@vrighter @ylai
That is a really bad analogy. If the "compilation" takes 6 months on a farm of 1000 GPUs and the results are random, then the dataset is basically worthless compared to the model. Datasets are easily available, always were, but if someone invests the effort in the training, then they don't want to let others use the model as open-source. Which is why we want open-source models. But not "openwashed" where they call it "open" for non-commercial, no modifications, no redistribution
If you find a piece of code in a forum without any license text, and you use it in your software, you could be in a lot of trouble, and not just because the code might be bad. Code without a license is NOT open source, nor public domain, nor free to use, https://opensource.stackexchange.com/a/1721. It needs to have a license that explicitly allows use, modification, redistribution, only then it is open source. You may have seen some "openwashing", someone trying to redefine the term to make them look good
@taladar Emphasis on "entire system".
@foosel Saying that I demand maintainers to cater to my requests can be easily disproven by just looking at my words above where I say the exact opposite. Then who is doing their best to misunderstand and turn words around?
@taladar Discussed in other threads here - forgejo.org is implementing forgefed which will do this, it's a work in progress, monthly reports here https://forgejo.org/tag/report/
@foosel So you want to continue sacrificing yourself? Your choice 🤷♂️
Now you are back to believing in open-source, so let's stop sending users to walled gardens, shall we?
@foosel But why do you feel like you "have to" do those things? Are you paid for it? Are you trying to sell the project? Are you looking for VC funding? Is someone threatening you if you stop fighting those fights? Those are all things from the commercial mindset, or things exploited by Jia Tan. Of course everybody likes when a project is maintained, good quality, free, but that should come from the cooperation and from the freedoms in the license and platform, not from your personal sacrifice
@mormund It's not about the privacy of the code, but the privacy of the users clicking on github and then reading some news. They aggregate behavioral data about you.
> the only thing that can be lost are issues and pull request histories
"Only"?? That's a HUGE problem. That's exactly one of the walls keeping people inside github. Git protocol could distribute that, but it doesn't suit the commercial platform's interests -> go to open platforms instead.
@foosel If that is the case, then how did the choice of using an open-source license even get through? It sounds like you are confusing commercial thinking (we have to get more users, we have to be where the users are, we have to support them, we have to meet the KPIs...) with the open-source. You don't have to do any of those.
@tyler Also note how you went from "we want projects with users" to "oh it's so hard to provide services to so many users".. at least stick to your argument. One thing is for sure - actively keeping users away from open platforms is not going to increase the users on these open platforms. Doesn't take a genius to figure that out. Do what you want, I'm just pointing out that you seem to be working against yourself.
@tyler So why are you doing open source anyway, if not for the philosophy? You are completely undermining that by forcing your contributors to stick to proprietary walled gardens. Last time I checked there were hosting providers for both gitea and matrix.
@OsrsNeedsF2P But that's the problem we need to fix, not the reason to give up. There will be more people on Gitea and Matrix if you try. There is also more people on Reddit and Twitter, yet here we are.
@Templa I'm actually using github.com/coder/code-server for that. It's also built only from the open-source parts of vscode, but it is made to run in browser. So I just deploy the docker container with code-server to the more powerful remote machine, and open my browser, where it can be used as PWA so it's almost unrecognizable from a native desktop app.
@pelespirit Not complicated at all. Llama is simply not open-source, the text of the license is in direct conflict with the meaning of open-source and any definition of that term.
in that case you are doing the whole "no-coding" all wrong. If you really need something that cannot be done by connecting the nodes and grouping them in flows, then instead of developing new function nodes, develop your own contrib packages. But there are a lot of existing packages, so maybe you don't need even that.
It all depends on how much you treasure that 2TB of data. If you cannot afford to lose it, then you don't really have any other option. All I'm saying is that self-hosting and backups go hand in hand. Self-hosting without backups is just asking for trouble. But what you are talking about sounds more like 2TB of old photos, that's not really self-hosting. Just get a second 2TB drive and copy it there. This thread started with talks about password managers and their backups - those aren't 2TB.
The federation between mastodon and lemmy is strange. If a M account wants to follow a L community, they need to follow an automated M account which represents the L community. But if any M post mentions that L community, the post will get boosted by the community's M account, so everybody who follows will get a notification. And I'm not sure if this can be moderated from the L side, because it seems like it never goes through L. Such as - do you see this @opensource ? Does a L mod see this?
