Some other fediverse software like hubzilla and sharkey let you migrate posts, so I wouldn't say it'll never happen. I don't think anyone is working on it though, so probably not anytime soon.

I feel like it's really far from being open. Besides the training data not being open, the more popular ones like llama and stable diffusion have these weird source available licenses with anti-competitive clauses, user count limits, or arbitrary morality clauses.

They probably argue that rescuing or even interacting with sick animals can spread disease and is therefore bioterrorism. If you stretch the definition enough almost anything can count.

No. It's got a "source available" license allowing only non-commercial use, and revokes the license for anyone who tries to sue them.

I think they switched to usually using bing results last year. Their support site mentions they use both backends. I'd guess which one you get depends on which API is cheaper for each country.

I think that's still closed, just poorly done in a way that isn't very accessible.

That's what burned in means.

I added an extra line break, but it already looked fine in the default webview and in Jerboa. Normally lists don't need line breaks around them.

For anyone who wants to know the difference between these terms:

• subtitles - just includes the dialogue
• captions - also includes description of other sounds
• closed - text is stored separately from the video, and can turn on and off while watching
• open - text is part of the video image itself

Why?! The whole point of federation is to let people join communities even when they don't have an account in the same server.

For people who've used lemmy or the rest of the fediverse yes, but most people don't know that yet. If someone shares a post from your site with their friends or a facebook group, they're not going to look into how lemmy works to sign up elsewhere.

1. people that are looking for a community in a niche interest, do not find it, and go back to Reddit.
2. people that are in a big instance and create (or sometimes, recreate) a community for a popular topic. This happens quite often and not because they were not satisfied with the existing communities, but just because they could not find them.

The idea of having topic-specific instances is an attempt to mitigate issue #2.

I'd prefer it if topic specific instances were more popular too. I just think that letting people making accounts tied to their favorite topics would get more people interested in joining them.

I feel a technical solution like federation pulling in lists of communities with would help more with discoverability.

Not my experience. A few examples:

• No one complained about the mods from !linux@lemmy.ml, yet I've witnessed endless discussions about moving away from lemmy.ml.

I'm not sure how that goes against what I said. That's mostly people disliking the admins.

• Beehaw defederated from LW, so this forced users of these instances to "choose" between the communities and/or create accounts on both of them if they wanted to keep following the whole conversation.

Similar issues could happen even if users are separate from the communities. Beehaw could defederate your instances, and lemmy world could defederate programming dev or something, and people would need other accounts if they want to see everything.

• Personally, I do not want to join or participate extensively in communities that are on LW if we have a topic-specific instance for it. I know that I am not the only one.

Me too. I usually avoid lemmy world communities unless there isn't an active community elsewhere.

New users to lemmy usually aren't going to join communities if they can't register there. And people who are really invested in a topic will want to have that domain for their account. You're cutting off a lot of the users that would grow your communities.

I don't mind the idea of a collective to handle a bunch of instances, but I feel like you're going about it the wrong way. When the same person make a bunch of instances about a variety of topics, it looks as if they aren't that invested in any specific community. From my experience, the most active communities start off with a few people who care almost obsessively about that topic.

Also the idea that communities can be 'neutral ground' doesn't make sense to me. People will leave or join based on how the admins and mods run them, whether or not the users are hosted there. In some situations it might work out fine, but if anyone thinks it's caused by how you're running your sites, they may defederate from the whole collection.

It makes sense that if you're designing a language, you'd like the language you made and would want to use it. It's fine for compilers like that to exist, and even be the main one used, but ideally it shouldn't be the only compiler.

But there are technically ways to bootstrap a language without writing it in another language (other than a small core in assembly or something). You could design a tiny compiler that only handles a small subset of your language, then write a better compiler using only the features available in that subset. You can do this for several layers of compilers until you have the full language.

That's already how it is now, we just don't usually think of it that way. You can't compile rust unless you already have a rust compiler. The current version was compiled in a previous version, which was compiled in a previous version, going through a chain of older versions and other languages. Anything along that chain could've theoretically had an influence on the current compiler.

It's not about the code itself being more trustworthy. The point is that when you bootstrap, you don't have to blindly trust any of the binaries, since it's source code the whole way down. Someone could bootstrap rustc like this, compare it to the binaries that already exist, and ideally they would be identical.

A lot of this bootstrapping stuff comes back to the 'trusting trust' attack. You could write a compiler that adds some malicious code to programs it compiles. But the thing is, it could also insert it's own malicious code when compiling a compiler. So you look at your code, and the code of your compiler, and everything looks fine, but the exploit is still there. Someone wrote an example in rust.

Theoretically there could also be bugs that propagate this way. So to fully trust your code is doing what you think it is, you'd need a chain of compilers back to hand coded assembly.

Yeah, I wrote the wrong language. I tend to lump those together in my head as 'big multi-paradigm languages I haven't bothered to learn yet.'

You can technically do it, but it's a convoluted path. The article talks about it. Basically to bootstrap that way you need to go through a lot of versions of rust, compile rust 0.7 in ocaml, compile ocaml in scheme, and compile scheme in C using gcc. For gcc you need to compile a chain of versions back to when it was written in C instead of C++, plus the whole TinyCC bootstrapping path.

edit: had listed scala instead of ocaml

The main thing is that TinyCC has already been bootstrapped.

Check out this page on bootstrappable.org. Basically they start with a 200 something byte binary (hex0) that can act as an assembler, then using a bunch of layers of tools and compilers you can bootstrap a whole system. I think they use stage0 to build M2-Planet, use that to build GNU Mes, and use that to build TinyCC.

So a project like this fits neatly into that bootstrapping path. It could be done other ways, but starting from a fairly complete C compiler makes it a lot easier than building an entire path from scratch.

The biggest thing is probably non-destructive editing, so you can do stuff like apply filters without them changing the underlying image. Gtk3 should add better support for tablets and wayland. There's also better layer tools and font support. A lot of it was on the backend, which should eventually allow for using other color spaces like cmyk natively.

It's too bad that GLIMPSE fork never took off.

They've been working on porting it since back in 2012, and didn't want to redo a bunch of the porting work before they even released it.

There's lem.el for Emacs.

There was another one but it doesn't work anymore. It hasn't been updated in 3 years.