5 ms·
> For better or worse, React is heavily overrepresented in the training weights, and “agent experience” is starting to matter more than developer experience. A
by chanux 13d ago
> For better or worse, React is heavily overrepresented in the training weights, and “agent experience” is starting to matter more than developer experience.
Another winner takes all situation. This and the other post on agent tool use[1] makes me think we are in a period of great homogenization. I'm sure there's a group that's delighted about this but for me, it just makes me feel a little uneasy.
[1] https://armature.tech/blog/which-tools-coding-agents-install https://armature.tech/blog/which-tools-coding-agents-install
- redwood 13d agoSee https://arxiv.org/abs/2407.02209 https://arxiv.org/abs/2407.02209 but I hope researchers are looking to tackle this problem
- cookiengineer 13d ago> period of great homogenization It's called alignment. And that's the correct word for it in my opinion. The most important part is that we have to find a way to tackle the knowledge enshittification problem. Because now that slopcode and slopposts are everywhere, the average quality level will decrease. We had the peak internet knowledge, and it's now gradually getting worse. That's not a personal opinion, that's how attention matrixes and their keys/values computation works.
- skew-aberration 13d agoIt was once called 'natural monopoly'
- bluefirebrand 13d ago> It's called alignment Alignment is such a polite term for "a giant boot that is crushing all of the uniqueness out" Once it is done crushing, congrats! Everything is aligned.. by being flattened
- bhawker 13d agoDoes the world need the "uniqueness" of 40 similar web frameworks?
- bluefirebrand 12d agoPerhaps not but the world does need the uniqueness of 40 different people who each made their own web frameworks
- camillomiller 13d agoAs long as there are standards, yes, why not. In Italian we have a saying: il mondo è bello perché è vario. The world is beautiful because it’s varied.
- arvinsim 6d agoImagine if we have 40 variants of assembly. Would software development have developed at the same pace? Personally for me, I am much less interested in the implementation in frontend and more interested in performance, accessibility, responsiveness, etc. Frontend frameworks are just means to an end. If we can standardize to one, then I will have more time to focus on the more important parts.
- jansan 13d agoSince a good LLM can handle any of those 40 web frameworks, I think having 40 web frameworks is becoming less of a problem than it used to be. Actually, implementing ideas in an individual manner will become easier. Remember the Bazaar and the Cathedral? The Bazaar kind of failed on the client side, but LLMs may revive it.
- sznio 13d agothe world certainly needs novel ideas. maybe eventually we'll come up with a web framework that doesn't suck - or rather we'll realise that Django had it right all along for all these years.
- gmac 13d ago
- altmanaltman 13d agoWasn't that true for tech in general though? I mean I have heard developers bitching about React for years now and every year I saw it get more adopted at the same time. Of course some obviously do not like it but the decision was usually driven by business and not coders. I feel its a simillar thing happening with AI code taste being primarily developed by big tech firms which ofc love homogenization. But its not all bad, I don't think people will just give up working on what they like because agent expeirenxe got better but who knows though
- afiori 13d agoA lot of the bitching about react always had the same flavour of the complaints that JavaScript dares to have a NaN To be fair it is stupid that NaN !== NaN but that is how IEEE and CPUs made floats to be
- jjgreen 12d agoIt's not stupid when you realise that NaN is not "not a number" but rather "not a particular number", this becomes clear from hypot(Inf, NaN) being Inf rather than NaN
- afiori 6d agoNaN means only NaN, 0/0 is NaN at the cpu level. to my weak undestanding only pow and hypot behave like this; pow for math reasons and i could guess that hypot does that to sorta have |x| <= hypot(x,y)
- bryanrasmussen 12d agoI mean to me the whole point of NaN is that it should not equal NaN, so it seems sort of like you are making the complaint in the first sentence in your second sentence, only in a slightly more abstracted way. Unless you are of the opinion there is another better point to NaN? funny story, I was once interviewing for a job where the tech guy did not know that NaN existed. I guess JavaScript wasn't his language, but not sure what was, and the code was in JavaScript for a Web Dev role.
- crabmusket 13d agoI wonder if this is overstated. Anecdotally, we use Vue at work and Claude and Codex handle it just fine. I'd expect Vue to be the third best-represented after React and Angular, but even then, we feel absolutely no need to change things up to get better results. Vue's reactivity is quite similar to Solid's, AFAIK, so I wonder why there would be so many issues with Solid. We have also started to have success getting LLMs to create custom lints for us to handle tricky edge cases. For example, discovering deep watches or tricky interactions - you can get a fair way asking an LLM to create a lint using a codemod library (in our case, vue-morph) that fires on this example but not on that example.
- zmmmmm 13d agoVue had a good tenure as a solid #2 to React, so I think it probably has got a lot of representation in training data. Things really splitered after that but it got a good foothold.
- gunalx 13d agoThe problem I see with vue is what vue patterns are you using. And the llms may be trained on more outdated ones more than current practices.
- zmmmmm 13d agoAgree. It wasn't as drastic but I think Vue did the 2=>3 update at an unfortunate time. They lost a lot of users along the way but they also confused a lot of the LLM training too, I am sure. Still, I have great success with it, especially with Typescript.
- sureglymop 12d agoWhat bothers me is that I notice it everywhere. Not only does AI have a distinct writing style, it has a distinct style of frontend UI/UX design. I notice it on more and more websites and it bores me. I want creativity and originality. I haven't really seen AI exhibit that at all. It can maybe be driven there through persistent human input but then I think it's still the human being original/creative.
- embedding-shape 12d agoYou'll get used to everything looking the same, this always goes in cycles and we're bound for it correcting itself shortly. In the past, we had this for web 2.0, Bootstrap, Tailwind (right before LLMs) and now LLMs, it all starts to look the same, then people get tired of it and everything looks unique and novel, until people get bored of that and a new unified design language starts to spread again, leading everything to look the same... It's all just cycles, seemingly forever, judging by the last ~20-30 years.
- FoundTheRetard 12d ago[dead]
- jojobas 12d agoCreativity and originality? Like when you try to scroll a page and it rotates an image gallery instead?
- sureglymop 12d agoNot sure if sarcastic but no. Check out for example this website: https://www.geofungames.com/ https://www.geofungames.com/ The functionality more or less works and the website is usable, but the design follows that "AI style". I'm not sure how else to describe it.
- properbrew 12d agoThis is absolutely a generic AI website. Do you feel the same about one I've developed using AI? https://whistle-enterprise.com/ https://whistle-enterprise.com/ I think the AIslop of websites is down to the lack of vision or user journey and just going with the default which will always be generic and bland. If it does feel in the "AI style" I'd love to understand what it is.
- totetsu 12d agoIs anyone keeping track of these .. second order effects.. of the proliferation of agents? Like this, or oss projects getting 10s of 000s of bug report, or the Linux kernel site getting ddosed by crawlers..