Recent Posts

• 2026-10-17

  Lotus Improv Pre-History

  In the last few days I've been asked a couple of time about Lotus Improv. I think it was triggered by a wonderful Improv analysis and retrospective by Christopher Drum who I never met, but really dug deep in his research. It was great to see such in depth attention to this app from 40 years ago. I mean really, that's research! I was asked on the discussion thread about where the ideas for the Pivot Table came from. Instead of giving a throwaway answer, I sat back and relived those early days...

  How the Pivot Table Came About

  Just for the heck of it,...

  Read more →
• 2026-09-09

  How I code today

  Over the last three months I have written thousands of lines of code and yet I didn't write any of them, claude.ai wrote them. The world of programming, software engineering, computer science, in my opinion has been turned upside down in a matter of months.

  How I code today

  I've created a process within the world of Claude Code which I use to design, plan and implement code. Many people have created similar ones and so instead of describing mine, in detail, let me explain the general idea. The heart of it is a series of feature specs (in individual numbered markdown files) and for each feature spec a task list (in a different markdown file as well) ...

  Read more →

  #ai #software-engineering
• 2026-09-08

  The Throw Keyword was a Mistake - By Chris Fox

  Originally Posted on: https://hackernoon.com/the-throw-keyword-was-a-mistake-l9e532di

  I came across this provocative piece arguing that the throw keyword was a mistake in language design, and while I don't fully agree with the conclusion, it raises some genuinely worthwhile caveats about exception handling that are worth sitting with. It immediately reminded me of Dijkstra's classic "Go To Statement Considered Harmful" — that same energy of challenging an accepted construct that most developers never think twice about.

  #programming #design
• 2026-05-08

  GPT in 60 Lines of NumPy - by Jay Mody

  Interesting article: GPT in 60 Lines of NumPy | Jay Mody. Implementing a GPT model from scratch in NumPy. This is a detailed article including python source code. Ive skimmed it and it does explain a lot. But even without my being a neural net, ai, machine learning, gpt, chatgpt expert, i think i would learn a lot. Also the article is very well written presented.

  Note: written a few years ago before we know how far the "AI thing" was going to go!

  #ai
• 2026-04-30

  Using Codex to create a Dockerfile from a running system

  I've been dreading this: My robot's rasberry pi 5 was built months ago. I have been using it a lot, installing, updating and changing code and config. All my code is separately in github repos. But there is a lot more state than the github repos. It is in config files, pip installs, and so on. I started to worry about the scenario where the MicroSD on the Pi just bites the dust. This does happen, randomly, and occasionally.

  

  I could make a full and faithful copy of the MicroSD. It's a little time consuming but not hard. I could start over with a clean Ub...

  Read more →

  #ai #robotics