I am just waiting for the time when all of us (already or soon) out-of-a-job folks will start reaping the "AI abundance" (that "AI leaders" can't get tired talking about, including our own Mr Huang) while some folks already make billions and millions in these trillion dollar IPOs (soon more to come).
Because if that doesn't happen then I have a feeling the world will proceed something like this: AIs paying for things to other AIs (as a circular gesture; possibly different from the circular finance of today) while some rich folks will sit in their safe island citadels or in NZ (etc) nuclear bunkers controlling the arms, ammunition, surveillance, medicine, finances, along with pretty much everything else of the world while the rest of humanity, the majority – who couldn't tap into that abundance, will finish each other off one by one – or maybe worse, they will be live to witness it all. I try really hard not to picture this post apocalyptic scenario but either all this IPO, moat, and whatnot are hogwash, or that is happening :)
Agreed, for once. We do not know what basis there was for the suspension. A trigger-happy registrar in the EU might have done the same.
As the domain was already restored (https://news.ycombinator.com/item?id=50545191 ), I imagine it had been suspended in error, probably in reaction to some bogus takedown notice from someone with an ax to grind.
Not exactly. Politicians write laws, law enforcement hires a prosecutor to enforce the law, prosecutors get a judge to issue a court order, the Company has to comply with the court order or their executives go to jail, the executives are not going to go to jail for you or anyone else. So yes relying on a US company means implicitly that you must follow US law.
But it's worse than that. The CLOUD Act means that even if they use a non-US company, that company can be compelled to release your data to the US, and possibly act on US demands, if the company does business with the US.
Maybe I'm too cynical, I wouldn't expect less from other countries.
If I run a US based company, say a SaaS business, and want to do business in the EU I have to follow their laws with regards to data protections, cookie popups, etc. How is it different for EU companies wishing to do business with US customers being expected to comply with US laws?
Yes but they could still use a registrar that operates under their laws, is more friendly to them and does helpful things like notify them of the problem and give them a chance to remediate it before suspension. Yes the US can still reach them but it would require due process. Of course we only have one side of the story here, it is possible Codeberg did not respond in the required timeframe for something.
Oh you can suspend and destroy anything – not limited to domain names. The important part is – can someone stop you, or can someone prosecute you by virtue of you being from certain countries and the power and reach of said countries?
I've learnt and acknowledged that all this talk of "use your own domain", "degoogle", "pay for the email" (and I do), etc are safe until someone resourceful enough eyes your domain with the wrong kind of look.
.eu or member state cc-TLD paired with providers with matching jurisdiction seem fairly solid, based on what I heard. They would have to ask a court at my place, and that court will discuss the matter with me in my preferred language before doing anything harmful - and apply a high standard to claims of urgency.
When I see one of these posts I feel quite left out, if not FOMO. I would never know when that browser named Arc arrived on the scene, did something (quick search kinda tells me: agentic browsing for humans? maybe?) and apparently broke many hearts in the process (or just left without saying goodbye?), and so badly that people haven't been able to get over it and are still trying to get it all over again. Must be something. And here I am, still stuck on, like a broken record, which browser consumes less RAM (fluctuating betwreen Safari and Firefox all this while) – periodically ending in very sad realisations that in most cases it's the websites now, not so much the browsers anymore, unitl the search resumes for the n'th time.
restic, tarsnap, borg, and kopia – I am planning a way to avoid backing up certain folders (so that FDA won't be needed for these apps; but certain special directory folder permissions will still be needed). Then there are a few objective-see apps, an app-uninstaller app, and a disk analysis app.
Askers? Too many. I have been removing them from there using the "–" button but I feel that might be counter-intuitive as these keep asking and then turning up there anyway. So I will stop doing it now.
Will this allow me to permit apps only "read" (and/or specific write i.e delete, change) access on chosen directories? Not clear from the article. I'd love it if I can give read-only access to many backup apps I use on specific folders.
Apple had the chance to be granular about directory permissions but it chose rather a mess – there's FDA and then there's also specific folder permissions and they sit separtely. UI adds to that mess – sometimes just by looking at the listing you cann't tell whether the permission is given or not because in "files and folders" permissiosn view it lists FDA if available and greyed and in some cases it doesn't even that app has FDA, and sometimes it lsits that FDA even if that app doesn't have FDA but asked for it. It will be even less sane if there's an app that asked for app data permission and you have many apps installed because now there are those entries for "apps' data directories". It's painful to both users (technical and non-tech nical) and the developers.
I think one of the reasons is Apple is paying the price of closed-source development. Something else is needed? Just add few hundred flags and make it happen one way or the other. Things might break? Well, add flags in other services/platforms as well, after all it's all in-house. User is still facing problems? Well, take them on phone calls, spend a few hours with them over a few calls at various levels of "seniority" (but actually showing no difference at all) and then make them reinstall the OS a few times and then close the case saying their exerience has been noted, without providing any issue/bug number or so. If they still insist? Tell them there's no next level of escalation and maybe possibly give them the option of dropping the device at an Apple centre (in most countries not the Appele stores) and remind them it can take from weeks to months. As a mac user it boggles my mind how unsafe of a ticking time-bomb of an OS we/I are using.
I just wanted to mention that there's a lovely little app called `2.1.278` and they also ask for FDA. I know the version is a bit old, can't seem to keep up :) Haven't moved all that behind containers and sandbox-exec is a bit unwieldy.
This trope would end soon. There's a tiny fraction of computer science students who will have "value" in the age of LLMs and they're the ones who actually would work on LLMs/models and stuff very close to those aspects of AI. They have value even today. Rest will be fair game, as they are today.
I'm very open-minded about entertaining drastic societal changes but I just don't think the facts support this prediction.
1. LLM performance has more or less peaked. Since Fable 5 released in June, every Claude model has been a ~1-2% improvement over the last one (if you go by benchmarks). Fable 5.2 marketing verbiage has shifted to how it's "lower cost per task" which imo marks a regime shift. I would bet money that this time next year, we're not going to have a model that's obviously and significantly better than the best ones we have today.
2. If you hold that this is about as good as LLMs are gonna get, we haven't found ourselves at a point where we can completely remove the software engineer from the seat. A software engineer is still required to create real commercial-quality software. The role hasn't been commodified either - a better software engineer still writes better software.
I’ll also anecdotally add that I’m busier at work than ever. Tasks can be completed quicker, (some much much quicker, some hardly at all) but there are so many more tasks than there used to be. Features/improvements/optimizations that used to be out of scope because they weren’t top priority, and just weren’t worth the time.
I find it hard to believe we’ll end up in a permanent state where Terence Tao is crushed but I’m not. I think more scale and better RL will work wonders for LLM performance over the coming 12 months.
Terence Tao is a robot who solves math puzzles. We built an actual robot that does that now, hence he got crushed. It's not very different from being good at a sport or board game. The space in which they operate is soothingly well-bounded.
Can you imagine LLMs replacing a small company's CEO for a non-trivial period of time (even forgoing physical requirements, i.e. all I/O may happen over email)?
> Can you imagine LLMs replacing a small company's CEO for a non-trivial period of time
Perhaps not right now. But performance on long-horizon "fuzzy" tasks is growing rapidly (see eg Fable 5.5 performance on remote work benchmarks). Why would that trend suddenly stop now?
(A human CEO wouldn't be very effective over e-mail alone either, but that's a separate issue.)
I'm sure people also thought that AGI was right around the corner when Deep Blue beat Kasparov.
LLMs are good at math because of the verification loop which allows them to implement a guided search algorithm. They don't "invent" new math, they explore many different branches applying pre-existing results and see what works.
Which is not an approach you can generalize to other fields (such as writing text which doesn't read like AI slop, or code which is architecturally sound)
> I'm sure people also thought that AGI was right around the corner when Deep Blue beat Kasparov.
I sure didn’t.
> Which is not an approach you can generalize to other fields (such as writing text which doesn't read like AI slop, or code which is architecturally sound)
Isn’t that just a question of building more / better RL environments? Sure, math is easier than most domains because a proof can be verified for correctness with certainty. But I don’t see why you couldn’t build RL environments that reward models eg for architecturally sound code.
That has been said over and over for the past ~4 years, with this many resources, the field will keep finding ways around bottlenecks. Even as it stands now, these models are good enough from an economic perspective to automate a huge share of work, which is why everyone is focusing on efficiency now
Idk about this. I've used almost 10B tokens of Fable 5.2 in the past week (ultra plan, and the 3 free resets they gave) and the output quality is absolutely step change above even Fable 5 or Opus 5.5. It's probably heavily dependent on the domain you work in.
It's actually so jarring to go back to Opus 5.5 and seeing it do all the usual LLM mistakes again.
There's a tiny fraction of developers who have to worry about silly things like memory allocation operations because that has been automated and we mostly work on a higher level. But it's still important to learn about it nonetheless. The same will happen with AI. It was already true in 1995 that you didn't need a degree to do software, but learning computer science was still valuable.
> I fundamentally don't want to use a node-based tool ever again
These are not (nostly) written in JS/TS, so I think these are not node based.
https://github.com/openai/codex (this is rust but I don't absurd dependency depths rust seems to be getting, so in that aspect rust tools is becoming like npm tools)
As for how reliable these are, you'd have to try them and see for your own usage. I am sure there are more. Lighter ones, maybe less famous, but I have tried these and I am back to using Claude Code most of the time :D. Though I am trying to get familiar with Goose. However given a choice between OpenCode and others, I won't even bother with OpenCode – it's really that bad. Though I've their Go plan and that has been very userful for me (super vfm).
I also don't get the point of Pi. Pi either has to become Claude/Codex for most coding tasks and then what's the point it, otherwise a tiny harness (or script) replaces it, so I've also stopped using Pi. Apparently there Pi ports in other languages.
> I also don't get the point of Pi. Pi either has to become Claude/Codex for most coding tasks and then what's the point it, otherwise a tiny harness (or script) replaces it
Do you have an example of tiny harness?
I use Pi, without any plugins and it is enough for me, I don't need more. The only issue is that I don't like JS/TS (its place is in the web, not in terminal) and I would prefer something written in a compiled language (C or Rust preferably) or at least Java/Python.
Oh, and I would need integration with Github Copilot and Claude APIs (api pricing) so I could use what I get at work.
Comment below mentions hax, it looks promising (I don't need mcp, having "code mode" would be nice, but I'm still not sure if it is useful, and I would prefer it to not use JS).
Since I started using my own domain at bsky I kinda like this platform. I never have to even worry about the "username" which used to be a major source of concern especially if someone had already picked my favourite ones :)
But Tangled, or any git platform (even free ones), feel a bit off as they don't allow private repos at all. This is a very conflicting aspect of some open and privacy promoting platforms. I am sure they have their reasons but that doesn't make it hard to reconcile wiht it.
Because if that doesn't happen then I have a feeling the world will proceed something like this: AIs paying for things to other AIs (as a circular gesture; possibly different from the circular finance of today) while some rich folks will sit in their safe island citadels or in NZ (etc) nuclear bunkers controlling the arms, ammunition, surveillance, medicine, finances, along with pretty much everything else of the world while the rest of humanity, the majority – who couldn't tap into that abundance, will finish each other off one by one – or maybe worse, they will be live to witness it all. I try really hard not to picture this post apocalyptic scenario but either all this IPO, moat, and whatnot are hogwash, or that is happening :)
reply