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

Vibe coding kills open source

arxiv.org

Post reply on HN
Print view View original

131–140 of 304 posts

Re: Vibe coding kills open source

#131

Post by wasmitnetzen »

WarmWash wrote: post #29

Small bespoke personalized on the spot apps are the future with LLMs. The future will absolutely not be "How things are today + LLMs" The paradigm now for software is "build a tool shed/garage/barn/warehouse full of as much capability for as many uses possible" but when LLMs can build you a custom(!) hammer or saw in a few minutes, why go to the shed?

Because I will probably ask the AI for a rock instead of a bespoke hammer. If I even know what a nail is.

I very much like to use the years of debugging and innovation others spent on that very same problem that I'm having.

Re: Vibe coding kills open source

#132

Post by softwaredoug »

tomaytotomato wrote: post #6

I have been trying to use Claude code to help improve my opensource Java NLP location library. However trying to get it to do anything other than optimise code or fix small issues it struggles. It struggles with high level abstract issues. For example I currently have an issue with ambiguity collisions e.g. Input: "California" Output: "California, Missouri" California is a state but also city in Missouri - https://gi…

Sounds a lot like model training and I’ve treated this sort of programming with AI exactly like that importantly making sure I have a test/train split

Make sure there’s a holdout the agent can’t see that it’s measured against. (And make sure it doesn’t cheat)

https://softwaredoug.com/blog/2026/01/17/ai-coding-needs-tes...

Re: Vibe coding kills open source

#133

Post by tomaytotomato »

epolanski wrote: post #118
tomaytotomato wrote: post #6

I have been trying to use Claude code to help improve my opensource Java NLP location library. However trying to get it to do anything other than optimise code or fix small issues it struggles. It struggles with high level abstract issues. For example I currently have an issue with ambiguity collisions e.g. Input: "California" Output: "California, Missouri" California is a state but also city in Missouri - https://gi…

One major part of successful LLM-assisted coding is to not focus on code vomiting but scaffolding. Document, document, document: your architecture, best practices, preferences (both about code and how you want to work with the LLM and how do you expect it to behave it). It is time consuming, but it's the only way you can get it to assist you semi-successfully. Also try to understand that LLM's biggest power for a dev…

I agree to an extent

My project is quite well documented and I created a Prompt a while back along with some mermaid diagrams

https://github.com/tomaytotomato/location4j/tree/master/docs

I can't remember the exact prompt I gave to the LLM but I gave it a Github issue ticket and description.

After several iterations it fixed the issue, but my unit tests failed in other areas. I decided to abort it because I think my opinionated code was clashing with the LLM's solution.

The LLM's solution would probably be more technically correct, but because I don't do l33tcode or memorise how to implement Trie or BST my code does it my way. Maybe I just need to force the LLM to do it my way and ignore the other solutions?

Re: Vibe coding kills open source

#134

Post by krona »

faxmeyourcode wrote: post #58
tomaytotomato wrote: post #6

I have been trying to use Claude code to help improve my opensource Java NLP location library. However trying to get it to do anything other than optimise code or fix small issues it struggles. It struggles with high level abstract issues. For example I currently have an issue with ambiguity collisions e.g. Input: "California" Output: "California, Missouri" California is a state but also city in Missouri - https://gi…

> LLMs dont care about the story, they just care about the current state of the code You have to tell it about the backstory. It does not know unless you write about it somewhere and give it as input to the model.

The commit history of that repo is pretty detailed at first glance.

Re: Vibe coding kills open source

#135

Post by groundzeros2015 »

WarmWash wrote: post #29

Small bespoke personalized on the spot apps are the future with LLMs. The future will absolutely not be "How things are today + LLMs" The paradigm now for software is "build a tool shed/garage/barn/warehouse full of as much capability for as many uses possible" but when LLMs can build you a custom(!) hammer or saw in a few minutes, why go to the shed?

Because it can’t really do that for any tools that matter.

Re: Vibe coding kills open source

#136

Post by groundzeros2015 »

pier25 wrote: post #75
WarmWash wrote: post #29

Small bespoke personalized on the spot apps are the future with LLMs. The future will absolutely not be "How things are today + LLMs" The paradigm now for software is "build a tool shed/garage/barn/warehouse full of as much capability for as many uses possible" but when LLMs can build you a custom(!) hammer or saw in a few minutes, why go to the shed?

I don't think apps where people spend a lot of time are equivalent to small tools. You can vibe code a calculator but you probably spend most of your time on much more complex software.

A calculator that uses doubles for everything I guess.

Re: Vibe coding kills open source

#137

Post by pixl97 »

tonyedgecombe wrote: post #106
contravariant wrote: post #54

I'm never quite sure what to think of papers that have a conclusion and then build a mathematical model to support it.

Science starts with hypothesis and predictions.

Yes, but it's easy, and incorrect, to start with an answer and build your work backwards rather than proving your hypothesis with evidence.

Re: Vibe coding kills open source

#138

Post by em-bee »

9dev wrote: post #83
ipaddr wrote: post #49

For me spending time on my open source projects doesn't make sense anymore. People (the community and employers) previously were impressed because of the amount of work required. Now that respect is gone as people can't automatically tell on the surface if this is a low effort vibe code or something else. Community engagement has dropped. Stars aren't being given out as freely. People aren't actively reading your cod…

Eh, I don't believe that. Smartphones have amazing cameras, and we still have photographers. There are CNC saws and mills that will ship you your perfectly realised CAD prints, yet there are still carpenters and a vibrant community of people making their own furniture. These examples go on and on. Without any kind of offence implied: As maintainer of a few open source projects, I'm happy if it stops being an employab…

exactly this. FOSS was always driven by those who could code and did so driven by their own intrinsic motivation. those people won't disappear. there may be less people because some are more driven by quick results and while in the past they had to code to get there, now they don't, which means they won't discover the joy of coding.

but for others coding will become an art and craft like woodworking or other hobbies that require mastery.

Re: Vibe coding kills open source

#139

Post by krona »

tomaytotomato wrote: post #6

I have been trying to use Claude code to help improve my opensource Java NLP location library. However trying to get it to do anything other than optimise code or fix small issues it struggles. It struggles with high level abstract issues. For example I currently have an issue with ambiguity collisions e.g. Input: "California" Output: "California, Missouri" California is a state but also city in Missouri - https://gi…

If Claude read the entire commit history, wouldn't that allow it to make choices less incongruent with the direction of the project and general way of things?

Re: Vibe coding kills open source

#140

Post by korenmiklos »

dizhn wrote: post #28

I don't really read papers and haven't read this one either but that summary. > In vibe coding, an AI agent builds software by selecting and assembling open-source software (OSS), Are they talking about indirectly due to prior training of the model? No agent I use is selecting and assembling open source software. That's more of an integration type of job not software development. Are they talking about packages and l…

Author here. By "returns" we mean any reward the developer is aiming for, whether money, recognition, world fame, future jobs, helping fellow developers. Sorry, econ jargon.

AI agents can select and load the appropriate packages and libraries without the user even knowing the name of the library, let alone that of the developer. This reduces the visibility of developers among users, who are now less likely to give a star, sponsor, offer a job, recommend the library to others etc.

Even as a business user, say an agency building websites, I could have been a fedn of certain js frameworks, hosting meetups, buying swags, sponsoring development. I am less likely to do that if I have no idea what framework is powering the websites I build.

Our argument is that rewards fall faster with vibe coding than productivity increases. OSS developers lose motivation, they stop maintaining existing libraries, don't bother sharing new ones (even if they keep writing a lot of code for themselves).

Post reply on HN