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

AI;DR (AI; Didn't Read)

rickmanelius.com

Post reply on HN
Print view View original

531–540 of 729 posts

Re: AI;DR (AI; Didn't Read)

#531

Post by hbcdbff »

seer wrote: post #444
LPisGood wrote: post #56

My coworkers continue to dump hundreds of lines of AI documentation in every PR and every other line of code has between one and ten lines of AI generated comments, talking about the real unlock and how things are byte for byte identical on the load bearing path or how the acceptance ladder is misleading. Features are coming out and metrics are improving, but we’re basically in a post readability code base, with the…

Honestly, I’ve stopped caring about code readability for a few months now. I want the code readable _to the agent_ not so much to me. I don’t trust it with code anyway - every feature needs comprehensive test, and then a live deploy on a real working test system before it is approved - I mostly measure success with - after deployment is it doing what it’s supposed to be doing. It’s like “helping another team managing…

> I’ve stopped caring about code readability for a few months now. […] I don’t trust it with code anyway

If you don’t trust it with code, surely you need the code to be readable so you can understand what it is writing?

Re: AI;DR (AI; Didn't Read)

#532

Post by eru »

LPisGood wrote: post #480
eru wrote: post #457

Earlier quoted context omitted.

Try property based testing perhaps.

I’ve heard of this, but I’m not really sure at all how to even get started. Are there any good guides out there?

Many people like https://fsharpforfunandprofit.com/series/property-based-test...

In this day and age, https://hypothesis.works/articles/claude-code-plugin/ might be useful.

Re: AI;DR (AI; Didn't Read)

#533

Post by gritzko »

preg_match wrote: post #382
koyote wrote: post #263

Earlier quoted context omitted.

It's not just claude, all AI is unable to produce something concise. On the surface everything looks 'good' whether code or prose, but then if you dig a bit, try and understand the whole text you quickly realise that 80% of it is unecessary and the whole thing could have been re-worded/re-coded into something a fraction of its size and complexity. I asked Sol to reduce the length of some documentation we had by makin…

I have to ask Claude to compact the comments every time, and I give specific criteria for it. Never ever reiterate what’s in the code, never mention decisions not made, never mention the conversation, etc etc. Even then it is conservative. For the love of God, compact the comments. Comments become a huge maintenance burden, especially in the age of AI. They just grow and grow, and then mislead the AI later on.

I set a line budget for comments (also wiki page parts, chat responses, etc). That only helps when I ask it to do a second pass to reword everything to the budget and add links. I think they tuned it this way to stash reasoning dumps in the code. Unlike human developers, it has no context in its head, other than general GitHub knowledge.

That is a dramatic shortcoming, but I fix it with permalinks to other files. https://replicated.live/blog/link

Re: AI;DR (AI; Didn't Read)

#534

Post by Gepsens »

We could put comments in separate files with an annotation format. For example: ion.cpp.cmt And autoremove all comments from source files. Then use plug-ins to toggle comment files within the text editor.

Re: AI;DR (AI; Didn't Read)

#537

Post by Bombthecat »

LPisGood wrote: post #56

My coworkers continue to dump hundreds of lines of AI documentation in every PR and every other line of code has between one and ten lines of AI generated comments, talking about the real unlock and how things are byte for byte identical on the load bearing path or how the acceptance ladder is misleading. Features are coming out and metrics are improving, but we’re basically in a post readability code base, with the…

You use AI to summarise it! That's the way to go lol

Re: AI;DR (AI; Didn't Read)

#538

Post by johnnyanmac »

x-complexity wrote: post #380
ryankrage77 wrote: post #311

Earlier quoted context omitted.

The stream-of-consciousness in the prompt is much better than the LLMs response. Yes, it's messy and unpolished, maybe even a little hard to parse, but it's so much richer for it. The way it is written lets me follow your thought process, shows that you were getting the words out quickly without much or any editing, and how you were exploring the idea and making it more concrete by giving it written form. The LLM res…

> Yes, it's messy and unpolished, maybe even a little hard to parse, but it's so much richer for it. ...This part immediately made me link it to the hipster stereotype: "oh, it's sooooo much better on vinyl because you can hear the cracks & pops" Neither pieces in the chat log work for me: The prompt was a wall of ramblespeak. The output was a presentation that should've been an email.

"More interesting" =/= "good". Yes, at the end of the day it's a very derisive rant against people the writer feels is too hard on AI and a defense on how they personally wield it. With a slightly ill intent to create a messy, unstructured writing in the prompt to support his point of "see, people like pretty words!".

Even the attempt of using Claude to be a fancy formatter failed when you look closely at the language employed in the final prompt. It's trying to convey the same ideas with better foratting, but still softens up the derision the author clearly has here. Which is the interesting part in my eyes.

Re: AI;DR (AI; Didn't Read)

#539

Post by filterfish »

roarcher wrote: post #519
doix wrote: post #510

Earlier quoted context omitted.

I hate that I have to change the way I write to avoid AI-isms. I loved using "load bearing" to describe weird code that you think you can delete but is actually holding everything together. Now people think I'm just parroting what Claude said. It sucks. I want my catch phrases back, I guess this is how em dash users felt

As a former em-dash enjoyer, I indeed feel your pain.

I'm with you an that one. RIP em-dash.

Re: AI;DR (AI; Didn't Read)

#540

Post by gavinsyancey »

teo_zero wrote: post #530

Whenever I'm tempted to base my judgment on how something's been created rather than what the end result is, I think about the scene where Sheldon throws away the french toast Penny made for him, despite smelling good, just because "Monday is oatmeal day."

Before LLMs came around, it took quite a bit more effort to write something than to read it. So the fact that someone had taken the time to create something was a signal that they thought it was worth that effort, and so it might be worth my (lesser) effort to read it. And even with that filter there was a lot of human-created crap out there.

I value my time. With a LLM, it takes very little effort to create a vast mountain of crap that will take a lot of time and effort to read through. True, it is also possible to use LLMs as a tool while creating things that are not worthless. But the fact that someone has not taken the time to make their thing not sound obviously AI-generated is a strong signal that it will not be worth my time.

Post reply on HN