Hallucinations in code are the least dangerous form of LLM mistakes

(simonwillison.net)

Show context

al2o3cr ◴[02 Mar 25 20:51 UTC] No.43234929[source]▶

    My cynical side suspects they may have been looking for
    a reason to dismiss the technology and jumped at the first
    one they found.

MY cynical side suggests the author is an LLM fanboi who prefers not to think that hallucinating easy stuff strongly implies hallucinating harder stuff, and therefore jumps at the first reason to dismiss the criticism.

replies(2): >>43235138 #>>43237917 #

williamcotton ◴[02 Mar 25 21:09 UTC] No.43235138[source]▶

>>43234929 #

What do you mean by "harder stuff"? What about an experimental DSL written in C with a recursive descent parser and a web server runtime that includes Lua, jq, a Postgres connection pool, mustache templates, request-based memory arena, database migrations and much more? 11,000+ lines of code with ~90% written by Claude in Cursor Composer.

https://github.com/williamcotton/webdsl

Frankly us "fanbois" are just a little sick and tired of being told that we must be terrible developers working on simple toys if we find any value from these tools!

replies(6): >>43235291 #>>43235372 #>>43235757 #>>43235877 #>>43239166 #>>43239764 #

1. Snuggly73 ◴[02 Mar 25 22:11 UTC] No.43235757[source]▶

>>43235138 #

..."request-based memory arena"...

there are some very questionable things going on with the memory handing in this code. just saying.

replies(1): >>43236158 #

2. williamcotton ◴[02 Mar 25 22:56 UTC] No.43236158[source]▶

>>43235757 (TP) #

Request-based memory arenas are pretty standard for web servers!

replies(1): >>43236510 #

3. Snuggly73 ◴[02 Mar 25 23:38 UTC] No.43236510[source]▶

>>43236158 #

Maybe be, after all - I dont write web servers (btw, the PQ and JQ libraries doesnt seem to use the arena allocator, which makes the whole proposition a bit dubious, but lets say that its me being picky).

What I meant was, that IMO the code is not very robust when dealing with memory allocations:

1. The "string builder" for example silently ignores allocation failures and just happily returns - https://github.com/williamcotton/webdsl/blob/92762fb724a9035...

2. In what seems most of the places, the code simply doesnt check for allocation failures, which leads to overruns (just couple of examples):

https://github.com/williamcotton/webdsl/blob/92762fb724a9035...

replies(1): >>43236644 #

4. williamcotton ◴[02 Mar 25 23:53 UTC] No.43236644{3}[source]▶

>>43236510 #

Thanks for digging in. Yup, those two libs don’t support custom allocators. I raised an issue in the jq repo to ask if they thought about adding it.

Great points about happy path allocations. If I ever touch the project again I’ll check each location.

Note to self: free code reviews of projects if you mention LLMs!

replies(1): >>43239971 #

5. namaria ◴[03 Mar 25 09:28 UTC] No.43239971{4}[source]▶

>>43236644 #

"People took a cursory look at a codebase I published and found glaring mistakes they discussed publicly as examples of how bad it is" is not the flex you think it is.

replies(1): >>43241341 #

6. williamcotton ◴[03 Mar 25 13:05 UTC] No.43241341{5}[source]▶

>>43239971 #

"Cursory", get it? I did indeed make it with Cursor! ;)

I hope you find yourself having a better day today than yesterday.

replies(1): >>43243292 #

7. namaria ◴[03 Mar 25 16:20 UTC] No.43243292{6}[source]▶

>>43241341 #

I hope you stop peddling AI slop

↑