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

Reports of code's death are greatly exaggerated

stevekrouse.com

Post reply on HN
Print view View original

41–50 of 486 posts

Re: Reports of code's death are greatly exaggerated

#41

Post by allthetime »

kstrauser wrote: post #19
pacman128 wrote: post #16

In a chat bot coding world, how do we ever progress to new technologies? The AI has been trained on numerous people's previous work. If there is no prior art, for say a new language or framework, the AI models will struggle. How will the vast amounts of new training data they require ever be generated if there is not a critical mass of developers?

That’s factually untrue. I’m using models to work on frameworks with nearly zero preexisting examples to train on, doing things no one’s ever done with them, and I know this because I ecosystem around these young frameworks. Models can RTFM (and code) and do novel things, demonstrably so.

Yeah. I work with bleeding edge zig. If you just ask Claude to write you a working tcp server with the new Io api, it doesn’t have any idea what it’s doing and the code doesn’t compile. But if you give it some minimal code examples, point it to the recent blog posts about it, and paste in relevant points from std it does incredibly well and produce code that it has not been trained on.

Re: Reports of code's death are greatly exaggerated

#42

Post by wiseowise »

bryanrasmussen wrote: post #24
cactusplant7374 wrote: post #17

Earlier quoted context omitted.

> But of course writing code directly will always maintain the benefit of specificity. If you want to write instructions to a computer that are completely unambiguous, code will always be more useful than English. Unless the defect rate for humans is greater than LLMs at some point. A lot of claims are being made about hallucinations that seem to ignore that all software is extremely buggy. I can't use my phone witho…

most human bugs are caused by failures in reasoning though, not by just making something up to leap to the conclusion considered most probable, so not sure if the comparison makes sense.

> most human bugs are caused by failures in reasoning though

Citation needed.

Re: Reports of code's death are greatly exaggerated

#43

Post by jedberg »

andrei_says_ wrote: post #36
jedberg wrote: post #25

Earlier quoted context omitted.

People are doing this now. It's basically what skills.sh and its ilk are for -- to teach AIs how to do new things. For example, my company makes a new framework, and we have a skill we can point an agent at. Using that skill, it can one-shot fairly complicated code using our framework. The skill itself is pretty much just the documentation and some code examples.

The question is, who made the new framework? Was it vibe coded by someone who does not understand its code?

No, it was created by our team of engineers over the last three years based on years of previous PhD research.

Re: Reports of code's death are greatly exaggerated

#44

Post by sd9 »

justonceokay wrote: post #22
pacman128 wrote: post #16

In a chat bot coding world, how do we ever progress to new technologies? The AI has been trained on numerous people's previous work. If there is no prior art, for say a new language or framework, the AI models will struggle. How will the vast amounts of new training data they require ever be generated if there is not a critical mass of developers?

Most art forms do not have a wildly changing landscape of materials and mediums. In software we are seeing things slow down in terms of tooling changes because the value provided by computers is becoming more clear and less reliant on specific technologies. I figure that all this AI coding might free us from NIH syndrome and reinventing relational databases for the 10th time, etc.

LLMs are very much NIH machines

Re: Reports of code's death are greatly exaggerated

#45

Post by jedberg »

NewJazz wrote: post #30
jedberg wrote: post #25

Earlier quoted context omitted.

People are doing this now. It's basically what skills.sh and its ilk are for -- to teach AIs how to do new things. For example, my company makes a new framework, and we have a skill we can point an agent at. Using that skill, it can one-shot fairly complicated code using our framework. The skill itself is pretty much just the documentation and some code examples.

A framework is different than a paradigm shift or new language.

Yes and no. How does a human learn a new language? They use their previous experience and the documentation to learn it. Oftentimes they way someone learns a new language is they take something in an old language and rewrite it.

LLMs are really good at doing that. Arguably better than humans at RTFM and then applying what's there.

Re: Reports of code's death are greatly exaggerated

#46

Post by allthetime »

suzzer99 wrote: post #39
drzaiusx11 wrote: post #31

Earlier quoted context omitted.

This is already happening, many days I am that grumpy "code janitor" yelling at the damn kids to improve their slop after shit blows up in prod. I can tell you It's not "fun", but hopefully we'll converge on a scalable review system eventually that doesn't rely on a few "olds" to clean up. GenAI systems produce a lot of "mostly ok" code that has subtle issues you on catch with some experience. Maybe I should just ret…

Yeah I imagine it has to be utterly thankless being the code janitor right now when all the hype around AI is peaking. You're basically just the grumpy troll slowing things down. And God forbid you introduce a regression bug trying to clean up some AI slop code. Maybe in the future us olds will get more credit when apps fall over and the higher ups realize they actually need a high-powered cleaner/fixer, like the Wol…

I’ve got a “I haven’t written a line of code in one year” buddy whose startup is gaining traction and contracts. He’s rewritten the whole stack twice already after hitting performance issues and is now hiring cheap juniors to clean up the things he generates. It is all relatively well defined CRUD that he’s just slapped a bunch of JS libs on top of that works well enough to sell, but I’m curious to see the long term effects of these decisions.

Meanwhile I’m moving at about half the speed with a more hands on approach (still using the bots obviously) but my code quality and output are miles ahead of where I was last year without sacrificing maintain ability and performance for dev speed

Re: Reports of code's death are greatly exaggerated

#47

Post by derrak »

pacman128 wrote: post #16

In a chat bot coding world, how do we ever progress to new technologies? The AI has been trained on numerous people's previous work. If there is no prior art, for say a new language or framework, the AI models will struggle. How will the vast amounts of new training data they require ever be generated if there is not a critical mass of developers?

Maybe you’re right about modern LLMs. But you seem to be making an unstated assumption: “there is something special about humans that allow them to create new things and computers don’t have this thing.”

Maybe you can’t teach current LLM backed systems new tricks. But do we have reason to believe that no AI system can synthesize novel technologies. What reason do you have to believe humans are special in this regard?

Re: Reports of code's death are greatly exaggerated

#48

Post by cratermoon »

Yet again we can pull out Edsger W.Dijkstra's 1978 article, "On the foolishness of "natural language programming""

"In order to make machines significantly easier to use, it has been proposed (to try) to design machines that we could instruct in our native tongues. this would, admittedly, make the machines much more complicated, but, it was argued, by letting the machine carry a larger share of the burden, life would become easier for us. It sounds sensible provided you blame the obligation to use a formal symbolism as the source of your difficulties. But is the argument valid? I doubt."

Re: Reports of code's death are greatly exaggerated

#49

Post by cratermoon »

suzzer99 wrote: post #9
soumyaskartha wrote: post #7

Every few years something is going to kill code and here we are. The job changes, it does not disappear.

For future greenfield projects, I can see a world where the only jobs are spec-writer and test-writer, with maybe one grumpy expert coder (aka code janitor) who occasionally has to go into the code to figure out super gnarly issues.

A good spec-writer, as the article notes, is writing code.

Re: Reports of code's death are greatly exaggerated

#50

Post by derrak »

rglover wrote: post #33

It's only dead to those who are ignorant to what it takes to build and run real systems that don't tip over all the time (or leak data, embroil you in extortion, etc). That will piss some people off but it's worth considering if you don't want to perma-railroad yourself long-term. Many seem to be so blinded by the glitz, glamour, and dollar signs that they don't realize they're actively destroying their future prospe…

> non-deterministic printer.

I interpret non-deterministic here as “an LLM will not produce the same output on the same input.” This is a) not true and b) not actually a problem.

a) LLMs are functions and appearances otherwise are due to how we use them

b) lots of traditional technologies which have none of the problems of LLMs are non-deterministic. E.g., symbolic non-deterministic algorithms.

Non-determinism isn’t the problem with LLMs. The problem is that there is no formal relationship between the input and output.

Post reply on HN