Skip Navigation

InitialsDiceBearhttps://github.com/dicebear/dicebearhttps://creativecommons.org/publicdomain/zero/1.0/„Initials” (https://github.com/dicebear/dicebear) by „DiceBear”, licensed under „CC0 1.0” (https://creativecommons.org/publicdomain/zero/1.0/)F
Posts
1
Comments
43
Joined
7 mo. ago

  • 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.

  • This looks promising, I'll be sure to give it a try! Is it better compared to Qwen3.6 in terms of getting stuck? So far, qwen is borderline unusable for me, for chatting/research, as it frequently gets stuck. Gemma4 is a lot more stable, but a lot worse, in my opinion so far. North mini looks promising if it can solve looping for me and be a bit smarter than Gemma.

  • 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.

  • Are you running an mlx model? If not, try that. My m4 macbook runs qwen3.6-35b-a3b lightning fast. Has its issues, but fast nonetheless.

  • No one is arguing that Bambu has the right to keep their sources closed. That is NOBODY's argument. Instead, you've not addressed the main argument and the point stated by Jeff - they threatened the open source developer with lawsuit and asked him to remove the code. The code doesn't violate anyting. At the same time, THEY based their whole product on opensource software. That makes them hypocrites and shitty people. That's the argument. Your post is long and doesn't address it at all.

    But let me also push back on some of the things you said.

    Bambu spends an enormous amount of money on r&d.

    So does Prusa. And Creality. And many others. They were all open source. They are now being very cautious about what they publish. It may not be Bambu's fault, but they certainly started the trend. It looks a lot like the ol' microsoft eee tactics again.

    The replacement parts are cheap. Comparable to other brands.

    They are also non-standard and not accessible to others to make and thus - the only guarantee they will stay cheap is their good word. Which is not worth much IMO. Cause I don't think they will allow competition to make the replacements without licenses.

    Why would you want to on a Bambu. ? It already works optimally.

    To customize it for your needs? What if you have no internet access on your print farm for a day and you require Bambu cloud services to work to be able to manage it? What if when you bought the printer, you could use it in LAN mode, but now after an update, you no longer can? Is it your fault still? This really happened already.

    if you want a tinker printer than get one that’s open. Most people don’t need that. They just want to print things.

    Fair point. But they did change the terms of usage after the sale once before. They're gonna do it again. This alone makes them shitty and unfit for a new printer purchase for me.

    Use a different brand if that’s a problem for you.

    I hope most people will. But again, this wasn't the original argument. Though even without the original argument, there are still reasons not to buy Bambu.

  • Smelled funny from very early on. I wonder if now all those creators on the bambu payrole will pretend like nothing happened. Channels like CNC kitchen or Tom Sanladerer come to mind. But also 99% of all other channels, bambu seems to be sponsoring a shitton of channels. At least Jeff spoke out immediately. I suspect most will keep quiet.

  • How so?

  • Tuxguitar is very good, IMO, like old guitarpro used to be (4/5 versions). It supports those formats directly.

    Other than that, there is ardour, though I am more used to reaper, which is not OSS.

  • Wait, but Mythos is the revolution in the software security world, it found 0-days in all popular OS's, including FreeBSD. I'm sure it would have found critical bugs in their own code! /s

  • Oh, definitely I'm not saying people should just jump the gun and replace their distro for one without systemd immediately. I certainly won't, at least not without thinking about it for a while. But I also think that denying the controversy exists is not good. This is definitely controversial, for some people even a deal breaker and there are valid, real reasons why. For the rest, it's good to look at what options there are, see that there really isn't an appropriate alternative for systemd in some cases and realizing that a successful fork would be a good thing. Also, a long time criticism of the community has been that systemd does too much and it being against basic Unix philosophy. I always thought of it not being a big deal, given its modularity. But I now realize that it centralizes control and design decisions to a single org and that is certainly a weak point IMO. So a fork makes a lot of sense, but it is at this point a mammoth of the project, so it will be really hard to maintain.

  • Yes, but what's wrong with this? If you gather engineers that are capable to maintain it - what is the downside? Systemd could always have used a bit of competition, I think most of us can agree. Most of the forks of systemd will fail, but most of all projects fail after some time. I don't think this situation will harm systemd ultimately and it shouldn't.

  • I agree with all that you've said. But why add it now? Why haven't they added it a long time ago? Or if now they remembered, why not other extra optional fields that some people might want, like gender, sex, any other field? Oh, it would be too political? I see...

  • I agree to some degree, but I think the issue of age verification is beyond this point. Yes, Linux users tend to be much nerdier and reactive than the general public. But they are the ones who use linux in the first place. Whether they gatekeep linux from others is another story, but the devs should know their audience by now - and hopefully care. And what's more - a lot of idealists (I wouldn't call them autistic, though that may be a factor) hate systemd in the first place. They already dont use it or don't want to use it. So the ones that do, I argue, are more mainstream. I am one of them. I don't want to go back to sysvinit and write a script for each new service. I also know that this doesn't end here. Today they add the field, tomorrow, some mainstream browser will depend on it existing and the frog will be boiled. Now it is not an API, but it's added in case anybody needs it. So you didn't even have to add it. And they didn't add a gender field in case anybody needed it, for example. Yes, Linux community would probably start arguing about that, but not nearly as much IMO. I think this is far more mainstream issue than you give it credit, honestly.

  • That is not the point. If it was so logical to add, why add it now, when you know it is controversial? The devs are aware of the controversy, they have made a political decision to do it this way. At the very least, they could've handled it with more care - as sensitive matters should. Turning a blind eye and pretending this is business as usual is very insulting. To me at least, and I'm sure to most who care. If you do this during "the surveillance state paranoia", you have to be aware you are contributing to more of it.

  • Wow, some good news on Lemmy? Sign me up!

  • Yeah, im in a similar situation. Curve doesn't work in my country and banks don't have their own solution. And google pay won't work on my grapheneos pixel.

  • At least the last one won't happen, as banks would have to be on board. And banks are not on your side with this one.

  • It would affect a lot of users, then it will indirectly affect you too, as a lot of devs won't be as interested in maintaining their apps for so few users. But I hope it will at least give a bit of a push to developing postmarket os. I personally am sure going to get a second hand phone to install postmarketos too and hope I can contribute at least a little bit. I am prepared to suffer, at least a little bit for the right cause.