I don't understand these posts. I've never been more excited to program. Everything that I disliked about it disappeared overnight.
Who likes typing code, writing boilerplate, reading bad documentation for nights looking for that small thing, asking around in forums, reading dependency source code, writing trivial unit/ui tests.
And who likes planning, architecting, directing a team, steering and giving advice, reviewing code, designing interfaces and APIs. Coding became more mentally exciting.
LLMs truly took the worst of this trade, and left all the enjoyable things (which they will be unable to do until AGI = for a long time if ever).
> Who likes typing code, writing boilerplate, reading bad documentation for nights looking for that small thing, asking around in forums, reading dependency source code
Me? I enjoy that stuff for what it is. Reviewing code is definitely not the truly enjoyable thing. Writing code, expressing my logic in code. That is enjoyable to me.
But you can still do that, much easier, with plain human words.
/yes I know a coding language is 100% better at expressing algorithmical logic, but it is definitely harder for humans to comprehend and express cleanly at a consistent rate. machines do that way better/
> but it is definitely harder for humans to comprehend and express cleanly at a consistent rate
Part of the skill and joy of programming is to constantly work on improving my ability 'to comprehend and express cleanly' my intent in code.
No, it's actually much, much harder unless you are ignoring any of the details, which are the things that actually matter ultimately. When you were first learning did they not do the "tell me how to make a PB+J sandwich" or whatever to show you the imprecision of plain human words? You may comprehend the LLM I/O easily, but you don't know that that matches what's actually happening.
> But you can still do that, much easier, with plain human words.
Not really? In my experience, to get anything precise done, you have to fight the LLM every step of the way. And then when you come back after a few days, you realize that it has overwritten the carefully crafted code or data structures.
Unrealistic deadlines, productivity being measured in PR counts and tokens spent, and generally a 100% focus towards having a software factory produce all code.
In our latest evaluation, one of the criteria was whether you trust AI, with "trust" being defined as letting the AI write all the code, without reviewing it. And of course, not trusting AI meaning that you were behind the curve.
It's like saying who likes to ride a bicycle in the world of cars, you can do it much easier with a plain steering wheel.
Well... sure it's more efficient (although in some cities and certain routes it's not; same with coding and LLMs). But some of us just love riding the bicycle and enjoying the nature, you know. Perhaps even strengthening our muscles as we do it, as a healthy side effect.
I definitely comprehend coding language better/faster when I want to fully understand a program. I usually skip docs and go to the sources when possible. I think it's deeper than just the code though, the biggest thing I haven't seen in using natural language is good organization to express the program.
poisonborz · · focus · HN ↗
Who likes typing code, writing boilerplate, reading bad documentation for nights looking for that small thing, asking around in forums, reading dependency source code, writing trivial unit/ui tests.
And who likes planning, architecting, directing a team, steering and giving advice, reviewing code, designing interfaces and APIs. Coding became more mentally exciting.
LLMs truly took the worst of this trade, and left all the enjoyable things (which they will be unable to do until AGI = for a long time if ever).
girvo · · focus · HN ↗
Me? I enjoy that stuff for what it is. Reviewing code is definitely not the truly enjoyable thing. Writing code, expressing my logic in code. That is enjoyable to me.
poisonborz · · focus · HN ↗
/yes I know a coding language is 100% better at expressing algorithmical logic, but it is definitely harder for humans to comprehend and express cleanly at a consistent rate. machines do that way better/
Matl · · focus · HN ↗
Part of the skill and joy of programming is to constantly work on improving my ability 'to comprehend and express cleanly' my intent in code.
I'd argue LLMs are anything but consistent.
cautiouscat · · focus · HN ↗
duskdozer · · focus · HN ↗
q3k · · focus · HN ↗
Skill issue.
lolakutty · · focus · HN ↗
If you are lucky, your question have a good representation in the training material and the randomness is on your side..
But the LLM won't tell you that, so until you test your luck by taking the answer from LLM and using it, you never know..
But then, the AI company would have run off with your token payments...
Yoric · · focus · HN ↗
Not really? In my experience, to get anything precise done, you have to fight the LLM every step of the way. And then when you come back after a few days, you realize that it has overwritten the carefully crafted code or data structures.
thunky · · focus · HN ↗
1. Review what it writes as it writes it
2. Give it a smaller, more focused, scope to make changes
Or, keep vibing but then don't get mad about it.
Yoric · · focus · HN ↗
Sadly, at my current company, this doesn't seem to be acceptable practice.
tome · · focus · HN ↗
girvo · · focus · HN ↗
Yoric · · focus · HN ↗
In our latest evaluation, one of the criteria was whether you trust AI, with "trust" being defined as letting the AI write all the code, without reviewing it. And of course, not trusting AI meaning that you were behind the curve.
aldanor · · focus · HN ↗
Well... sure it's more efficient (although in some cities and certain routes it's not; same with coding and LLMs). But some of us just love riding the bicycle and enjoying the nature, you know. Perhaps even strengthening our muscles as we do it, as a healthy side effect.
dylanowen · · focus · HN ↗