Live data from Hacker News

Show HN: I made a "programming language" looking for feedback

github.com

31–40 of 42 posts

Re: Show HN: I made a "programming language" looking for feedback

#31

Why would I want this?

README.md looks like

$$$PROGRAM vector X=[12. 17] -> rules

sequential output so AI doesn't hallucinate.

I have no problem harnessing LLMs for building my application. I don't need another unreadable mess. Why do I need this?

You fail to communicate the problem this solves.

It's a more involved way to prompt?

Re: Show HN: I made a "programming language" looking for feedback

#32

But… why not just write pseudo code, or any language you actually know, and just ask the AI to port it to the language you want? That’s a serious question by the way, is there some use case here I’m not seeing where learning this new syntax and running this actually helps instead of being extra steps nobody needs?

Indeed, it seems to occupy a middle ground between fast-and-easy AI prompting, and slow-but-robust traditional programming. But it's still relying on AI outputs unconstrained (as far as I can tell) by more formal methods and semantic checks. But it's also hard for me to grasp the exact value add from the README, or why I should buy their story, so I'm not sure.

Install my executable bro, trust me just one more tool and you will be the 10x engineer!!!

Re: Show HN: I made a "programming language" looking for feedback

#33
post #13

Earlier quoted context omitted.

What does temperature have to do with anything?

Or however deterministic inference is supposed to happen. I don’t know LLMs.

https://152334h.github.io/blog/non-determinism-in-gpt-4/

Re: Show HN: I made a "programming language" looking for feedback

#34
post #16
post #15

Earlier quoted context omitted.

Well, it's still not deterministic even at temp 0. The tech described in my comment's parent is speculative, and technically it's not even inference, once it's perfectly reproducible. At that point it's retrieving results from a database. EDIT: how would OP address my main point, which is that det. inference is functionally equivalent to any arbitrary keyed data storage/retrieval system?

Deterministic inference isn't speculative, it's achievable if you want it. It's just not the default.

It seems the TL:DR is race conditions, rounding and other RTE specific factors.

Re: Show HN: I made a "programming language" looking for feedback

#37

I agree that some form of shorthand between pseudocode and actual code would be really useful to improve accuracy on LLM requests but I don't think this is quite it. Ideally it would be as simple as possible, but not rely on language-specific paradigms. Sort of a pidgin that everyone would understand, that used white space and indentation to indicate things like loops and such. Something a normal person could look at…

I wrote IntentCode, but it seems to have gone under the radar: https://github.com/jfilby/intentcode
Post reply on HN