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

Cognitive Debt: When Velocity Exceeds Comprehension

rockoder.com

Post reply on HN
Print view View original

181–190 of 234 posts

Re: Cognitive Debt: When Velocity Exceeds Comprehension

#181

Post by lurkshark »

the_arun wrote: post #75
jasode wrote: post #48

Not to disagree with anything the article talks about but to add some perspective... The complaint about "code nobody understands" because of accumulating cognitive debt also happened with hand-written code. E.g. some stories: - from https://devblogs.microsoft.com/oldnewthing/20121218-00/?p=58... : >Two of us tried to debug the program to figure out what was going on, but given that this was code written several year…

Probably, we need to start saving prompts in Version Control. Prompts could be the context for both humans & machines.

I agree with this, I like spec-driven-development tooling partially for this reason. That being said, what I’ve found is often that I don’t include enough of the “why” in my prompt artifacts. The “what” and “how” are pretty well covered but sometimes I find myself looking back at them thinking “Why did I do this?” I’ve started including it but it does sometimes feel weird because I feel like “Why would the LLM ‘care’ about this story?”

Re: Cognitive Debt: When Velocity Exceeds Comprehension

#182

Post by direwolf20 »

shimman wrote: post #170
direwolf20 wrote: post #168

Earlier quoted context omitted.

Writing is better than reading, but doing is better than writing.

How does this apply to coding when the act of writing IS doing? Or do you mean like coding "on your own" versus following a tutorial for example?

Writing code is doing

Re: Cognitive Debt: When Velocity Exceeds Comprehension

#183

Post by roywiggins »

Bengalilol wrote: post #119

Nothing new here, but the article is so well written and clear in how it presents the effects that it is a must-read. One could argue with its stance, but I took it as a given (the equation for cognitive debt touches on science). It feels entirely logical to view LLMs/coding agents as an almost final step in the short-term focus the overall system has been thriving on.

It's AI-written. Every heading is "The X Problem" or "The Y dilemma."

Re: Cognitive Debt: When Velocity Exceeds Comprehension

#185

Post by jonator »

I’ve found a good counter to this is having agents visualize and explain the architecture of the system. Then I gain just enough context to figure out what I’m trying to accomplish.

Also, as always, a highly modular codebase is very important. If I only have to reason about a single module then I don’t have to have full context on system.

It seems we’re now in a world where engineers are responsible for creating a good environment where an agent is able to gain context on the architecture and validate its work via tests (e2e, unit, smoke, etc). Then it can get into its own feedback loop and find the correct solution on its own much faster.

Re: Cognitive Debt: When Velocity Exceeds Comprehension

#186

Post by layer8 »

copperx wrote: post #56
senko wrote: post #35

Earlier quoted context omitted.

I recently did some work on a codebase I last touched 4 years ago. I didn't remember every line but I still had a very good grasp of how and why it's put together. (edit: and no, I don't have some extra good memory)

Lucky you. I always go "huh, so I wrote this?". And this was in the pre-AI era.

There is probably a bias here, because you notice the times where the code is unfamiliar more than the times when it’s still familiar. You wouldn’t go “huh” if not remembering was the normal case. If it were, you’d rather go “huh” if exceptionally you do remember.

Re: Cognitive Debt: When Velocity Exceeds Comprehension

#187

Post by AlecSchueler »

jonator wrote: post #185

I’ve found a good counter to this is having agents visualize and explain the architecture of the system. Then I gain just enough context to figure out what I’m trying to accomplish. Also, as always, a highly modular codebase is very important. If I only have to reason about a single module then I don’t have to have full context on system. It seems we’re now in a world where engineers are responsible for creating a go…

> It seems we’re now in a world where engineers are responsible for creating a good environment where an agent is able to gain context on the architecture and validate its work via tests

Part of me feels like we could have increased both velocity and comprehension a great amount twenty years ago already if we'd only had the same considerations for our fellow developers.

Re: Cognitive Debt: When Velocity Exceeds Comprehension

#188

Post by lurkshark »

layer8 wrote: post #172
the_arun wrote: post #75

Earlier quoted context omitted.

Probably, we need to start saving prompts in Version Control. Prompts could be the context for both humans & machines.

I wonder how scalable that is. After the twentieth feature has been added, how much connection will the conversation about the first feature still have with the current code? And you’ll need a larger and larger context for the LLM to grok the history; or you’ll have to have it rewrite it in shorter form, but that has the same failure modes why we can’t just have it maintain complete documentation (obviating the need…

Things like MemGPT/Letta, ToM-SWE, and Voltropy have made long context documentation pretty manageable. You could probably build some specialized tooling/prompts for development artifacts specifically too. But I’ll be the first to admit this is basically “Throw more agents at the problem”

Re: Cognitive Debt: When Velocity Exceeds Comprehension

#189

Post by layer8 »

vjvjvjvjghv wrote: post #116
senko wrote: post #35

Earlier quoted context omitted.

I recently did some work on a codebase I last touched 4 years ago. I didn't remember every line but I still had a very good grasp of how and why it's put together. (edit: and no, I don't have some extra good memory)

I definitely understand my own code better than what other people wrote, even from 10 years ago. I often see code and think "this makes sense to do it this way". Turns out I wrote it years ago.

That means your personal growth has plateaued. ;)

Re: Cognitive Debt: When Velocity Exceeds Comprehension

#190

Post by CrzyLngPwd »

Back in the day, in the 90s, I was taught that software is twice as hard to debug and maintain as it is to write, so you need to write it half as smart as you think you are.

Now that we have coding assistants and so-called AI, 'software developers' are prompting code that far exceeds their abilities.

The piper will need to be paid, one way or another.

Post reply on HN