11·
13 days agoThe weights for the neural network or the embeddings?
- 56 Posts
- 1.59K Comments
Joined 2 years ago
Cake day: September 1st, 2023
You are not logged in. If you use a Fediverse account that is able to follow users, you can follow this user.
364·13 days agoIn a decade, most games will be cross platform but compiled for
windowsproton and people will have forgotten why. Then somebody or some group will come up with “cross platform compilation” and the circle will start a new only to return to proton or some form of it.
What is a “kernel” in this context? It doesn’t seem to be related to the OS kernel but some kind of graphics kernel? Whatever that is…
5·13 days agoAre you doing his in any official capacity? Who are you? And who is “us”?
3·14 days agoHow is it possible that Huawei is the third biggest phone vendor and most apps still have trouble running without Google Services? I thought they were not able to come preloaded with Google Services for a few years and that they have many customers outside of China. Are all those customers just limited to Chinese apps?
Did they resolve their issue with that other company? I can’t remember what was going on but the owner of automattic was pissed about them not contributing back or something? Did something change?
Anything built on top of atproto I tend to distrust. Doesn’t it all hinge upon the makers of bluesky hosting their central node for it to work?
2·24 days agoWhy wasn’t free threading a new version of python? Having everything suddenly going from thread safe to C level insecurity is a major change, not just a small feature update. In fact, if they had to redactor the interpreter to support it, that sounds like even more of a reason to make it a major version update. Doesn’t python follow semver?
Same. I’m thinking of cancelling my subscription and just sticking with what works. I’m not sure I had a really useful update in a while.
What are you doing to get into dependency hell? Never had that problem. Are you running “pip install” in your venv at will?
Right now, the jetbrains IDEs are my favourite because they are proper IDEs, not some editor with a bunch of scripts in a trenchcoat pretending to be an editor. But the company is starting to lose touch with its customers: developers who want an IDE for productivity, not a VS Code lookalike. It’s like the company is finally being taken over by managers who don’t know lick about development and it’s starting to show (at least to me).
Now, I’m on the market for a new editor and even willing to pay, even though I’d prefer paying for an open source IDE. Right now, Zed is looking interesting. The only thing that bothers me is how loud people were about it. Hype destroys my faith in stuff as it’s often just good marketing. Another thing that bugged me is that when they started, they were “Mac first, Linux maybe”. But now that the hype has died down, there’s much less “omg, zed is the new editor and it will be better anything else” type posts, and it supposedly works on Linux, I can give it a try.
Execs that float this shit can suck it. If they don’t want to hire, don’t let them saddle you with more work. I work my hours, do what I can and clock out. If they ask for more, I tell them what I’m working on, how much it takes, and when their stuff will be able to get slotted in. They can try and squeeze more out of you, but only if you let them. Don’t let them.
They can fire you and try to hire a sucker who will - or keep running through suckers that do, but it won’t do their company nor their clients good, at least not in the long term. But these people are only around for the short term gains.
Why do people always have to use some freemium offering when there’s an opensource, self-hosted or already hosted variant out there? I don’t get it. Just riding the wave I guess.
27·25 days agoI honestly thought this was going to be a post about how they finally are going to trademark nix and the logo. Can they even trademark it? My feeling is that “nix” is just too ambiguous a term to trademark.
Even better!
Company creates problem. Requires users to change because of created problem. You defend company creating problem.
That’s the logical flaw.
If you see no flaws in defending a monopolist, well, you cannot be helped then.
I would like to remind you that you are arguing for a monopolist. I’d agree with you if it were for a startup or mid-sized company that had lots of competition and was providing a good product being abused by competitors or users. But Github has a quasi-monopoly, is owned by a monopolist that is part of the reason other websites are being bombarded by requests (aka, they are part of the problem), and you are sitting here arguing that more people should join the monopoly because of an issue they created.
Can you see the flaws in reasoning in your statements?
OK I understand your concerns better. Thank you for explaining.
I am less concerned and don’t have such a negative relationship with crypto. As long as it’s not the selling point of something and decoupled from the actual project or product, that’s fine to me. That others don’t feel the same way is understandable.
For me, radicle is the fastest way to get off of github. All my projects are now there and anybody can contribute without signing up to yet another website i.e they don’t need to have a login for each individual forgejo or gitlab instance. One radicle identity is all you need to contribute to a radicle project on any seed node.
If (when?) forgejo finally gets federation, I’d be more open to using it, but at the moment, it barely provides an advantage over radicle.
The Gemini.com article looks like AI slop to me, honestly.
In lieu of traditional client-server architecture, Radicle Link uses a Directed Acyclic Graph (DAG) as the core of its P2P network, a distributed ledger technology similar to blockchain that excels in speed and scalability.
DAGs are a distributed ledger? Wat?
Also if you actually looked at the code of radicle, you wouldn’t find rad tokens, erc-20, or whatever else. If you further looked at the protocols you’d see that they aren’t using a blockchain. Repository ownership is not handled by smart contracts either - it’s all public key cryptography, which (again) is not crypto in the sense you’re talking about.
To be fair, the article is old and describing radicle version 2. You can find the code here, but I can’t find ERC tokens or anything like that in there, which further makes me think the authors of the article are very confused, AI, or misrepresenting the project on purpose. Of course, it’s possible that all references to crypto were removed from the archive, but it would be good to provide a link to that if you found it.
$RAD is the native token of the Radworks Network, used as the primary means to coordinate all actors, govern the treasury, and (later this year will) reward infrastructure providers on top of the Radicle network.
This I didn’t know of. But I’m curious how that will be done. It is not proof of crypto being within the radicle protocol or codebase (because it isn’t, I looked - maybe I missed it, but I’d like proof thereof). It might be put in there in the future but I’m pretty sure they know it would piss off people to do that.
My guess is that theyll do it like IPFS, which I don’t think has crypto with the protocol but has filecoin on top to reward people who pin things in IPFS. But IPFS users can completely ignore filecoin and aren’t required to use it.
Hopefully this also means monetary investment in open source, not just open source usage without a support contract or contributing back. Matrix is a great example of an open source project that is being used by governments but struggling to get paid because governments are employing their own support staff and making internal forks.
But the more governments, agencies and individuals switch, there greater the chance they’ll pay the developers and maintainers for support or features.
Anti Commercial-AI license