I agree on everything you’ve said. Rust hype is a bit annoying and can be off putting. AI bros seem to have amplified the effect and I strongly dislike it. GUI is probably the worst area of Rust, but given the universality of Rust, I think it’s expected. I found egui easy to use, though, but far from swift ui, still.

Having tried both Swift and Rust - you can hardly compare them. For me Rust is in a different league. Swift is a modernization and a step forward compared to ObjC, but Rust is a different paradigm altogether. Swift has IMO done a great job for Apple as a platform, but has not achieved universality and had not transcended the platform. Rust has achieved universality, but has not captured any platform in particular yet. Both have not and may never replace C/C++, though my biased opinion is rust may have a chance (at least with C++)

What?! Well, surely Nintendo is better then!

I really hope so. 200k is peanuts for google, but probably significant for Motorola.

Honestly, I expect it will become increasingly hard getting a company willing to do something like this. Lately, all of the companies have started heavily enshittifying and I don’t know many that still care about their customers. And it’s a domino effect - they know you ain’t going anywhere, all the others are just as shit. I remember a time where you had a 100 phones officially supported by Lineageos and were cheap. You could get a Xiaomi, Huawei for like a $100 and they would run lineageos almost out of the box. Or a Nexus 5 for a few extra bucks. As soon as they eliminated the competition, they’ve become closed entirely and are basically spying machines. I don’t trust Motorola will be trustworthy in the long term, if they capture any market. Just a few months ago, they were caught hijacking affiliate links on their phones (link). I don’t know when the consumer situation will change, but I’m not holding my breath for Motorola. Of course, if they make a good phone running GOS, I might try it if my pixel kicks the bucket.

Well, we had a good run. It was nice while it lasted…

True, People can be obnoxious. But you should still try it at some point, there is something to the hype.

Yeah, you know that episode where Homer does pranks for Mr Burns and goes in and buys a no 1 spiderman from the comic book guy and eats it in front of him. That’s how most people feel about book destruction. Nothing legally wrong, but disgusting nonetheless.

It works for them not scraping their own slop back into training data. I assume that is actually the real purpose of the system. They don’t want to share the key with the public. But they probably will with other llm companies in exchange for theirs.

My tinfoil head tells me they don’t want to ingest their own slop for training. But this only helps them if others do the same. Data poisoning is a real problem for llm future.

This won’t work - you can always use a dumber, open model to rephrase and remove the watermark. All this does is marketing for anthropic, which I see most people are buying. Gives you a false impression you will easily be able to figure out if it’s done by AI, which it is not. Liars are still gonna lie, cheaters still gonna cheat.

Endgame is training data is audited and ethically or at least legally acquired for ALL companies making models.

Ugh… I don’t know whether to upvote or downvote…

So AI is whatever makes you right and me wrong, basically? Of course the ones responsible are the people building products based on AI. Nobody is arguing LLMs themselves are evil. What I’m talking about are massive datacenters built overly fast so that they need to use evaporative cooling and micro-turbines is an utterly irresponsible use of that technology. It’s like rushing to build a faster car than China and so you don’t install brakes and your cars start killing everyone on the streets. Of course a car isn’t evil, it’s the people irresponsibly designing them. AI tech is constantly doing propaganda fear mongering that if you don’t use their product RIGHT NOW - you’re getting left behind. If you don’t integrate them in your business RIGHT NOW, you’re gonna get overrun by competition. And the product is not nearly as reliable as they push it and not nearly as powerful as they claim. It’s a universal tool that is inefficient at any specific, specialized task. In fact, it’s painfully INEFFICIENT. But you and I wouldn’t know it, because nobody is publishing exactly how much it costs to run and how much they are subsidizing the use of it. The technology is useful, even revolutionary, but the way it’s overwhelmingly being used right now is bullshit. This way is not sustainable, not profitable and will end up costing the entire world. That’s what’s gonna happen. The technology will stick and people will find best uses for it and start paying attention about its use. But I doubt it will survive in this current form.

Dude, USA affects all of the world already in a very significant way. Western world especially. While the decay of American imperialism is something a lot of us would like to see, it will, in the short term affect everyone negatively. The USA will not go out without a fight. Already, people across the world suffer because of it. It is precisely because of the USA and their economy. Economy, which is undoubtedly, by their own accounts, dominated by the AI tech industry. So, although indirectly, people outside the USA are negatively impacted by the AI. Not to mention direct impact it has (military literally being driven by AI).

Are you me? Cause I have exactly the same setup. Navidrome, tailscale, the whole thing. I also use strawberry. It’s OK but a bit basic. I recently tried Nocturne for desktop, looks promising, but still somewhat buggy.

Hahah :(

Why is everywhere you look all of the leadership doubling down on idiocy? All of the western world, not just usa.

IMO, qwen3.6 a3b is smarter seeming but gets stuck way too often, Gemma 4 is stupider, but doesn’t get stuck almost at all. I’m trying to find middle ground, but so far - no luck.

I do, but I am becoming increasingly more disappointed as time goes on. Not just self hosted, llms in general. They sometimes help, but they mislead so many times and waste time that you don’t even notice. I think that’s the trap. When you succeed at a task, you become impressed but don’t notice how many times it failed doing a simple task. And as soon as you scratch the surface, you see how you would have done it differently and perhaps in a better way. Even just googling is bad. It does research for you, but it has no critical thinking and can’t decide what is better from the results it gets (other than google ranking) so it often leads you to think it did as good as you would, when it’s nowhere near as good. Every time I did the googling myself after it did, I did it much better. And I mean MUCH better. Ask it to find the app, it misses the most important ones, hallucinates a bunch, for ex. I found this to be the case with frontier models as well.

Self hosting has its benefits, but seeing how the ecosystem looks right now, concluding this is a huge bubble is inevitable. It reminds me of crypto so much. It looks rich and plentiful, but as soon as you dig a mm under the surface - nobody has tested it, it’s got a critical bug, it is overblown and there are issues with no response. No docs, no info, no nothing. For the biggest thing in technology in history, it is awfully hollow. I don’t mean it in a condescending way, in fact community is enthusiastic and very helpful, it’s just that it doesn’t live up to what most would expect.

A caveat I need to mention is I have not used it for coding - I have an irrational fear and resistance towards it, being a programmer. I just won’t touch it, even if it means the end of my career. I’m trying to be grown-up about it, but so far, I dont want to use it, for good and bad reasons.

I have a model with 64GB of ram. I’ve limited context to 16k, in an effort to make it more stable, but tbh - it is rather unreliable no matter what I do. With my setup - mlx_lm and webui, it frequently collapses or loops, no matter the settings. I have done a lot of debugging and have concluded it is probably inherent model behavior.