✉️ Not subscribed yet? Subscribe to the Newsletter
Future of Coding Weekly 2025/08 Week 3
2025-08-17 23:33
🛸 A moldable Common Lisp object inspector 🧑🔬 Trying out "Reflective Prompt Evolution Can Outperform Reinforcement Learning" 📄 Hierarchical File Systems are Dead
Share Your Work
🗨️ guitarvydas: 📝 De-Spaghettifying Control Flow & 📝 Towards Asynchronous Programming Workflows and Languages
In my mind, CPUs implement subroutines, and, "functions" are but artificial constructs built on top of subroutines. Here are some more attempts to explain this hardware-centric view (see, also, the playlist of short videos in the first article).
📝 De-Spaghettifying Control Flow
📝 Towards Asynchronous Programming Workflows and Languages
🗨️ Robin Heggelund Hansen:
I'm working on a language called Gløgg, which is a fully declarative, relational language where the source code is stored in a sqlite3 database as AST.
I now just got vite support and (proof of concept) dom integration working.
I'll be doing a whole presentation on Gløgg at the upcoming JavaZone conference (september 3-4).
🗨️ Scott: 📝 The System Inside the System
I've been thinking about the Viable System Model as a framework for AI agents since I came across it, mentioned it here actually: https://futureofcoding.slack.com/archives/C5T9GPWFL/p1748381760844169
Finally put together some rubygems to explore it - vsm the framework and airb a CLI-based agent built with it and wrote up some more detailed thoughts about them on my newsletter:
📝 The System Inside the System
Announcing two new Ruby gems: vsm and airb
DevLog Together
🗨️ Konrad Hinsen: 🛸 A moldable Common Lisp object inspector based on CLOG
I finally implemented dataflow-based reactive views for my moldable inspector for Common Lisp. It's much shorter than the limited input handling I had before, and also more versatile. I am working on a demonstration/example, stay tuned...
There seem to be two widely used dataflow implementations for Common Lisp, out of which I picked the simple one (https://github.com/kchanqvq/lwcells) over the complex one (https://github.com/kennytilton/cells).
In the process, I ran into my first experience with ambiguities in the Common Lisp specification, with one of my new indirect dependencies.
Thinking Together
🗨️ Joel Neely:
- The clean way that the if/fi and do/od constructs embraced non-determinism still looks cleaner to me than the contemporary toolkits of various mainstream languages and frameworks.
Linking Together
🗨️ Ivan Reese:
Let's have a discussion that'll hopefully generate lots of links.
Microsoft have just placed GitHub under AI in their org chart.
If I don't want to stick around and find out what fresh hell comes next, where should I go instead? Gitlab? Codeberg? Sourceforge? Send me links to your favourite social network for git repos!
🗨️ Bill Mill:
a semi-related opinion that I don’t want to clutter that thread with:
self-hosting is awful* and there’s a big opportunity to develop structures for community-hosting things like git forges, chat servers, &c. Spread the load and develop social experience for sharing resources
*: too much work for too little use, too expensive, nobody wants to be on the hook all the time for maintenance so it rots
🗨️ Chris Knott:
I thought I would hate this paper (Hierarchical File Systems Are Dead) because I actually zealously love the hierarchical directory metaphor and hate the modern broth of files approach. In fact it is basically neutral on the user facing metaphor, recognises the necessity of replicating posix if anything will ever get traction, and is overall very good.
More like - how do we reify semantic search and tagging to the lowest level of the OS
📄 Hierarchical File Systems are Dead
AI
🗨️ Tom Larkworthy: 📝 Trying out "GEPA: Reflective Prompt Evolution Can Outperform Reinforcement Learning"
I completed a toy setup of GEPA with some fricken good results! I am very happy to have a methodology that removes some of the guess work with prompt design. Its a really simple algorithm that orchestrates a genetic evolution where the mutation operator is asking a LLM to improve the prompt based on the diagnostic trace your evals. So simple! I also enabled websearch during reflection so it can actually do its research when improving the prompt. This means it would adapt the web documentation to suit the prompt format automatically. Very good, been looking for something like this and it did not disappoint.
🧑🔬 Trying out "GEPA: Reflective Prompt Evolution Can Outperform Reinforcement Learning"
Present Company
🗨️ Ivan Reese:
We still have nobody offering to demo this month at the weirdly timed meetup (North American Tuesday Evening, or NATE). If nobody volunteers I'll have to get weird with the format. We need your help to make sure that doesn't happen: 👉 Call Now 👈
👨🏽💻 By 🐘 @marianoguerra@hachyderm.io 🐦 @warianoguerra
💬 Not a member yet? Check the Future of Coding Community
✉️ Not subscribed yet? Subscribe to the Newsletter / Archive / RSS
🎙️ Prefer podcasts? check the Future of Coding Podcast