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

What I'm Hearing About Cognitive Debt (So Far)

margaretstorey.com

Post reply on HN
Print view View original

121–130 of 150 posts

Re: What I'm Hearing About Cognitive Debt (So Far)

#121

Post by mishellaneous »

nsoonhui wrote: post #105

To me, the cognitive debt incurred by Agentic AI described here is not so different from the cognitive debt incurred by code written by someone else. Even when you are the reviewer of your colleague’s code, you can’t just grok everything as if the code were written by you. What more to say if you are not even the reviewer. And that’s okay! Much like it’s okay to let other people write the code. What is important is t…

yes, from the point of view of an individual programmer, there is no difference between code created by AI or by some other programmer.

but from the point of view of the company as a whole there is a difference: code created by some programmer is understood by that programmer, while code created by AI is understood by nobody

Re: What I'm Hearing About Cognitive Debt (So Far)

#122

Post by svnt »

mishellaneous wrote: post #121
nsoonhui wrote: post #105

To me, the cognitive debt incurred by Agentic AI described here is not so different from the cognitive debt incurred by code written by someone else. Even when you are the reviewer of your colleague’s code, you can’t just grok everything as if the code were written by you. What more to say if you are not even the reviewer. And that’s okay! Much like it’s okay to let other people write the code. What is important is t…

yes, from the point of view of an individual programmer, there is no difference between code created by AI or by some other programmer. but from the point of view of the company as a whole there is a difference: code created by some programmer is understood by that programmer, while code created by AI is understood by nobody

> code created by some programmer is understood by that programmer

was at some point understood by that programmer, at least for a few months, before they moved on to something else and forgot important context and detail, or left the company

Re: What I'm Hearing About Cognitive Debt (So Far)

#123

Post by jdw64 »

i_think_so wrote: post #119
jdw64 wrote: post #67

Earlier quoted context omitted.

As someone from outside the Anglophone cultural sphere, when I first learned to write in English, the kind of writing that AI now often produces was taught to me as “formal" writing. But these days, when I write in that formal style, people sometimes say it sounds like AI. That has been a difficult and frustrating point for me. I still find the subtle difference hard to understand.

I was raised and educated well inside the Anglosphere (USA) and was also taught to write formally in that way. Do the people who say you sound like AI give you any specifics? Also, if you don't mind, what was your English education like? I understand that quite a few Americans work in South Korea as teachers but I have no details about how that manifests.

That used to happen to me more often. When I first came to HN, and even now if I am not careful, my comments can get flagged. Also, when I translate from Korean using DeepL and paste the result, people often say it sounds flagged, awkward, or unnatural. I studied English more seriously in graduate school, although I dropped out. In Korea, there are quite a few Americans who teach English. Public schools often have native English-speaking instructors, but in my case I learned English more seriously at graduate school, and universities also make students study English almost semi-compulsorily.

In Seoul, there are probably many teachers who mainly teach middle and high school students, but a lot of that is through private education rather than the public school system.

Re: What I'm Hearing About Cognitive Debt (So Far)

#124

Post by eithed »

hogehoge51 wrote: post #14

Unfortunately I think Cognitive Debt is the cry of the software craftsperson who thought they were an Engineer. Upon working with the agent subcontractor, the agent factory, the agent part vendor, they approached it as a craft; they found themselves wanting to walk through the offices of the subcontractor reviewing screens, inspect pieces at the factory, and get the internal design for the parts they ordered. It's na…

I think this is nothing new and comes down to just working with people, even without LLMs - many times I found myself puzzled by "what does this code do that was merged in by this person without my approval?"; without specs and docs it's inevitable, unless you're maniacally inspecting every PR and holding this knowledge in your head.

Re: What I'm Hearing About Cognitive Debt (So Far)

#125

Post by Yokohiii »

raincole wrote: post #112
Yokohiii wrote: post #76

Earlier quoted context omitted.

If the default mode is that LLMs generate crap code that you have to fix with even more LLMs, then something is fundamentally wrong. In terms of over engineering, I wouldn't be surprised if the human tendency for skeuomorphism (combined with an loss of technical skill) will create even weirder code.

The default mode of the whole startup world is humans generate crap code that you have to fix with even more humans. And it has been like that for at least two decades.

Humans are accountable.

Re: What I'm Hearing About Cognitive Debt (So Far)

#126

Post by erikerikson »

i_think_so wrote: post #118
erikerikson wrote: post #19

Earlier quoted context omitted.

Or, stay with me here, the LLMs were trained on how we, statistically, write.

When you say "we" you're talking about Twitter, right?

I used that once, during a conference about 6 years ago and never again since. My use of "we" references humanity.

Re: What I'm Hearing About Cognitive Debt (So Far)

#127

Post by margaretstorey »

Nice to read the discussion here.

A few mentioned cognitive debt is not new, I agree. Large systems have always been beyond the reach of individual understanding (Naur and others have written about this for decades).

What I'm curious to investigate is whether AI is amplifying and changing the nature of cognitive debt as code and system features are generated faster than teams can build shared understanding.

Some of the discussion around specs, contracts and accountability touch on how we can manage intent, especially as systems evolve.

I've been working on a broader frame around technical, cognitive, and intent debt if folks are interested: https://arxiv.org/abs/2603.22106

Re: What I'm Hearing About Cognitive Debt (So Far)

#128

Post by Gravityloss »

Aperocky wrote: post #63
efitz wrote: post #52

There are many projects that should not be built agentically. For things that are appropriate to build with agents, I have come to hold the strong opinion that you need to go all-in. If you built it with an agent, then you fix it with an agent, you debug it with an agent, and you change it with an agent. In that case you should not consider yourself the steward of the source code and worry about “cognitive debt”- it’…

This is correct, but it misses out an important dimension. You can inject philosophy into the agent and ensure that it sticks to it. The LLM will, with sufficient drilling, begrudgingly implement it, most important of which is SIMPLE>COMPLEX on all levels and you have to either manually or agentically continuously monitor this. Alternatively, LLM will use its tiny context window to build a true spaghetti that even it…

I wonder if there is some easy way to have a "spaghetti index" of a codebase. One could then have different criteria for different projects / parts of projects. If it's a business critical thing that also expects to see a lot of future development, one could then communicate to management that the spaghetti index is too high and that must be first lowered before further development can happen. And on the other hand, if it's a throwaway tool that's used for a limited time, no matter.

Re: What I'm Hearing About Cognitive Debt (So Far)

#129

Post by datsci_est_2015 »

Aperocky wrote: post #46

Struck a cord, and I think I managed this prior to LLM in a certain degree. My primary editor is vim, and for a significant amount of time I was using it almost in puritan fashion, this was before LLM was mainstream. However, I could not use vim to edit java, even with language server - I tried, but each time I went back to intellij - the rest of the code base in python, ruby and typescript was typically fine. The re…

> The reason was two fold, because everyone was using all of the features that intellij had to offer, the code was structured similar to intellij and obviously the java design patterns that was popular at the time. Everything went through factories and managers and interfaces and tracking them through a pure editor was almost impossible. The IDE handled it for you.

One of my journeyman theories of programming is that your IDE has a way bigger impact on how you write your code than would be implied based on how much people talk about it online.

Trivial example: worked at a startup with a coworker who double spaced all of his code. He should have just set his editor / IDE up to have a larger margin between lines. The lines were too close for his comfort in his IDE.

I often find juniors will write a bunch of “utility functions” if their IDE lets them quickly hover or ctrl-click for function definitions. Then they just create a “utils.py” and store all of those “utility functions” there. Now their code is a spaghetti of indirection without any conceptual groupings.

Re: What I'm Hearing About Cognitive Debt (So Far)

#130

Post by marcosdumay »

scorpioxy wrote: post #61
marcosdumay wrote: post #42

Earlier quoted context omitted.

In fact, agile has been extremely successful. It's the people that claim to "do agile" that invariably don't do it. But software development used to fail most of the time, and it doesn't do that anymore.

What makes you say that it has been extremely successful? And when you say doesn't fail anymore, do you mean it doesn't go over budget and/or changes scope?

It goes severely overbudget less often, it rarely goes unused, and it causes organizational changes more often than not.

Except on the places that "do agile", those got worse. There is an organization that tracks that, I don't remember the name.

Post reply on HN