data format: Amazon Ion

Amazon Ion is a high-performance, self-describing, typed data serialization format designed as a JSON superset, offering both a compact binary form for efficiency and a readable text format. Unlike rigid schema-based formats like Protocol Buffers, Ion allows flexible, dynamic data evolution, making it ideal for microservices and data storage.

Comparing Data Formats: Avro, Parquet, Ion and OpenAPI | by Ronald Ssebalamu | Medium

Key Comparisons with Similar Formats

• vs. JSON: Ion is a superset of JSON, meaning all valid JSON is valid Ion. However, Ion provides richer types (decimals, blobs, timestamps), better performance, and binary encoding.
• vs. Protocol Buffers (Protobuf) & Apache Avro: While Ion offers similar performance, it does not require a pre-defined schema to interpret data, allowing for easier, schema-less evolution, unlike the stricter, schema-bound nature of Protobuf or Avro.
• vs. MessagePack: Both are binary formats aimed at replacing JSON. Ion offers stronger typing (specifically for timestamps and decimals) and more extensive metadata handling compared to MessagePack.
• vs. YAML: Ion is more structured and efficient for machine processing, whereas YAML is primarily designed for human readability and configuration.

Amazon Ion

Amazon Ion is a richly-typed, self-describing, hierarchical data serialization format offering interchangeable binary and text representations. The text format (a superset of JSON) is easy to read and author, supporting rapid prototyping. The binary representation is efficient to store, transmit, and skip-scan parse. The rich type system provides unambiguous semantics for long-term preservation of data which can survive multiple generations of software evolution.

Ion was built to address rapid development, decoupling, and efficiency challenges faced every day while engineering large-scale, service-oriented architectures. It has been addressing these challenges within Amazon for more than a decade, and we believe others will benefit as well.

Available Languages: C – C# – Go – Java – JavaScript – Python – Rust
Community Supported: D – PHP – Ion Object Mapper for .NET
Related Projects: Ion Hash – Ion Schema
Tools: Ion CLI – Hive SerDe
Support: Discord

vs

• Ion (Amazon Ion) — JSON-like, binary + text, streamable
• BSON (Binary JSON, MongoDB)
• CBOR (Concise Binary Object Representation)

TOON | Token-Oriented Object Notation

Implementations | TOON

Local network windows web server

 Many dev tools can run simple test web server, but to be able to access them for other computers on local network need to configure Windows firewall. This is the command, need to run as admin.

netsh advfirewall firewall add rule name="Node Port 3000" dir=in action=allow protocol=TCP localport=3000

const http = require('http');
const fs = require('fs');
const path = require('path');

const PORT = 3000;
// Use '0.0.0.0' to listen on all network interfaces, not just localhost
const HOST = '0.0.0.0'; 

const server = http.createServer((req, res) => {
  // Build file path from current directory and requested URL
  let filePath = path.join(__dirname, req.url);
  // If the request is for a directory, try to serve index.html
  if (req.url === '/') {
    filePath = path.join(__dirname, 'index.html');
  }
  // Check if the file exists
  fs.readFile(filePath, (err, content) => {
    if (err) {
      // Handle file not found (404)
      res.writeHead(404, { 'Content-Type': 'text/plain' });
      res.end('404 Not Found');
    } else {
      // Serve the file with a 200 OK status
      // In a real-world scenario, you would want a more robust MIME type handling, 
      // but this is a minimal example.
      res.writeHead(200, { 'Content-Type': 'text/html' }); 
      res.end(content, 'utf-8');
    }
  });
});
server.listen(PORT, HOST, () => {
  console.log(`Server running at http://${HOST}:${PORT}/`);
  console.log('Serving files from directory:', __dirname);
});

or just this with python

> python3 -m http.server 8000 --bind 0.0.0.0
or

import http.server
import socketserver

PORT = 8000
# By using "", the server will listen on all interfaces (0.0.0.0)
HANDLER = http.server.SimpleHTTPRequestHandler
with socketserver.TCPServer(("", PORT), HANDLER) as httpd:
    print(f"Serving at port {PORT}...")
    # This will print the actual address the server is binding to
    print(f"Accessible via: http://{httpd.server_address[0]}:{PORT}") 
    httpd.serve_forever()

or with Go
package main
import (
	"log"
	"net/http"
)
func main() {
	// Serve files from the "static" directory.
	// You can change "static" to any directory path you wish to serve.
	fs := http.FileServer(http.Dir("./static"))
	// Handle all requests with the file server handler.
	// The "/" pattern in Go matches all request paths.
	http.Handle("/", fs)
	// Listen on all network interfaces ("") on port 8080.
	// Use ":8080" to listen on localhost only, or "" to listen on all addresses.
	port := ":8080"
	log.Printf("Serving on all addresses on port %s...\n", port)
	err := http.ListenAndServe(port, nil)
	if err != nil {
		log.Fatal(err)
	}
}

or with DotNet/C#

// Program.cs
var builder = WebApplication.CreateBuilder(args);
// Configure Kestrel to listen on all IP addresses (0.0.0.0) on port 8080
builder.WebHost.UseUrls("http://*:8080"); // Or use "http://*:*" to use default ports
var app = builder.Build();
// Enable serving static files (e.g., from the 'wwwroot' folder)
app.UseStaticFiles();
// Optional: Add a simple minimal API endpoint
app.MapGet("/", () => "Hello from Minimal File Server!");
app.Run();

AI: Government vs Anthropic

does this make any sense?

The Government just blacklisted Anthropic... - YouTube by Matthew Berman

Core Issues and Disagreement

• Safety Constraints: Anthropic, known for focusing on AI safety, placed restrictions on its technology, prohibiting its use in "killer robots" (fully autonomous weapon systems) and for spying on American citizens.
• Pentagon Demands: The Department of Defense (DoD), led by Secretary Pete Hegseth, demanded unrestricted"free rein" to use the technology for "any lawful use" refusing to accept these safety guardrails.
• The Breakdown: Following negotiations, Anthropic refused to lift its safeguards, leading the Pentagon to deem the company a "supply chain risk"

Statement from Dario Amodei on our discussions with the Department of War \ Anthropic

Two such use cases have never been included in our contracts with the Department of War, and we believe they should not be included now:

• Mass domestic surveillance
• Fully autonomous weapons

Trump has ordered government agencies to stop using Anthropic AI tools @BBC

Why the Trump administration is clashing with AI-firm Anthropic - YouTube @PBS

Software Stocks Implode, Claude's Hit List, State of the Union Reactions, Trump's Tariff Pivot - YouTube  @ All-In Podcast - YouTube

Gmail -= POP3

Google removed POP3 access to Gmail accounts, used by some email clients;
There is still IMAP protocol available, potentially more secure.

 Google’s Gmail Upgrade—Millions Of Accounts Now At Risk

Google Confirms Bad News For Gmail Users—Older Accounts Now Blocked

Build Your Own Gmail Reader with Node.js and Google OAuth2! - YouTube

Send and receive email in NodeJS and JavaScript (and create new email addresses) - YouTube

JS Web API: console.table()

 console: table() static method - Web APIs | MDN

The console.table() static method displays tabular data as a table.

console.table(data) console.table(data, columns)

The data to display. This must be either an array or an object. Each item in the array, or property in the object, is represented by a row in the table. The first column in the table is labeled (index) and its values are the array indices or the property names.

If the elements in the array, or properties in the object, are themselves arrays or objects, then their items or properties are enumerated in the row, one per column.

Note that in Firefox, console.table() is limited to displaying 1000 rows, including the heading row.

// an object whose properties are objects

const family = {};

family.mother = new Person("Janet", "Jones");
family.father = new Person("Tyrone", "Jones");
family.daughter = new Person("Maria", "Jones");

console.table(family);

(index)	firstName	lastName
daughter	'Maria'	'Jones'
father	'Tyrone'	'Jones'
mother	'Janet'	'Jones'

Languages => Web Assembly (WASM)

appcypher/awesome-wasm-langs: 😎 A curated list of languages that compile directly to or have their VMs in WebAssembly @GitHub

WebAssembly, or wasm for short, is a low-level bytecode format that runs in the browser just like JavaScript. It is designed to be faster to parse than JavaScript, as well as faster to execute which makes it a suitable compilation target for new and existing languages.

This repo contains a list of languages that currently compile to or have their VMs in WebAssembly(wasm)

• README
• Code of conduct
• Contributing

Awesome WebAssembly Languages 

WebAssembly, or wasm for short, is a low-level bytecode format that runs in the browser just like JavaScript. It is designed to be faster to parse than JavaScript, as well as faster to execute which makes it a suitable compilation target for new and existing languages.

This repo contains a list of languages that currently compile to or have their VMs in WebAssembly(wasm) 

Contents

• 🐔 - In Production.

  • C
  • C++
  • Rust
  • Go

• 🐥 - Stable for production usage.

  • .Net
  • AssemblyScript
  • C#
  • Dart
  • TypeScript
  • WebAssembly
  • Zig
    ...

WebAssembly | MDN

Web Assembly should be a default binary target - DEV Community

(login required, free)

it is more efficient, and it is standard, can run anywhere, that is why

Coding raw WebAssembly Text - Dive into Wat - YouTube

Creating Components - The WebAssembly Component Model

Android PC: Android + ChromeOS ("Aluminum OS " 2026)

Google confirms Android-PC merger launches next year | The Tech Buzz

Excited About Android for PC? This Leak Gives Us First Glimpse of New OS | PCMag

Google plans to replace ChromeOS with a new platform called Aluminium OS, combining the best of its existing computing software with Android. We may now have seen the first glimpse of the software, but not as Google would have intended.

Bill Gates PANICS as the World ABANDONS Windows for Android! - YouTube

huaweicentral.com/after-harmonyos-pc-google-confirms-to-launch-android-pc-in-2026/

Book: Competitive Programming CP4

To "compete" with AI, one may need this book :)

Competitive Programming Book

"This Competitive Programming book, 4th edition (CP4) is a must have for every competitive programmer. Mastering the contents of this book is a necessary (but admittedly not sufficient) condition if one wishes to take a leap forward from being just another ordinary coder to being among one of the world's finest competitive programmers."

Competitive Programming Book CP4

Competitive Programming 4 - Book 1: The Lower Bound of Programming Contests in the 2020s: Halim, Steven, Halim, Felix, Effendy, Suhendry: 9781716745522: Amazon.com: Cell Phones & Accessories

Competitive Programming 4 - Book 2: The Lower Bound of Programming Contests in the 2020s: Halim, Steven, Halim, Felix, Effendy, Suhendry: 9781716745515: Amazon.com: Cell Phones & Accessories


 stevenhalim/cpbook-code: CP4 Free Source Code Project (C++17, Java11, Python3 and OCaml) @GitHub

"objective in writing this book is similar with... to further improve humanity by training current students to be more competitive in programming contests. 

The possible long term effect is future Computer Science researchers who are well versed in problem solving skills. 

We use C++ (primary), Python (secondary), Java (tertiary), and OCaml (optional) code to illustrate the algorithmic concepts, i.e., we dislike vague pseudo-code commonly found in many other Computer Science textbooks. 

We also built and heavily use our-own visualization tool: VisuAlgo to help explain the data structure and algorithm concepts to our book readers and beyond."

Leetcode/competitive programming by steven & felix halim.pdf at master · jaytorasakar8/Leetcode

Competitive-Programming/cpResources/cpBooks/CompetitiveProgramming4Book2.pdf at main · QKnot/Competitive-Programming

visualgo: Visual Data & Algorithms

 visualising data structures and algorithms through animation - VisuAlgo

"visualising data structures and algorithms through animation"

"LEGO"-like "click" houses 1/2 price ?

 THE LEGO CLIK REVOLUTION: Why Robots Are Building the Future of Housing - YouTube

AI: Gemini 3.1 Pro: high intelligence, low price

Google's New AI Is Smarter Than Everyone's But It Costs HALF as Much. Here's Why They Don't Care. - YouTube

What's really happening when Google ships the smartest AI model on the planet, prices it at a seventh of the competition, and doesn't care if you keep using Claude or ChatGPT? The common story is that this is another benchmark race—but the reality is more interesting when the company generating $100 billion in annual free cash flow is playing a fundamentally different game.

javascript package manages: pnpm vs npm

pnpm vs npm | pnpm

"npm maintains a flattened dependency tree as of version 3. This leads to less disk space bloat, with a messy node_modules directory as a side effect.

On the other hand, pnpm manages node_modules by using hard linking and symbolic linking to a global on-disk content-addressable store. This lets you get the benefits of far less disk space usage, while also keeping your node_modules clean. There is documentation on the store layout if you wish to learn more.

The good thing about pnpm's proper node_modules structure is that it "helps to avoid silly bugs" by making it impossible to use modules that are not specified in the project's package.json."

I Finally Changed Package Managers - YouTube by Theo - t3․gg - YouTube

Code Is Cheap. Software Is Expensive?

What happens now? - YouTube by Theo - t3․gg - YouTube

Code Is Cheap Now. Software Isn’t. — Chris Gregori

1. The Collapse of the Barrier to Entry

• Point: Building functional code is no longer a privilege held only by professional developers.

• Justification: Tools like Claude Code and Claude Opus 4.5 allow non-technical "builders" to architect their own tools (e.g., subscription trackers or niche Chrome extensions) simply by describing what they need.

2. The Rise of "Disposable" Personal Software

• Point: We are moving away from permanent SaaS platforms toward ephemeral "scratchpad" software.

• Justification: Because the cost and time to generate code have dropped so low, it is now viable to build a tool for a single, one-off task and discard it immediately after, similar to how one uses a spreadsheet for quick calculations.

3. Code is Cheap; Software is Expensive

• Point: Writing lines of code is easy, but building robust, real-world software remains difficult.

• Justification: LLMs can generate syntax, but they don't solve for "the friction of the real world," such as maintenance, edge cases, UX debt, and breaking changes in third-party APIs or DOM structures.

4. The Shifting Role of the Engineer

• Point: The value of a software engineer has moved from the "how" (syntax) to the "what" and "why" (systems).

• Justification: AI hides complexity rather than managing it. Engineers are still required for architectural rigor—knowing how to manage distributed caches, rate-limiting, and data security—tasks that AI-native "weekend apps" often fail to address.

5. Distribution is the New Bottleneck

• Point: Engineering leverage is no longer a primary differentiator for a business.

• Justification: Since anyone can generate a product quickly, the competitive advantage has shifted to factors that are harder to automate: taste, timing, and a deep understanding of the audience. The hardest part is no longer building, but "finding a way to get people to care."

AI-Powered Flutter App with Antigravity

 How to Build an AI-Powered Flutter App with Google Antigravity: A Hands-On Tutorial @freecodecamp

use Antigravity to let AI agents plan, write, test, and show video walkthroughs of your app. This “vibe coding” style means that you describe what you want, review plans, and approve changes – all while agents handle the heavy lifting.

AI Agents: machine that builds the machine

Build the machine that builds the machine | LinkedIn
Paul Dix
Cofounder and CTO at InfluxDB, series editor on Addison Wesley's "Data & Analytics

Building the machine that builds the machine with Paul Dix, Co-founder & CTO at InfluxDB (Changelog Interviews #676)

Paul Dix joins us to discuss the InfluxDB co-founder’s journey adapting to an agentic world. Paul sent his AI coding agents on various real-world side quests and shares all his findings: what’s going to prod, what’s not, and why he’s (at least for a bit) back to coding by hand.
Update: He’s back to letting the AIs write code, but with a lot more oversight. For now

Paul Dix on X: "2026: The Great Engineering Divergence" / X

Message from InfluxDB

AI tool: SmolAgents by Hugging Face

 Understanding AI Agents and Agentic AI: Concepts, Tools, and Implementation with SmolAgents @CodeMag

Today, there are numerous frameworks available for building AI agents, each offering its own philosophy and strengths. Popular options include SmolAgents by Hugging Face, LangChain, LangGraph (both by LangChain, Inc.), and Microsoft’s AutoGen. Although all of them support agent development, they differ in complexity, flexibility, and intended use cases.
In this article... focus on building agents using SmolAgents, a lightweight and developer-friendly Python framework.

huggingface/smolagents: 🤗 smolagents: a barebones library for agents that think in code. @GitHub

AI models: Gemini 3.1 Pro, Sonnet 3.6

Google launches Gemini 3.1 Pro, retaking AI crown with 2X+ reasoning performance boost | VentureBeat

The most significant advancement in Gemini 3.1 Pro lies in its performance on rigorous logic benchmarks. Most notably, the model achieved a verified score of 77.1% on ARC-AGI-2.

This specific benchmark is designed to evaluate a model's ability to solve entirely new logic patterns it has not encountered during training.

This result represents more than double the reasoning performance of the previous Gemini 3 Pro model.

• Input Price: $2.00 per 1M tokens for prompts up to 200k; $4.00 per 1M tokens for prompts over 200k.
• Output Price: $12.00 per 1M tokens for prompts up to 200k; $18.00 per 1M tokens for prompts over 200k.
• Context Caching: Billed at $0.20 to $0.40 per 1M tokens depending on prompt size, plus a storage fee of $4.50 per 1M tokens per hour.
• Search Grounding: 5,000 prompts per month are free, followed by a charge of $14 per 1,000 search queries.

Introducing Sonnet 4.6 \ Anthropic

Claude Sonnet 4.6 is most capable Sonnet model yet. It’s a full upgrade of the model’s skills across coding, computer use, long-context reasoning, agent planning, knowledge work, and design. Sonnet 4.6 also features a 1M token context window in beta.

For those on our Free and Pro plans, Claude Sonnet 4.6 is now the default model in claude.ai and Claude Cowork. Pricing remains the same as Sonnet 4.5, starting at $3/$15 per million tokens.

Anthropic just dropped Sonnet 4.6... - YouTube Matthew Berman

DotNet/C#: DB => Mermaid charts

 SimonCropp/DbToMermaid: Generate Mermaid ER diagrams from SQL Server databases or Entity Framework Core models. @GitHub

Entity Relationship Diagrams | Mermaid

Docker Sandboxes (for AI code)

Docker Sandboxes | Docker Docs

Docker Sandboxes lets you run AI coding agents in isolated environments on your machine. Sandboxes provides a secure way to give agents autonomy without compromising your system.

Why use Docker Sandboxes

AI agents need to execute commands, install packages, and test code. Running them directly on your host machine means they have full access to your files, processes, and network. Docker Sandboxes isolates agents in microVMs, each with its own Docker daemon. Agents can spin up test containers and modify their environment without affecting your host.

 cd ~/my-project
 docker sandbox run AGENT

Replace AGENT with your preferred agent (claude, codex, copilot, etc.). The workspace defaults to your current directory when omitted. You can also specify an explicit path:

 docker sandbox run AGENT ~/my-project

Hanselminutes Technology Podcast - Fresh Air and Fresh Perspectives for Developers - Run your AI Agent in a Sandbox, with Docker President Mark Cavage