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

AI makes the easy part easier and the hard part harder

blundergoat.com

Post reply on HN
Print view View original

171–180 of 382 posts

Re: AI makes the easy part easier and the hard part harder

#171

Post by fooker »

ThunderSizzle wrote: post #120
zozbot234 wrote: post #80

Earlier quoted context omitted.

That's nice if it works, but why not look at the plan yourself before you let the AI have its go at it? Especially for more complex work where fiddly details can be highly relevant. AI is no good at dealing with fiddly.

That's what you can do. Tell the AI to make a plan in an MD file, review and edit it, and then tell another AI to execute the plan. If the plan is too long, split it into steps.

This has been a well integrated feature in cursor for six months.

As a rule of thumb, almost every solution you come up with after thirty seconds of thought for a online discussion, has been considered by people doing the same thing for a living.

Re: AI makes the easy part easier and the hard part harder

#172

Post by fsmv »

phpnode wrote: post #133
AuthAuth wrote: post #54

Earlier quoted context omitted.

Its license washing. The code is great because its already a problem solved by someone else. The AI can spit out the solution with no license and no attribution and somehow its legal. I hope American tech legislation holds that same energy once others start taking American IP and spitting it back out with no license or attribution.

The other day I had an agent write a parser for a niche query language which I will not name. There are a few open source implementations of this language on github, but none of them are in my target language and none of them are PEGs. The agent wrote a near perfect implementation of this query language in a PEG. I know that it looked at the implementations that were on github, because I told it to, yet the result is…

It would be nice to give them some kind of attribution in the readme or something since you know which projects you referenced

Re: AI makes the easy part easier and the hard part harder

#173

Post by raw_anon_1111 »

apsurd wrote: post #161
raw_anon_1111 wrote: post #139

Earlier quoted context omitted.

I agree completely. I just did my first “AI native coding project”. Both because for now I haven’t run into any quotas using Codex CLI with my $20/month ChatGPT subscription and the company just gave everyone an $800/month Claude allowance. Before I even started the implementation I: 1. Put the initial sales contract with the business requirements. 2. Notes I got from talking to sales 3. The transcript of the initial…

That sounds really powerful, but also like burden shifts to the people that will maintain all this stuff after you're done having your fun. Tbh, I'm not exactly knocking it, it makes sense that leads are responsible for the architecture. I just worry that those leads having 100x influence is not default a good thing.

My thought is that the markdown is the code and that Claude code/Codex is the “compiler”.

The design was done by me. The modularity, etc.

I tested for scalability, I checked the IAM permissions for security and I designed the locking mechanism and concurrency controls (which had a bug in it that was found by ChatGPT in thinking mode),

Re: AI makes the easy part easier and the hard part harder

#174

Post by lock1 »

marcus_holmes wrote: post #140
johnbender wrote: post #134

Earlier quoted context omitted.

Compilers don’t do this error free of course BUT if we want them too we can say what it means for a compiler to be correct very directly _one time_ and have it be done for all programs (see the definition for simulation in the CompCert compiler). This is a major and meaningful difference from AI which would need such a specification for each individual application you ask it to build because there is no general speci…

> there is no general specification for correct translation from English to Code. that's an interesting point. Could there be? COBOL was originally an attempt to do this, but it ended up being more Code than English. I think this is the area we need to get better at if we're to trust LLMs like we trust compilers. I'm aware that there's a meme around "we have a method of completely specifying what a computer system sh…

No, there isn't.

I guess you could pick a subset of a particular natural language such that it removes ambiguity. At that point, you're basically reinventing something like COBOL or Python.

Ambiguity in natural languages is a feature, not a bug. While it's better not to be an unintentional pun or joke instruction that might get interpreted as "launch the missile" by computer.

However, each project error tolerance is different. Arguably, for an average task within the umbrella of "software engineer", even current LLMs seem good enough for most purposes. It's a kind of similar transition to automatic memory managed language, trading control for "DX".

Re: AI makes the easy part easier and the hard part harder

#175

Post by whaleidk »

marcus_holmes wrote: post #140
johnbender wrote: post #134

Earlier quoted context omitted.

Compilers don’t do this error free of course BUT if we want them too we can say what it means for a compiler to be correct very directly _one time_ and have it be done for all programs (see the definition for simulation in the CompCert compiler). This is a major and meaningful difference from AI which would need such a specification for each individual application you ask it to build because there is no general speci…

> there is no general specification for correct translation from English to Code. that's an interesting point. Could there be? COBOL was originally an attempt to do this, but it ended up being more Code than English. I think this is the area we need to get better at if we're to trust LLMs like we trust compilers. I'm aware that there's a meme around "we have a method of completely specifying what a computer system sh…

No, there can’t be. Code keywords are tied to concrete mathematical concepts. Human languages are not. and even if you tried, the more languages you add to the LLM’s pool, misinterpretation chances increase exponentially. You can’t just choose English to be the programming language either, because then you would be asking every non-English speaking developer in the world to first learn the entirety of the English language which is way harder than just learning a programming language. Why are programmers so scared of code and math??

Re: AI makes the easy part easier and the hard part harder

#176

Post by blibble »

bethekidyouwant wrote: post #50
zozbot234 wrote: post #34

Earlier quoted context omitted.

The most credible argument against AI is really the expense involved in querying frontier models. If you want to strengthen the case for AI-assisted coding, try to come up with ways of doing that effectively with a cheap "mini"-class model, or even something that runs locally. "You can spend $20k in tokens and have AI write a full C compiler in a week!" is not a very sensible argument for anything.

How much would it cost to pay developer to do this??

zero

because they tell you to stop being so stupid and run apt install gcc

Re: AI makes the easy part easier and the hard part harder

#177

Post by _zagj »

AuthAuth wrote: post #54
zjp wrote: post #24

Earlier quoted context omitted.

I call these "embarrassingly solved problems". There are plenty of examples of emulators on GitHub, therefore emulators exist in the latent spaces of LLMs. You can have them spit one out whenever you want. It's embarrassingly solved. There are no examples of what you tried to do.

Its license washing. The code is great because its already a problem solved by someone else. The AI can spit out the solution with no license and no attribution and somehow its legal. I hope American tech legislation holds that same energy once others start taking American IP and spitting it back out with no license or attribution.

> The AI can spit out the solution with no license and no attribution and somehow its legal

Note that even MIT requires attribution.

Re: AI makes the easy part easier and the hard part harder

#178

Post by whaleidk »

djx22 wrote: post #148

Don't let AI write code for you unless it's something trivial. Instead use it to plan things, high level stuff, discuss architecture, ask it to explain concepts. Use it as a research tool. It's great at that. It's bad at writing code when it needs to be performant or needs to span over multiple files. Especially when it spans over multiple files because that's where it starts hallucinating and introducing abstraction…

IME it’s actually really terrible at discussing architecture. It’s incredibly unimaginative and will just confirmation-bias whichever way you are leaning slightly more towards

Re: AI makes the easy part easier and the hard part harder

#179

Post by _zagj »

bsenftner wrote: post #65

People need to consider / realize that the vast majority of source code training data is Github, Gitlab, and essentially the huge sea of started, maybe completed, student and open source project. That large body of source code is for the most part unused, untested, and unsuccessful software of unknown quality. That source code is AI's majority training data, and an AI model in training has no idea what is quality sof…

> huge sea of started, maybe completed, student and open source project.

Which is easy to filter out based on downloads, version numbering, issue tracker entries, and wikipedia or other external references if the project is older and archived, but historically noteworthy (like the source code for Netscape Communicator or DOOM).

Re: AI makes the easy part easier and the hard part harder

#180

Post by ares623 »

Qworg wrote: post #37
isodev wrote: post #33

Earlier quoted context omitted.

And what if the foundation was made by the AI itself? What’s the excuse then?

Your responsibility as a developer in this new world is design and validation. A poor foundation is a design problem. Throw it away and start again.

But how will new developers learn to design and validate in the future?
Post reply on HN