logoalt Hacker News

jltsiren • yesterday at 7:03 PM • 2 replies • view on HN

That just follows from the fact that people can learn. Human organizations always have to balance between tacit knowledge and explicit documentation. Document too little, and onboarding new people will be more difficult, and you will lose information when people quit. Document too much, and you are just wasting time without doing anything productive, because people would have figured out that extra information in less time than it took to write it down.

The fundamental problem is that current AI tools don't learn. They make the same mistakes over and over again, unless you give them explicit instructions to avoid those mistakes. Maybe we need to accept that all work is now done by new hires, who have read all the documentation but have no experience in anything. Or maybe AI tools will get better, and all that tedious documentation effort was just a waste of time.


Replies

pixl97 • yesterday at 8:35 PM

>current AI tools don't learn

and

>and all that tedious documentation effort was just a waste of time.

Seeing where the AI is failing and then putting out good public documentation is a good way to ensure the next released model learns.

➕ show 1 reply
CharlieDigital • yesterday at 7:10 PM

    > The fundamental problem is that current AI tools don't learn
They don't learn, but they are very, very good at searching. That is why writing things down is important, but so is getting that into some sort of agent accessible index.