choosing the most expensive part in every category. The priority is gaming performance, then reliability, cooling and enough headroom to keep the system useful for years. Read more
The Modem Restart That Gave Me More Upload Speed
Published:
There was a period when internet problems were not background annoyances. They were events. You noticed every kilobit, knew what speed you were supposed to have, knew what speed you actually had, stared at modem statistics, restarted hardware, and called customer support. And sometimes, after thirty minutes of explaining a technical problem, the recommended solution was the most universal instruction in the history of consumer electronics: turn it off and turn it on again. Read more
Apple Introduces the iPhone 4
Published:
Apple announced the iPhone 4 at WWDC on June 7, 2010, with Steve Jobs calling the design the most beautiful thing Apple had ever made. It went on sale June 24 at $199 (16GB) and $299 (32GB) on two-year contracts. The phone replaced the iPhone 3GS’s plastic back with a glass-and-stainless-steel sandwich design: flat Corning Gorilla Glass on front and back, with a stainless-steel frame that doubled as the antenna system. The A4 processor — the same chip in the first iPad, manufactured by Samsung on a 45nm process — combined a 1 GHz ARM Cortex-A8 CPU and PowerVR SGX535 GPU in Apple’s first in-house SoC design. Read more
Liking Computers Is Not the Same as Studying Computer Engineering
Published:
There is a very understandable way to choose computer engineering. You like computers, you like the internet, you spend a lot of time in front of a screen — therefore, computer engineering. The logic feels almost perfect when you are eighteen. Read more
Never Use a Floating-Point Loop Counter
Published:
There are programming mistakes you make once and remember forever. For me, one of them involved a for loop and a floating-point variable. Years ago, I wrote a very simple warning: do not use double or float as the loop variable, otherwise arithmetic may not behave the way you expect, and you may find yourself stuck halfway through the problem. The exact wording was a little dramatic. The lesson was not. Read more
The Algorithm for Asking Someone for Directions
Published:
There is a special kind of problem that exists only because you decided to think about it too much. Asking someone for directions is a good example. You are somewhere unfamiliar, you do not know where to go, and there are people around. The sensible solution is obvious: ask one of them. That should be the entire algorithm — input: lost person, process: ask nearby person, output: direction, complexity: almost constant. Read more
Google Open-Sources the VP8 Video Codec
Published:
Google open-sourced the VP8 video codec on May 19, 2010 at Google I/O (May 19–20, 2010), alongside the announcement of the WebM media format, approximately three months after completing its $124.6 million acquisition of On2 Technologies (February 19, 2010). On2’s VP8 was the eighth-generation codec from the company; VP7 had been the video codec used in Skype calls. Google published the VP8 specification and reference encoder/decoder (libvpx) under a BSD-style license with an irrevocable royalty-free patent grant, making both the standard and the implementation free to use and implement without licensing fees or formal negotiations. WebM combined VP8 video with Vorbis audio (the royalty-free audio codec from Xiph.org) in a container derived from the Matroska (.mkv) container format, all under open, royalty-free terms. Mozilla, Google, and Opera committed immediately; Apple declined to add WebM to Safari, citing hardware H.264 decoder support in iOS devices and existing H.264 infrastructure; Microsoft announced it would support WebM in Internet Explorer 9 only if users installed the codec separately. Read more
The iPad And The Question Of What A Computer Is
Published:
Apple announced the iPad on January 27, 2010 and began selling Wi-Fi models on April 3. The device starts at $499 for 16GB. It runs a version of iPhone OS 3.2 on an Apple A4 chip. It has a 9.7-inch multi-touch display at 1024×768 pixels. It has no camera, no USB port, no SD card slot, no Flash support. Applications are installed only through the App Store. Read more
Apple Releases the First iPad
Published:
Apple CEO Steve Jobs unveiled the first iPad at a press event on January 27, 2010, introducing it as a device that filled “a third category between a laptop and a smartphone” for use cases including web browsing, email, photos, video, music, games, and e-books. The Wi-Fi-only models went on sale April 3, 2010; Wi-Fi + 3G models followed April 30, 2010. Pricing: $499 (16 GB Wi-Fi), $599 (32 GB), $629 (64 GB); Wi-Fi + AT&T 3G models added $130 to each tier. The 3G data plans were offered without contracts — $14.99/month for 250 MB or $29.99/month for unlimited, purchased directly from Apple — the first major smartphone-class device with month-to-month no-contract data. The iPad’s 9.7-inch IPS display at 1024×768 resolution (132 ppi) was the same aspect ratio as the original iPhone’s display but 7.8× larger in area. Apple’s A4 SoC (ARM Cortex-A8 at 1 GHz, PowerVR SGX535 GPU, 256 MB RAM) — designed by Apple’s chip team after its acquisition of PA Semi (April 2008, $278 million) — was the first chip Apple had designed in-house for a shipping product. The A4 was 9.56 mm thick, weighed 1.5 lb (Wi-Fi), achieved Apple’s claimed 10-hour battery life in typical use, and produced no audible fan noise (fanless thermal design inherited from iPhone). Read more
NVIDIA Announces Fermi GeForce GTX 480
Published:
NVIDIA launched the GeForce GTX 480 on March 26, 2010 at $499, introducing the Fermi architecture (GF100 chip) as its first GPU designed explicitly to serve dual roles as a graphics processor and a general-purpose parallel computing platform. NVIDIA had first revealed Fermi at its GPU Technology Conference in September 2009 with a detailed architecture paper, generating anticipation for what appeared to be a significant compute-capability advance. The GF100 die was manufactured on TSMC’s 40nm process and contained 3.0 billion transistors across a 529 mm² die — more than double the GT200 chip (1.4 billion transistors, 470 mm²) used in the GTX 280 (2008). The GTX 480 shipped with 480 CUDA cores (15 Streaming Multiprocessors × 32 cores each) out of GF100’s full complement of 512 cores (16 SMs), with one SM disabled — a common yield management practice for complex dies. Each SM in Fermi was substantially redesigned from the GT200: it gained a configurable L1 cache (16 or 48 KB, shared with shared memory in a 48/16 or 16/48 KB split), connected to a 768 KB unified L2 cache shared across the entire chip — a conventional memory hierarchy absent from earlier CUDA architectures that had instead relied on purely programmer-managed shared memory and no L2. The GTX 480 offered 1.35 TFLOPS of single-precision and 168 GFLOPS of double-precision floating-point performance, with the 1:8 ratio between DP and SP reflecting a consumer-grade configuration (the professional Tesla C2050/C2070 cards, launched in 2010, offered 1:2 DP:SP ratio for scientific computing). Read more
The Browser Is Becoming A Platform
Published:
For most of the web’s history, a browser was a program for reading documents. HTML was a document format. JavaScript was a scripting language for small interactive touches — form validation, a dropdown menu, an image that changes when you hover over it. Serious applications ran on the desktop. Read more
Microsoft Azure Becomes Commercially Available
Published:
Microsoft Azure became commercially available in 2010, marking Microsoft’s serious entry into cloud computing as a public platform. At the time, the cloud conversation was often associated with Amazon Web Services and fast-moving startups. Azure showed that the shift was also becoming unavoidable for established enterprise software companies. Read more
Google Introduces the Nexus One
Published:
Google unveiled the Nexus One at a press event on January 5, 2010, with the phone going on sale the same day at $179 with a T-Mobile contract or $529 unlocked — a price significantly higher than competitors expected from a carrier-free unlocked Android phone at the time. HTC manufactured the hardware; Google designed the software experience and sold the phone directly from google.com/phone without requiring a carrier store or contract. The phone ran Android 2.1 (Eclair) on a 1 GHz Qualcomm Snapdragon QSD8250 processor with 512MB RAM, a 3.7-inch 480×800 AMOLED display, and a 5MP camera. Read more
2009
Google Public DNS Launches
Published:
Google launched the public DNS resolvers at 8.8.8.8 and 8.8.4.4 on December 3, 2009, offering a globally operated alternative to the DNS recursive resolvers that most users received automatically from their Internet Service Providers. DNS (Domain Name System) is the distributed hierarchical database that translates human-readable domain names like example.com into IP addresses that computers use for routing, resolving the mapping by querying a chain of authoritative name servers from the DNS root through top-level domain servers to the specific domain’s authoritative server, and caching responses to avoid repeated queries. Before Google Public DNS, the primary alternative to ISP DNS was OpenDNS (founded 2006, primary address 208.67.222.222), which offered faster resolution and optional content filtering but monetized NXDOMAIN responses (redirecting mistyped domain names to search/ad pages). ISP DNS resolvers varied dramatically in quality: some performed NXDOMAIN hijacking (redirecting failed domain lookups to ISP-branded search pages rather than returning NXDOMAIN responses), some had poor cache TTL compliance (serving stale records beyond their intended expiry), some lacked DNSSEC validation support, and many had higher latency than a globally anycast resolver could achieve. Google’s anycast deployment announced the same 8.8.8.8 IP address from data centers in multiple locations worldwide, routing each user’s query to the geographically or topologically nearest Google data center — reducing average query latency from typical ISP values of 30-60 ms to Google’s claimed 11-12 ms median. Read more
When Solid State Drives Start To Feel Real
Published:
Hard drives have always made computers feel mechanical. You can hear them spin up at startup, click during seeks, and spin down when idle. A significant part of the computer is electronic, but for decades storage has depended on a physical disk rotating at 5,400 or 7,200 RPM and a read/write head moving across its surface. The average seek time for a 7,200 RPM hard drive — the time required to physically position the head over the right track — is around 8–10 milliseconds. Read more
Google Announces the Go Programming Language
Published:
Google announced Go as an open-source project on November 10, 2009, releasing the initial source code at golang.org. The language was created by Robert Griesemer, Rob Pike, and Ken Thompson — Pike and Thompson were the principal designers of Unix at Bell Labs in 1969 and Plan 9, and had strong opinions about language design informed by decades of systems programming. The project began September 21, 2007 as an internal Google effort motivated by frustration with C++ compilation times: a large Google service could take 30–45 minutes to compile in C++, making development iteration slow and build farms expensive. Go targeted C++-scale systems programming with compilation speeds measured in seconds. Read more
Remembering Atatürk on November 10, 2009
Published:
Every November 10, Turkiye remembers Mustafa Kemal Ataturk. Read more
Windows 7 Is Released
Published:
Microsoft released Windows 7 on October 22, 2009, at retail prices of $119.99 (Home Premium), $199.99 (Professional), and $219.99 (Ultimate) — with launch edition upgrade pricing of $49.99 for Home Premium available through the summer pre-sale. Windows 7 RTM (Release to Manufacturing) had been signed off on July 22, 2009 as build 7600.16385, available to OEM partners in August and to MSDN/TechNet subscribers before the retail launch. The release came 34 months after Windows Vista’s consumer release (January 30, 2007), which had been criticized heavily: Vista’s User Account Control dialog appeared 4–6 times during a typical session (Apple’s “Get a Mac” ads dramatized this repeatedly), hardware requirements were high (1 GB RAM minimum, 2 GB practical), driver compatibility was poor at launch (printer and scanner drivers in particular broke badly), and early versions had performance regressions on the hardware typical at the time of release. Windows 7 kept Vista’s underlying architecture — NT 6.1 kernel (Vista was NT 6.0), WDDM 1.1 graphics driver model, BitLocker, Windows Defender — but fixed the UAC behavior to prompt far less frequently (background configuration changes by administrators no longer triggered prompts), reduced minimum memory requirements, and benefited from three years of Vista driver maturation in the hardware ecosystem. Read more
Windows 7 Reaches RTM
Published:
Windows 7 reached RTM (Release to Manufacturing) on July 22, 2009, with build 7600.16385 signed as the final code. TechNet and MSDN subscribers could download the RTM build on August 6; OEM distribution — Dell, HP, Lenovo, Toshiba, and others receiving the build to preinstall on new PCs — began August 24. The public retail launch was October 22, 2009, with Home Premium priced at $119.99 upgrade / $199.99 full, Professional at $199.99 upgrade / $299.99 full. Microsoft also introduced a temporary three-license “Family Pack” at $149.99 for Home Premium to address household multi-PC licensing. Read more
Snow Leopard Reaches Gold Master Development
Published:
Mac OS X Snow Leopard reached its gold master stage in 2009, signaling that Apple’s next Mac operating system was ready for final release. Unlike many operating-system launches, Snow Leopard was not remembered mainly for dramatic visual changes. Its reputation came from refinement. Read more
The Comfort Of Long Forum Conversations
Published:
Stack Overflow launched last September and has grown quickly. It does something that older technical forums often failed at: it surfaces the best answer efficiently, rewards clarity, and is designed for people who want a fast, correct response. I use it. It is genuinely useful. Read more
Google Chrome OS Is Publicly Announced
Published:
Google publicly announced Chrome OS in a blog post on July 7, 2009, authored by Sundar Pichai (VP of Product Management) and Linus Upson (Engineering Director). The announcement came ten months after Google Chrome browser launched on September 2, 2008 — the OS would be the browser made into a full operating system. The post framed the project around the 2009 netbook market: millions of consumers were buying $299–$399 Windows XP or Windows 7 Starter laptops (Acer Aspire One, ASUS Eee PC) that were slow, prone to malware, and required constant maintenance. Chrome OS targeted that segment directly with a Linux-kernel OS in which the Chrome browser was the only execution environment — no native app installation, no file system users needed to manage. Read more
Microsoft Launches the Bing Search Engine
Published:
Microsoft launched Bing on June 3, 2009 — replacing Live Search, which had itself replaced MSN Search in 2006. CEO Steve Ballmer had positioned it as a “decision engine” rather than a search engine, emphasizing structured results for specific high-value query categories: flight prices and schedules (Farecast acquisition, 2008), restaurant and business reviews, health conditions and drug information, and shopping comparison. The homepage featured a rotating scenic background photograph with mouse-over hotspots linking to related searches — a deliberate visual departure from Google’s sparse interface. Read more
Google Wave Is Revealed
Published:
Google revealed Wave at Google I/O on May 27–28, 2009, in an 80-minute live demo by Lars Rasmussen — the same engineer who had previously co-created Google Maps with his brother Jens at Where 2 Technologies before Google acquired it in 2004. A “wave” was a hybrid object combining a shared document, a conversation thread, and a live collaboration session. Every participant saw other users’ keystrokes appear character by character in real time as they typed, rather than waiting for messages to be sent. Waves were nestled: sub-conversations (wavelets) could branch inside a parent wave, and any participant could edit any message in the conversation’s history, unlike email’s immutable sent-message model. Read more
Oracle Announces Its Plan to Acquire Sun Microsystems
Published:
Oracle announced its plan to acquire Sun Microsystems in 2009, and the news immediately felt larger than a normal enterprise acquisition. Sun had shaped major parts of computing culture: Java, Solaris, SPARC, NFS, workstations, servers, and important open-source projects all carried Sun’s influence. Read more
The Web Is Moving Into Our Pockets
Published:
Browsing the web on a phone used to mean visiting a stripped-down WAP page — minimal text, no images, a navigation designed around a numeric keypad. That was the mobile web until recently: a simplified shadow of the real internet, tolerated rather than enjoyed. Read more
Intel Launches Xeon 5500 Nehalem-EP
Published:
Intel launched the Xeon 5500 series (Nehalem-EP) on March 30, 2009, in a joint announcement with HP and Dell at their respective server events. The lineup ran from the Xeon E5504 (4 cores, 2.0 GHz, $188) to the Xeon X5570 (4 cores, 2.93 GHz, 8MB L3, $1,386), manufactured on Intel’s 45nm High-K metal-gate process. Nehalem-EP supported two-socket configurations — the “EP” designation indicated dual-processor capable — with quad-channel DDR3 memory per socket (up to 144GB per socket at launch). Read more
WhatsApp Is Founded
Published:
Jan Koum and Brian Acton incorporated WhatsApp Inc. in California on February 24, 2009. Both had recently left Yahoo — Koum after nine years as an infrastructure engineer, Acton after eleven years in various engineering and product roles. They had applied to Facebook in 2009 and been rejected. Koum’s initial concept was for an app that would show a short status beside each contact in your address book — not messaging, but presence: “at the gym,” “on a call,” “battery low.” Read more
Learning From Stack Overflow
Published:
A new programming question-and-answer site called Stack Overflow has become one of the places I enjoy visiting. The idea is simple. A programmer asks a clear question. Other programmers answer it. Useful answers move upward through voting, and the discussion stays focused on solving the problem. This sounds small, but it feels different from searching through long forum threads. Read more
Windows 7 Beta Shows a New Direction for Microsoft
Published:
Microsoft released the Windows 7 public beta on January 9, 2009 at CES, with Steve Ballmer announcing it on stage. Build 7000 downloaded 2.5 million times in the first week before Microsoft throttled access. The release came less than two years after Windows Vista’s January 30, 2007 launch, making it one of the fastest major Windows release cycles, driven by the need to recover from Vista’s poor reception: Vista’s UAC (User Account Control) generated constant prompts for routine tasks, it required 2+ GB RAM to run smoothly on hardware where XP used 512 MB, and driver compatibility failures at launch had left many users unable to use printers, scanners, and peripherals. Read more
2008
Cloud Computing Becomes a Serious Industry Strategy
Published:
By 2008, cloud computing had moved from a curious phrase into a serious industry strategy. The idea was still surrounded by marketing language, but the underlying shift was real: companies were beginning to treat computing resources as services that could be provisioned, scaled, and billed more flexibly than traditional infrastructure. Read more
Cloud Computing Still Feels Like A Strange Name
Published:
The phrase “cloud computing” is appearing more often in December 2008, and I am still not sure I like the name. The cloud usually means somebody else’s computers, located somewhere I cannot see, providing storage or processing through a network connection. Calling this a cloud does not add clarity — it is a metaphor that obscures the physical infrastructure behind the word. Read more
Google Releases Android Source Code
Published:
Google released the Android source code in 2008, shortly after the first commercial Android phone reached the market. The release mattered because Android was not presented only as a product; it was also framed as a platform that manufacturers, developers, and the wider industry could build around. Read more
Remembering Atatürk on November 10, 2008
Published:
Every November 10, Turkiye pauses to remember Mustafa Kemal Ataturk. Read more
Android 1.0 Arrives
Published:
Android 1.0 launched on October 22, 2008 with the T-Mobile G1 (HTC Dream) — the first commercial Android device, sold at $179 with a two-year T-Mobile contract or $399 unlocked. The hardware was built around a Qualcomm MSM7201A processor at 528 MHz with 192 MB RAM and 256 MB internal storage, a 3.17-inch 320×480 touchscreen, and a physical QWERTY keyboard that slid out from beneath the display (Android 1.0 had no virtual keyboard). The Android Market launched the same day with approximately 50 available applications, using a 70/30 revenue split (developer/Google) matching the App Store’s model. Google Maps, Gmail, YouTube, and Google Talk were preinstalled, with native Google Calendar and Contacts sync through Google’s servers. Read more
A Personal Website Feels Like A Notebook
Published:
A personal website is not a very practical object. It needs time and occasional maintenance. Pages break. Links go dead. Designs become outdated. Sometimes you spend an hour on a change that almost nobody will notice. None of this is a good use of an evening, rationally speaking. Read more
Google Launches the Chrome Browser
Published:
Google launched Chrome in 2008 into a browser market that already seemed crowded. Internet Explorer still held enormous share, Firefox had become the favorite of many power users, and Safari was important on Apple’s platforms. Chrome’s argument was different: the browser was no longer just a document viewer. It was becoming the runtime for applications. Read more
Intel Introduces the First Mainstream SSDs
Published:
Intel announced the X25-M (M for Mainstream) solid-state drive in August 2008, with retail availability beginning in September 2008 at $595 for 80GB and $945 for 160GB. The X25-M used 50nm multi-level cell (MLC) NAND flash manufactured by the Intel-Micron joint venture IMFT (IM Flash Technologies), with an Intel-designed proprietary controller. Specifications: sequential read 250 MB/s, sequential write 70 MB/s, and — most significantly — 35,000 random 4KB read IOPS. Contemporary 7,200 RPM hard drives delivered approximately 80–120 MB/s sequential with about 80–100 random IOPS, constrained by the physical seek time (averaging 8–12 milliseconds) required to move the read head to a new location. The SSD’s NAND flash had no moving parts; a read operation completed in microseconds rather than milliseconds, making the 35,000 IOPS figure roughly 400× better than HDD random access. Read more
Apple Opens the App Store
Published:
Apple opened the iPhone App Store on July 10, 2008 alongside iPhone OS 2.0, launching with approximately 500 applications — roughly 200 free and 300 paid — after Apple had announced the iPhone SDK (Software Development Kit) on March 6, 2008. The announcement reversed Steve Jobs’s original iPhone software strategy: at iPhone’s launch on January 9, 2007, Jobs had said the way to deliver third-party applications on iPhone was through web apps accessed in Safari, stating “the full Safari engine is inside of iPhone, and it runs on the Objective-C run-time, so you can write amazing Web 2.0 and Ajax apps.” Developer community pressure and the commercial opportunity of a native application platform led Apple to reverse course; the SDK was announced eight months after the original iPhone launch. The iPhone OS 2.0 SDK required Xcode (Mac only) and used Objective-C with Apple’s UIKit framework — the same framework used for Apple’s own iPhone applications — giving third-party developers access to the same UI components, animation APIs, multitouch gesture recognizers, and hardware interfaces (camera, accelerometer, GPS, Bluetooth) that Apple’s own engineers used. Apple provided code signing certificates through its iPhone Developer Program ($99/year for individuals, $299/year for enterprise), and all App Store applications had to pass Apple’s review process before being listed — the first time a mobile platform had implemented systematic pre-distribution review of applications. Read more
Git And The Strange Idea Of Distributed Version Control
Published:
Version control becomes important the first time a project grows large enough that “final”, “final2”, and “really-final” stop being funny file names. A proper system keeps the history of changes and allows us to return to an older version when something goes wrong. Read more
Mozilla Releases Firefox 3
Published:
Mozilla released Firefox 3 on June 17, 2008, accompanied by a “Download Day” campaign that set a Guinness World Record for most software downloads in a single 24-hour period: 8,002,530 downloads. Firefox 3 had been in development for nearly two years since Firefox 2 shipped in October 2006 and incorporated a rebuilt Gecko 1.9 rendering engine with full compliance for CSS 2.1, SVG, and Canvas, a new Places bookmarking system storing history and bookmarks in SQLite rather than flat MRK files, and the Awesome Bar (formally the Location Bar) — an autocomplete field that searched URLs, page titles, and tags simultaneously, presenting results by frecency (a combination of frequency and recency) rather than only alphabetically. Read more
Multi-Function Display
Published:
A Multi-Function Display, usually called an MFD, is one of the cockpit systems that interests me because it brings many different kinds of information onto a single screen. Instead of giving every function its own instrument, the display can change according to what the pilot needs at that moment. Read more
Microsoft Releases Windows XP Service Pack 3
Published:
Microsoft released Windows XP Service Pack 3 on May 6, 2008 via Windows Update — more than six and a half years after Windows XP RTM (released to manufacturing August 24, 2001) and fifteen months after Windows Vista had launched on January 30, 2007. SP3 bundled approximately 1,174 previously released hotfixes and security patches, including every update released from Windows XP SP2 (August 10, 2004) through early 2008. Several capabilities from Vista and the Windows Server 2008 codebase were backported: Network Access Protection (NAP) client support, which allowed enterprise networks to verify the security state of connecting machines and deny access to unpatched systems; Black Hole Router Detection, improving TCP reliability on networks with Path MTU Discovery problems; and updated cryptographic components including support for later TLS cipher suites. SP3 did not include Internet Explorer 7 (available separately since October 2006) or Windows Media Player 11 (available separately since November 2006) as mandatory updates, nor did it add Vista’s visual shell features — Aero Glass, the Sidebar, or User Account Control. This was by design: SP3 was a security and reliability update for the existing XP user base, not an attempt to bridge to Vista’s feature set. The service pack weighed approximately 316 MB as a standalone download, making it one of the largest service packs Microsoft had released. Read more
Google App Engine Opens to Developers
Published:
Google announced App Engine on April 7, 2008, opening a preview beta limited to 10,000 developers who could apply for access. The initial release supported Python 2.5 applications running in a sandboxed runtime with specific API access: the Datastore (a schemaless distributed database built on Google’s Bigtable), Memcache (distributed caching), URLFetch (for making HTTP requests to external services), and the Users service (authenticating users via Google accounts). Applications could not open raw sockets, write to the local filesystem, or spawn threads — restrictions that forced a specific application architecture while enabling Google to run many customers on the same underlying machines without isolation failures. Read more
Open Source Changed How I Learn
Published:
One of the nicest things about programming is that sometimes we can look inside the thing we use. With a textbook example, the code exists to illustrate a concept and has been cleaned and simplified for that purpose. With open-source software, the code exists to solve a real problem and carries the marks of that: old decisions left in place because changing them is risky, variable names chosen in 2003 that nobody has renamed since, workarounds for bugs in other people’s software, patches contributed by people who never met each other. Read more
Apple Releases the iPhone SDK
Published:
Apple released the iPhone SDK on March 6, 2008, at a press event where Steve Jobs presented it alongside the announcement of the App Store. The original iPhone had launched in June 2007 with no SDK; Jobs had told developers at WWDC 2007 that the way to build iPhone applications was to create web apps in Safari, arguing that iPhone’s full Safari browser meant web apps could be “as good as any native application.” Third-party developers found this inadequate — web apps had no access to the camera, microphone, accelerometer, GPS, contacts, or phone, and had no presence on the home screen. Read more
Microsoft Opens More of Its Server Protocol Documentation
Published:
On February 21, 2008, Microsoft announced that it would publish technical specifications for more than 30,000 pages of Windows client and server protocol documentation, committing to make these specifications freely available online. The announcement directly responded to the European Commission’s 2004 antitrust ruling, which had ordered Microsoft to provide “complete and accurate” interoperability documentation for its server protocols to allow competing server and workgroup software to interoperate with Windows clients. Microsoft had been slow to comply: the Commission fined Microsoft €280.5 million in July 2006 and €899 million in February 2008 for insufficient compliance — the €899 million fine was one of the largest EU antitrust penalties at the time, announced the same week as the documentation commitment. Read more
Netbooks And The Return Of The Small Computer
Published:
Personal computers have spent most of their history growing more powerful. Faster processors, larger screens, more RAM, stronger graphics: these improvements have been easy to understand and easy to market. The ASUS Eee PC, which launched in October 2007 at $299, is suggesting that another direction can be valuable: a computer that becomes useful by becoming smaller, cheaper, and simpler. Read more
MacBook Air Redefines the Thin Laptop
Published:
Steve Jobs introduced the MacBook Air at Macworld on January 15, 2008, sliding it out of a manila envelope to illustrate its thinness. The MacBook Air was 0.16 inches at its thinnest edge and 0.76 inches at its thickest, 3 lb (1.36 kg), with a 13.3-inch 1280×800 display. It launched at $1,799 with a 1.6 GHz Intel Core 2 Duo (a custom LV — low voltage — version using a 45nm die in a smaller package than standard laptops), 2GB RAM soldered to the board, an 80GB 4200 RPM PATA hard drive, or optionally a 64GB SSD for $999 more. Ports: one USB-A, one Micro-DVI, one 3.5mm headphone jack. No Ethernet, no optical drive, no removable battery. Read more
2007
Amazon Expands the Idea of Cloud Computing
Published:
Amazon Web Services launched Amazon S3 on March 14, 2006 ($0.15 per GB stored, $0.20 per GB transferred out) and EC2 public beta on October 25, 2006, at $0.10/hour for the smallest (m1.small: 1 vCPU, 1.7 GB RAM) through $0.80/hour for m1.xlarge (4 vCPUs, 15 GB RAM). Both ran on Xen hypervisors in Amazon’s data centers, exposing infrastructure that Amazon had built for its own retail operations. In November 2007, Amazon added SimpleDB (a schemaless, eventually-consistent key-value store) as AWS’s first managed database service. By the end of 2007, S3 was storing tens of billions of objects, and notable startups including Dropbox (which launched in April 2007 with S3 as its storage backend), SmugMug, Reddit, and Animoto had built their core infrastructure on AWS. Read more
Google Announces Android
Published:
Google announced Android in 2007 as part of the Open Handset Alliance, and the announcement immediately changed the shape of the mobile software conversation. The iPhone had already shown that the phone could become a general-purpose computing device. Android suggested that this new world would not belong to a single hardware maker or a single closed ecosystem. Read more
Remembering Atatürk on November 10, 2007
Published:
Every November 10, Turkiye remembers Mustafa Kemal Ataturk with silence, respect, and reflection. Read more
Ubuntu 7.10 Expands Desktop Linux
Published:
Ubuntu 7.10 “Gutsy Gibbon” was released on October 18, 2007, shipping Compiz desktop compositing effects enabled by default for systems with supported graphics hardware — the first Ubuntu release to make GPU-accelerated desktop effects standard rather than optional. Compiz on Gutsy required a compatible NVIDIA or ATI driver; the Restricted Drivers Manager (introduced in Ubuntu 7.04) simplified installing these proprietary drivers through a GUI dialog, eliminating the manual xorg.conf editing that had previously required technical knowledge. With Compiz enabled, Ubuntu 7.10 offered the rotating desktop cube, wobbly window animations, and the Expo plugin showing all workspaces simultaneously — effects that had attracted significant attention when Novell demonstrated them via Xgl in early 2006. Read more
Wikipedia And The Changing Habit Of Looking Things Up
Published:
When I was younger, looking something up usually meant finding the right book. A home encyclopedia — Britannica, Larousse, or whatever the household owned — had weight, order, and roughly 120,000 articles. If the information was not there, the search had to continue elsewhere. The library. A teacher. A more specialized reference. Read more
Apple Introduces the iPod Touch
Published:
Apple announced the iPod touch at “The Beat Goes On” event on September 5, 2007, with the first units available September 14, 2007 at $299 (8 GB) and $399 (16 GB). The iPod touch used the same Samsung S5L8900 ARM1176JZF-S processor at 412 MHz, the same 3.5-inch 320×480 multi-touch display, and the same Safari mobile browser as the original iPhone (launched June 29, 2007), but omitted the cellular radio, microphone (added in the 2nd generation September 2008), earpiece speaker, camera (first appeared in 4th generation September 2010), and GPS. Wi-Fi (802.11b/g) was retained, enabling Safari, YouTube, Google Maps, and the iTunes Wi-Fi Music Store — Apple’s feature allowing direct music purchase over Wi-Fi without connecting to a computer, first enabled on iPod touch at launch. The device was thinner than the iPhone at 8.0 mm versus 11.6 mm, because the cellular hardware occupied a significant portion of the iPhone’s internal volume. The iPod classic (announced the same day) continued as a hard drive-based music player, and a new thinner iPod nano was also announced, but the iPod touch represented Apple’s bet that the touchscreen interface would replace the click-wheel-and-physical-button paradigm for the premium iPod. Read more
VMware Fusion Brings Virtualization to Intel Macs
Published:
VMware Fusion brought VMware’s virtualization experience to Intel-based Macs in 2007, arriving at a moment when the Mac platform was changing quickly. Apple’s move from PowerPC to Intel processors made it far easier to run x86 operating systems on Mac hardware, and virtualization turned that capability into a practical daily workflow. Read more
Tactical Environment Software
Published:
Tactical environment software is an interesting example of software that must work with a large amount of changing information. In a military or operational environment, a user may need maps, weather data, positions, communication information, surveillance reports, and mission plans at the same time. The important problem is not only collecting this information. It is presenting the right part of it clearly enough to support a decision. Read more
Intel and the Industry Push Multi-Core Computing
Published:
Intel’s Core 2 Duo launched July 27, 2006, marking the mainstream shift to dual-core x86 desktop processors. By mid-2007, Intel expanded to quad-core with the Core 2 Quad Q6600 (January 2007, $851 initially, reduced to $266 by July 2007) and the Core 2 Extreme QX6850 at 3.0 GHz ($999 at launch, July 2007). AMD had been first to mainstream dual-core with the Athlon 64 X2 in May 2005, but Intel’s Core 2 microarchitecture (Conroe, 65nm) reclaimed performance leadership, and the move to quad-core in 2007 extended Intel’s lead as AMD’s Phenom quad-core didn’t ship until November 2007. The clock-frequency race that had defined the PC industry since the 1980s had effectively ended: Intel’s Pentium 4 Prescott (90nm, 2004) reached 3.8 GHz but generated over 100 watts of heat in high-performance configurations, making 4+ GHz architecturally untenable. Adding cores was the only remaining path to increased throughput within thermal limits — Steve Ballmer summarized this publicly: “core counts will double every two years” in place of clock frequency. Read more
The Phone Is Becoming a Computer
Published:
Two days ago, Apple began selling the iPhone in the United States. I have been reading everything written about it and trying to understand whether this is a genuinely different kind of device or only a well-designed version of things that already existed. Read more
The First iPhone Goes on Sale
Published:
The first iPhone went on sale in the United States on June 29, 2007 at 6 PM local time in each timezone. Steve Jobs had introduced it at Macworld on January 9, 2007, describing it as “an iPod, a phone, and an internet communicator” — three products in one. The original iPhone had a 3.5-inch 320×480 capacitive multi-touch screen (163 PPI), a 2MP camera, Samsung-manufactured ARM 11 processor running at 412 MHz, 128MB RAM, and either 4GB or 8GB of storage, priced at $499 and $599 respectively on a two-year AT&T exclusive contract. AT&T was the only US carrier; GSM-only, operating on EDGE (2.5G) data rather than 3G. Read more
Google Street View Appears
Published:
Google launched Street View on May 25, 2007 with panoramic street-level imagery from five US cities: San Francisco, New York City, Las Vegas, Miami, and Denver. The feature integrated into Google Maps as an alternative to the top-down satellite/map view, allowing users to navigate a 360-degree photograph taken from street level at specific addresses. The initial imagery was captured using cars equipped with camera systems from Immersive Media — their Dodeca 2360 spherical camera array captured images in 11 directions simultaneously — combined with GPS, inertial measurement units, and laser rangefinders to record the precise position and orientation of each photograph. Processing the raw camera output into Street View panoramas required multiple pipeline stages: synchronizing images from multiple cameras taken at slightly different times, lens distortion correction, color calibration across cameras, photographic stitching into equirectangular panoramas (2D projections of the full sphere), GPS/IMU fusion to place each panorama at an accurate geographic coordinate, and compression for web delivery. The initial Street View cars were Chevrolet Cobalt sedans with the camera assembly mounted on the roof; Google subsequently developed its own Street View camera generations, reaching 15 cameras plus additional sensors by later versions. Read more
Ubuntu 7.04 Feisty Fawn Is Released
Published:
Ubuntu 7.04, known as Feisty Fawn, was released in April 2007 during a period when desktop Linux was becoming more approachable for curious PC users. Ubuntu had already built a reputation for taking Debian’s strengths and presenting them through a friendlier installation and update experience. Feisty continued that work. Read more
Adobe Announces Creative Suite 3
Published:
Adobe announced Creative Suite 3 in 2007, and the release represented more than another bundle of design applications. It arrived at a moment when print design, web design, photography, video, and interactive media were converging inside the same professional workflows. Designers increasingly needed to move between formats rather than live inside one tool. Read more
History of Early Personal Computers and Gaming Consoles
Published:
When I look at today’s computers, it is easy to forget how young personal computing really is. Many of the machines that helped shape home computing and video games appeared only a few decades ago. They had tiny amounts of memory and very limited processors compared with modern systems, but they created a new relationship between people and computers. Read more
Virtualization Becomes a Mainstream PC Topic
Published:
Intel shipped VT-x (Vanderpool Technology) in mainstream processors beginning with the Pentium 4 600 series in June 2005, and AMD shipped AMD-V (Pacifica) with Athlon 64 X2 processors beginning in May 2006. By February 2007, both Intel Core 2 and AMD Athlon 64 processors sold in standard retail configurations included hardware virtualization support, making the technology available to any developer or power user who bought a modern PC. Before hardware virtualization, VMware had used binary translation — rewriting x86 privileged instructions in real time to safe equivalents — and achieved surprisingly good performance, but at significant engineering complexity. Intel VT-x and AMD-V added a new CPU execution mode (VMX/SVM mode) in which the hypervisor ran at a higher privilege level than ring 0, so guest OS code could execute native privileged instructions that would trap to the hypervisor for emulation rather than causing system failures. Read more
Learning Linux From A Live CD
Published:
Trying another operating system used to feel like a significant commitment. Installing it meant repartitioning a drive, potentially overwriting the existing system, configuring a bootloader, and accepting some risk that something would go wrong. For a curious user, the barrier was technical and emotional at once. Read more
Windows Vista Reaches Consumers
Published:
Windows Vista launched at retail for consumers on January 30, 2007, following a November 30, 2006 release for volume-license business customers. Retail editions: Home Basic ($199 upgrade), Home Premium ($239 upgrade), Business ($299 upgrade), and Ultimate ($399 upgrade). Vista “Premium Ready” hardware required a 1 GHz CPU, 1 GB RAM, 128 MB dedicated VRAM, and a DirectX 9-capable GPU with WDDM driver support for Aero Glass transparency effects — specifications that exceeded the average PC sold in 2005–2006, when many machines shipped with 512 MB RAM and integrated graphics. Read more
2006
AMD Moves Desktop Processors to 65 Nanometers
Published:
AMD’s move to 65 nanometer desktop processors in 2006 was not the kind of launch that instantly changed what users saw on the screen. There was no new operating system, no dramatic interface, and no single feature that made a computer feel transformed overnight. Still, process transitions like this are part of the quiet machinery behind every generation of faster and more efficient hardware. Read more
The Last Days Of The Floppy Disk
Published:
For most of my computing life, the floppy disk was one of the most normal objects in the room. Documents, homework, small programs, configuration files: these moved between computers on a square piece of plastic small enough to disappear into a pocket. The 1.44 MB capacity of the 3.5-inch high-density disk — the format that became standard in 1987 — felt reasonable because many of the things we wanted to move were small enough to fit. Read more
Windows Vista Launches for Business Customers
Published:
Microsoft released Windows Vista to volume-license business customers on November 30, 2006 (build 6000.16386 RTM), allowing enterprises to begin testing and deployment before the January 30, 2007 retail consumer launch. The release concluded a five-year development cycle that began as “Longhorn” in 2001 and was reset in August 2004 after the WinFS transactional filesystem — intended to make all Windows data addressable through a unified query interface — proved undeliverable. The reset moved Vista’s codebase onto Windows Server 2003 SP1, trading the ambitious WinFS vision for a more conservative but shippable release. Read more
Remembering Atatürk on November 10, 2006
Published:
Every November 10, Turkiye stops at 09:05 to remember Mustafa Kemal Ataturk. Read more
Firefox 2 Arrives
Published:
Mozilla released Firefox 2 on October 24, 2006, approximately two years after Firefox 1.0 (November 9, 2004) and one year after Firefox 1.5 (November 29, 2005). Firefox 2 ran on the Gecko 1.8.1 rendering engine and introduced features that addressed the practical experience of heavy browser users: session restore (automatically recovering all open tabs after a browser crash, the first major browser to offer this as a built-in feature), inline spell checking for text fields and text areas using the same red-underline convention as desktop word processors (checking as users typed in web forms and comment boxes), a redesigned tab bar with a close button on each individual tab (earlier Firefox versions had a single close button at the far right of the tab bar, requiring users to select a tab before closing it), a search plugins list in the search box dropdown allowing one-click installation of new search engines from any website, and enhanced phishing protection using Google’s Safe Browsing API to warn users before loading known phishing URLs. The anti-phishing feature was significant: IE 7 (released October 18, 2006, just six days before Firefox 2) also added phishing filtering as a response to the same threat, revealing that both browsers were tracking the same security problem simultaneously. Read more
Facebook Opens Registration Beyond Colleges
Published:
Facebook opened registration to anyone aged 13 or older with a valid email address on September 26, 2006. At that moment, the service had approximately 9.5 million registered users, almost entirely university and high school students. Myspace had approximately 100 million registered users at the same time and was the dominant social network; Friendster had peaked and was declining; LinkedIn served professionals. The open-registration decision came three weeks after a product change that would define Facebook’s growth: the News Feed, launched September 5, 2006, which displayed a continuous stream of friends’ activity (status updates, photo uploads, relationship changes, group joins) on the home page — replacing the model of navigating to individual profiles to discover activity. Read more
SPEC CPU2006 Redefines Processor Benchmarking
Published:
The Standard Performance Evaluation Corporation (SPEC) released SPEC CPU2006 in August 2006, replacing SPEC CPU2000 (released August 2000). The six-year gap between releases reflected how dramatically hardware had changed: CPU2000’s workloads fit comfortably in cache and DRAM configurations typical of 2000, making results in 2006 more reflective of L2 cache performance and compiler optimization than real-world application behavior. CPU2006 contained two suites: CINT2006 (12 integer workloads) and CFP2006 (17 floating-point workloads), all substantial programs compiled from source and run to completion rather than synthetic microbenchmarks. Workloads included gcc (compiler), bzip2 (compression), h264ref (H.264 video encoding), perlbench (Perl interpreter), namd (molecular dynamics simulation), milc (quantum chromodynamics lattice simulation), and astar (path-finding algorithm) — programs with diverse memory access patterns and instruction mix characteristics. Read more
RSS And The Quiet Pleasure Of Following The Web
Published:
The web is becoming larger faster than anyone can follow by visiting sites individually. Technorati, which tracks blog growth, estimates that a new blog is created every second in 2006 and that the blogosphere doubles roughly every six months. Technology sites, news pages, personal journals, project logs: the volume is growing faster than reading time. Read more
Intel Launches the Core 2 Duo
Published:
Intel launched the Core 2 Duo desktop processor family on July 27, 2006 under the codename Conroe, simultaneously releasing the Core 2 Extreme X6800 ($999, 2.93 GHz), Core 2 Duo E6700 ($530, 2.67 GHz), E6600 ($316, 2.4 GHz), and E6400 ($224, 2.13 GHz). The launch ended Intel’s four-year pursuit of the NetBurst microarchitecture that had defined Pentium 4 (launched November 2000) and Pentium D (launched May 2005). NetBurst had relied on very long pipeline stages — up to 31 stages in Prescott (2004) — to achieve high clock frequencies, reaching 3.8 GHz, but the deep pipeline imposed penalties on branch mispredictions, reduced instructions-per-clock efficiency, and generated substantial heat: Prescott-based Pentium 4 chips at 3.6–3.8 GHz consumed 84–115W TDP and required elaborate cooling solutions. Intel’s own engineers had begun calling the Prescott thermal situation “the burning building.” The Core microarchitecture Intel introduced in Core 2 derived instead from the Pentium M (Banias/Dothan) lineage — the laptop processor Intel Israel had designed in 2003 based on the P6 architecture of the Pentium Pro (1995) and Pentium III. Core used a 14-stage pipeline (versus Prescott’s 31), executed up to 4 instructions per clock cycle (versus Pentium 4’s 3-wide issue), and shared a single 4 MB L2 cache between both cores (versus the Pentium D’s two separate L2 caches with no direct inter-core communication path). The 65nm manufacturing process (Intel P1264 node) allowed higher transistor density and lower leakage versus the previous 90nm Prescott. Read more
Blu-ray Enters the Consumer Market
Published:
Blu-ray Disc reached consumers on June 20, 2006, when Samsung launched the BD-P1000 player in the United States at $999. Sony, Pioneer, and Panasonic shipped additional players in July and August 2006, priced between $799 and $1,299. Blu-ray’s 405nm blue-violet laser (from which the format took its name) allowed the pit size on disc to shrink to 0.32 micrometers, versus 0.74 micrometers for DVD’s 650nm red laser — producing 25 GB per layer versus DVD’s 4.7 GB. Dual-layer Blu-ray stored 50 GB, enabling 1080p video at 40+ Mbps bitrates alongside lossless Dolby TrueHD or DTS-HD Master Audio tracks, compared to DVD’s standard 9 Mbps MPEG-2 video. Read more
The Web Is Becoming More Social
Published:
The web feels different lately. A few years ago, most websites were closer to digital brochures. You visited, read something, and left. The website belonged to whoever built it, and visitors were an audience, not participants. Read more
Windows Vista Beta 2 Reaches the Public
Published:
Windows Vista Beta 2 reached a broad public audience in 2006 and gave many PC users their first serious look at Microsoft’s next desktop operating system. After the long life of Windows XP, expectations were high. Vista promised a more modern interface, stronger security boundaries, better search, and a platform prepared for newer hardware. Read more
Apple Introduces Boot Camp for Intel Macs
Published:
Apple released Boot Camp as a public beta on April 5, 2006, three months after the first Intel Macs shipped in January 2006. The software was a free download including a startup disk partition assistant, a Windows driver disc creator (packaging Apple’s drivers for the GPU, audio, keyboard, trackpad, iSight camera, and Bluetooth onto a CD for installation inside Windows), and firmware updates enabling a dual-boot menu at startup. Read more
Amazon S3 Launches and Changes Web Infrastructure
Published:
Amazon launched Simple Storage Service, better known as S3, in March 2006. At first glance it looked like a simple storage product: put files somewhere on the internet, retrieve them later, and pay for what you use. The deeper change was that storage stopped being something every small team had to buy, rack, monitor, replace, and over-plan before a product had even found its audience. Read more
XGL Brings 3D Effects to the Linux Desktop
Published:
Novell engineer David Reveman demonstrated Xgl and the Compiz window manager at a Linux conference in January 2006, and a video of the demo spread rapidly on the early internet through February 2006 — showing Linux desktop effects that looked more advanced than what Windows XP offered: windows that wobbled like jelly when moved, a rotating cube exposing four virtual desktops, transparent window decorations, and smooth minimizing animations that collapsed windows into the taskbar with a fire effect. The demo reached millions of viewers through Google Video and early YouTube, generating significant attention for desktop Linux among users who had never seen 3D compositing in any OS. Read more
AJAX And The Web Without Reloading
Published:
For a long time, using a website has followed a familiar rhythm. I click a link or submit a form, the browser waits, and then the whole page appears again. It works, but it often reminds me that I am moving from one document to another. Recently, some websites have started to feel different. Read more
Apple Begins the Intel Mac Era
Published:
Steve Jobs opened Macworld Expo 2006 on January 10, 2006 at the Moscone Center in San Francisco with the announcement that Apple had shipped its first Intel-based Macs — available for purchase that day: an updated iMac with Intel Core Duo at 1.83 and 2.0 GHz (17-inch at $1,299; 20-inch at $1,699) and the MacBook Pro, a new name replacing the PowerBook G4 line, at 1.67 and 1.83 GHz ($1,999 and $2,499). Jobs had announced the intention to switch from PowerPC to Intel at WWDC on June 6, 2005, setting a two-year transition timeline; the actual transition completed in 14 months. The processors were Intel’s dual-core Core Duo (codename Yonah) built on Intel’s 65nm process — the same platform used in Centrino laptop processors — with a shared 2 MB L2 cache and a 31W TDP, making it suitable for the iMac’s all-in-one thermal design and the thin MacBook Pro chassis. The Core Duo delivered substantially better performance per watt than the PowerPC G5 IBM had supplied for the Power Mac and the PowerPC G4 Freescale had supplied for the PowerBook; the G5’s heat output had prevented Apple from shipping a G5-based laptop, leaving the PowerBook G4 at 1.67 GHz as the maximum while Intel laptops had advanced significantly. Jobs demonstrated the iMac Core Duo running a video transcode “four to five times faster” than the Power Mac G5 iMac it replaced. Read more