Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

A case for Go as the best language for AI agents

getbruin.com

Post reply on HN
Print view View original

261–270 of 310 posts

Re: A case for Go as the best language for AI agents

#261

Post by jaen »

moritz wrote: post #28

C.f., from 25d ago: “Why Elixir is the best language for AI” https://news.ycombinator.com/item?id=46900241 - for comparison of the arguments made - features a bit more actual data than “intuitions” compared to OP - interesting to think about in an agent context specifically is runtime introspection afforded by the BEAM (which, out of how it developed, has always been very important in that world) - the blog post has…

For convenience, a link to the original paper ("AutoCodeBench: Large Language Models are Automatic Code Benchmark Generators"):

https://autocodebench.github.io/#:~:text=Experimental%20Resu...

Go as a language for LLM generation actually trails a lot of other languages, at least according to this research...

Re: A case for Go as the best language for AI agents

#262

Post by g947o »

wakawaka28 wrote: post #218
jimbokun wrote: post #207

Earlier quoted context omitted.

It’s about the best possible documentation.

It isn't documentation. It is example code, in the best case. That shit belongs in other files, not in the main file. There is also a reason why literate programming never took off in general. Good luck getting anything done when 80% (conservatively) of the stuff you have to scroll through contributes nothing to the actual execution of the program and might actually be giving you false impressions of how things need…

I have yet to see a single Rust file where the test comes before source and takes 80% of the file content.

Re: A case for Go as the best language for AI agents

#263

Post by re-thc »

hu3 wrote: post #134
gf000 wrote: post #118

Earlier quoted context omitted.

Most of these reasons apply to Java as much, if not more. It's an even more popular language with even more training data and also has a better type system so more validation on LLM output, etc.

Java has decade(s) of cruft and breaking changes which LLMs were trained on. It's hard to compare. Plus Go compilation speed/test running provides quick iteration for LLMs.

> Plus Go compilation speed/test running provides quick iteration for LLMs

How is Java slow in any of these? Plenty of tricks e.g. AppCDS before the tests.

Re: A case for Go as the best language for AI agents

#264

Post by ithkuil »

stouset wrote: post #260
wakawaka28 wrote: post #216

Earlier quoted context omitted.

It is literally being changed constantly. They chose to add it to the Linux kernel but they have to build the features they need into the language as they go. It's not backward compatible like C and C++ either. Sure, there are a couple of versions you could try to pin to in Rust, but then all your libraries need to be pinned to the same version (to be fair, this last part is an assumption).

I think very nearly every assertion in this statement is demonstrably untrue.

I think the relevant problem of rust to this thread is not that it's unstable (it isn't unstable!) but it can be argued that it's a moving target

Re: A case for Go as the best language for AI agents

#266

Post by RedNifre »

grey-area wrote: post #72
thot_experiment wrote: post #53

Earlier quoted context omitted.

Why is this a meaningful distinction to you? What does "reason" mean here? Can we construct a test that cleanly splits what humans do from what LLMs do?

Sure, things like counting the ‘r’s in strawberry, for example (till they are retrained not to make that mistake).

They don't see the letters, so how could they possibly succeed at that? It's like asking a human how many infrared flowers they see.

Re: A case for Go as the best language for AI agents

#267

Post by wakawaka28 »

g947o wrote: post #262
wakawaka28 wrote: post #218

Earlier quoted context omitted.

It isn't documentation. It is example code, in the best case. That shit belongs in other files, not in the main file. There is also a reason why literate programming never took off in general. Good luck getting anything done when 80% (conservatively) of the stuff you have to scroll through contributes nothing to the actual execution of the program and might actually be giving you false impressions of how things need…

I have yet to see a single Rust file where the test comes before source and takes 80% of the file content.

Probably because all the tests are trivial, and people have the bias to not add all the testing that is needed inline with the code.

Re: A case for Go as the best language for AI agents

#268

Post by wakawaka28 »

stouset wrote: post #260
wakawaka28 wrote: post #216

Earlier quoted context omitted.

It is literally being changed constantly. They chose to add it to the Linux kernel but they have to build the features they need into the language as they go. It's not backward compatible like C and C++ either. Sure, there are a couple of versions you could try to pin to in Rust, but then all your libraries need to be pinned to the same version (to be fair, this last part is an assumption).

I think very nearly every assertion in this statement is demonstrably untrue.

It's not, especially the part about it being actively changed to adapt to Linux. I've heard several complaints about this already. I may have been a little harsh on the stability concerns but I stand by my assertions in general.

Re: A case for Go as the best language for AI agents

#269

Post by mpalmer »

CuriouslyC wrote: post #213
mpalmer wrote: post #43

Have yet to find a better choice than OCaml: - Strongly typed, including GADTs and various flavors of polymorphism, but not as inscrutable as Haskell - (Mostly) pure functions, but multiple imperative/OO escape hatches - The base language is surprisingly simple - Very fast to build/test (the bytecode target, at least) - Can target WASM/JS - All code in a file is always evaluated in order, which means it has to be def…

Seems like if you're in that boat you'd just want to go with F# though?

Nope

Re: A case for Go as the best language for AI agents

#270

Post by datsci_est_2015 »

st3fan wrote: post #231
hu3 wrote: post #134

Earlier quoted context omitted.

Java has decade(s) of cruft and breaking changes which LLMs were trained on. It's hard to compare. Plus Go compilation speed/test running provides quick iteration for LLMs.

All you need is a JAVA_CODING_GUIDELINES.md with some hints about what kind of Java code you like the agent to write.

All you need is [to encode all of the lessons you’ve learned about style and best practices over your entire career into a single markdown file].

Sorry, couldn’t help myself.

Post reply on HN