décentraliser les couts de l’IA ne la rend pas plus efficace. Le local c’est pour des questions de vie privée, pas d’environnement
On est quand même en droit de penser qu’il vaut mieux avoir des GPU à 400W qui font des spikes occasionnels et déversent quelques calories dans le buffer thermique d’une résidence (où elles sont compensées mécaniquement par moins de chauffage etc…), que des data centers en spike permanent qui dumpent des MW de chaleur dans leur environnement immédiat.
It must be nice to have technical opinions that don’t need to be grounded in facts. Why would you check your assumptions when you can just vibe.
I mean the ones in the latest release of rsync, tf does nginx have to do with anything ?
Which of the CVEs in question are you referring to?
What about the 6 critical security bugs he fixed in that release. Didn’t rsync need those “contributions”?
It was a mess from superfluous complexity from adopting every buzzword along his career, cloud, microservices, configuration management
That was the bane of my existence before AI and i suspect AI will only compound this issue.
If you do “artisan vibe coding”, acting like a very hands on CTO that challenges decisions and reviews most of the code produced, you get a modest productivity boost in the 20 to 40% range, and a large reduction in cognitive load which can help you think bigger thoughts on the longer term. The quality can be as high as you want it to be in that setup.
But if you do fully agentic unsupervised vibe-coding, it’s easy to get into a mess because it’s like having a team of junior/mids paid by the line churning out complexity all day long. The productivity boost can be a large multiple but the quality suffers because you have to ignore a lot of the stuff and the devil is in the details so he will certainly get you at some point.
Maybe you’re right
oh man while i do love being correct, deep down i hope you’re the one who’s right and i’m overreacting
If you’re making a broader point don’t single me out
Yes that is a communication lapse on my end, i try to make issues personal to emphasize them but it’s not always relevant. I don’t think it should detract from the broader point, sorry if it does.
My sensation is that we’re doing fine for now
I think the data shows that we’re far from fine and already resource-constrained on most critical projects. It’s not that people stop caring about open source, it is still fundamental to the way the web works. It’s just that they don’t feel personally compelled to pitch in because they think we’re doing fine now. The wikipedia analogy works well here : it is still fundamental to the way people get information, but it’s chronically understaffed and may already be in a death spiral.
I do contribute time and donate money to open source project so… miss?
You’re missing the point. Sure you do, that’s a nice anecdote, but the data shows most people don’t. You are part of a shrinking cohort that is already insufficient to maintain what we need in the long run.
If we can’t get enough resources to support even the most basic infrastructure then the experiment will end
And then what ? Only large corporations can finance their own in-house tools and they gain even greater advantage against the rest of society ? What a great outcome…
Your point of view is not crazy but i think it suffers from too much optimism in the face of bleak data.
I don’t know why you’re insisting on this anyway
my guess would be low executive function and a smidge of autism
Also, having critical software depend on one guy is not safe. We should avoid that. If critical software depends on one guy it should be phased out.
I’m sorry to say 90% of the internet’s load bearing infrastructure is in this situation. It’s just how the story goes, everybody wants to build low-stakes toy projects, nobody wants to do high-effort low-reward infrastructure work.
“Writing something new using modern tools” is all fun and sparkles, but then you run into the same issues as rsync except without the experience. Then you get attention from attackers, you get security issues, which you have to patch with defensive code which is not appealing to read and zero fun to write. Before you know it your project is “decades of Rust/Zig/Lisp” which nobody wants to touch and you’re back at square one. All you’ve accomplished is give the attackers a few years of low hanging fruit and easy exploits.
There’s a reason why we get a million shiny toys a year but solutions like rsync stay entrenched for decades.
I think what you’re missing is that the number of people doing step 4 has been going downhill steadily since the 2000s. People start open source projects yes, which for 99% of them don’t bring in any users and barely get maintained over the long run, but the pool of people willing to contribute to large established projects is so small it is becoming problematic.
Even Wikipedia is having its own editor crisis, where most of the power editors are greying out and barely anyone is stepping up to replace them.
And this is happening exactly because most people, like you, think that the free infrastructure around us is a fait accompli which doesn’t require us to personally get involved in their maintenance, and that we can even afford to scare away those that do contribute.
Je comprends pas, c’est un clone de minecraft avec une génération de monde tellement précise que chunkbase marche ?
ain’t that a spicy squirrel…
I think you misread my comment. I’m depressed that people are harassing open source devs, not that open source devs use LLMs.
I don’t give a shit whether a maintainer like Tridge uses AI, because i trust them to review the AI’s code like they’ve reviewed human contributions since forever.
I mean obviously mileage does vary from project to project and task to task, but i think you might be overestimating mid-level developers. Or you’ve been really lucky with your recruitment ! Cause i would describe them just the way you described Opus. Pretty eager, kind of try-hard, decent engineering chops but often misdirected with dumb approaches.
Of course my experience is limited and i’ve never really been in a managing role but i’ve been the adult in a fair number of rooms and i’ve done my share of “grooming sprints” and dispatching tasks.
That being said, there are projects that are horribly reluctant to agentic coding. It’s pretty rare as most codebases nowadays are bog standard and rely on roughly the same abstractions, but i’ve seen it happen. It can come from the complexity of the domain, or of the codebase, or from the way documentation and tribal knowledge clash, or a myriad other reasons. Often it’s the kind of projects that require more mature devs and can’t really onboard juniors/mids.
When it digs itself into a hole, it’s very bad at trying to amend the mess that has accumulated
Oh yeah definitely. Once it’s in the hole you better scratch that branch off and restart with more specific instructions cause agents are very “additive”, they don’t often think to remove stuff and change their approach. Again, kind of like mid devs once they’re committed to an implementation plan.
Oh man I’m like super agreeing with you. Also I’m in a place that actually has universal healthcare, so it’s not like it’s unworkable
I most certainly won’t lol
No disagreement here but I’m not sure that’s on the menu right now
Pour le coup je pense que l’océan c’est le pire endroit possible pour dumper autant de chaleur. C’est vraiment très problématique surtout que ça nuit à la capacité de l’eau à absorber le CO2 de l’air donc ça a un effet composé.
Après pour les résaux de chaleur c’est super intéressant je ne sais pas dans quelle mesure c’est possible mais du coup un DC avec énergie décarbonée + production de chaleur pour la communauté locale ça devient assez attrayant comme proposition.