Cool project, I’m always weary by projects that take so much dev time to build out, but I’ve been keeping my eye on this one.
I’ve been thinking about git collaboration slightly differently from the rest of the code forges. I think git collaboration could be easier for a huge chunk of use cases.
I don’t see why we necessarily need code forges for collaboration. We just need something a little better than emailing patches around.
It’s why I’ve been building a pastebin super charged for git collaboration. It’s still very much a wip but for anyone curious: https://pr.pico.sh
I'll get into radicle when they make a properly namespaced p2p alternative to crates.io out of it. I'll make it my religion. If I'm going to have all of these packages on my system at all times, I can continually dedicate 50-100K of my bandwidth to sharing them; everyone could, really.
Rust has a Microsoft dependency in crates.io. We all have a Microsoft dependency with github.
edit: iirc, they're extreme CoC warriors and crypto-connected, so that's going to limit the reach of the project.
Now its been a long time since I last checked at radicle but if they are using checksum as the thing then you can definitely hack through something with ktra which pulls the content from radicle nodes and gives it back I suppose
not sure what you mean by namespaced p2p thing mate, at best maybe we could have something like nostr in an ideal world where we sort things through on time and sort of but that is really finnicky, I have worked on something similar but that would involve crypto which I or you also don't want to involve it seems.
The best idea could be a sort of lookup table that can be operated by rust foundation etc. which just links names to their hashes etc., iirc nix can definitely do something similar with nix flakes in the sense that they also follow the hash based approach but don't take my word for it as I am not sure and nix is also not the point of this discussion.
If this is something that you are really passionate about, maybe you should give it a go! Have a nice day!
Yeah, "make it my religion" is a bit extreme - these are, as you say, software tools. Nonetheless, I do think it is bad that a lot of digital infrastructure for free/open source software, such as crates.io, is dependent on centralized services run by large corporations such as Microsoft-owned Github. It is important for free software hackers to develop free alternatives to this infrastructure that don't rely on e.g. having to have an account on a service wholly owned and run by Microsoft. Radicle is a really neat software project precisely because it is an attempt at building something that does the same job as Github, but without having to use proprietary software hosted by Microsoft.
WSL... filesystem? Either way, I firmly disagree, there are not many cases where I've been unable to do dev work on WSL. Only when I need particularly weird / specific networking or hardware (ie. GPU, which might work now) have I had significant problems.
File IO from windows into the wsl disk and vice-versa is significantly slower so it's not great to, for example, use wsl git on a project living under your windows user directory or visual studio on a directory under your wsl home.
I think they're just using FUSE to make it work but don't quote me on that part.
Last I heard, using native git on Windows was slow anyways? Something about how NT handles files. So even with git it was already best to keep things inside WSL and just use ex. VS Code with a remote in WSL, and at that point it doesn't matter if you use radicle in WSL instead of git.
Sure, not doubting that, I'm also a developer, and use Windows. But not because I want to, which I feel is a pretty common position to be in, at least around me.
The motivation you had to move to Linux evaporated when you started regularly using Linux, and it's important to consider Windows developers, for whom it's now easy to use Linux without completely moving to Linux.
Issue tracking support is great. I hope they add web login (maybe through ActivityPub and/or OAuth), and a wiki tab. This would put it on par with Fossil, which is similarly decentralized but has a bespoke version control which makes rewriting history and other git workflows difficult.
Fossil is primarily a DVCS, like git, but aimed at a different set of users (e.g. smaller projects) and includes decentralised tickets etc., but in practice it is easiest to have a centralised server. Its easy to run your own, but there is hosting available for Fossil: https://chiselapp.com/
Redicale seems much more decentralised. Rather than making it easy to run your own server it eliminates the server altogether. A more radical approach :).
it would probably be easier to build something like Radicale on top of Fossil because all that is missing is a way of finding to peers, and connecting to them without a server (e.g. from dynamic IPs, behind NAT etc.).
Cool project, I’m always weary by projects that take so much dev time to build out, but I’ve been keeping my eye on this one.
I’ve been thinking about git collaboration slightly differently from the rest of the code forges. I think git collaboration could be easier for a huge chunk of use cases.
I don’t see why we necessarily need code forges for collaboration. We just need something a little better than emailing patches around.
It’s why I’ve been building a pastebin super charged for git collaboration. It’s still very much a wip but for anyone curious: https://pr.pico.sh
All your work at pico.sh never cease to amaze me!
This seems like one of the most interesting P2P projects out there.
It seems like it could be a really nice backend sync engine for offline first apps, aside from the lack of official support for Android.
(2024)
https://radicle.xyz/
Some various discussions:
Radicle 1.0 – A local-first, P2P alternative to GitHub
https://news.ycombinator.com/item?id=41509713
Jujutsu and Radicle
https://news.ycombinator.com/item?id=44900455
I'll get into radicle when they make a properly namespaced p2p alternative to crates.io out of it. I'll make it my religion. If I'm going to have all of these packages on my system at all times, I can continually dedicate 50-100K of my bandwidth to sharing them; everyone could, really.
Rust has a Microsoft dependency in crates.io. We all have a Microsoft dependency with github.
edit: iirc, they're extreme CoC warriors and crypto-connected, so that's going to limit the reach of the project.
> I'll make it my religion.
Hey mate, these are just tools but I love the energy I suppose.
Pretty sure that you can hack something like this though
Just checked and saw this https://github.com/moriturus/ktra
Now its been a long time since I last checked at radicle but if they are using checksum as the thing then you can definitely hack through something with ktra which pulls the content from radicle nodes and gives it back I suppose
not sure what you mean by namespaced p2p thing mate, at best maybe we could have something like nostr in an ideal world where we sort things through on time and sort of but that is really finnicky, I have worked on something similar but that would involve crypto which I or you also don't want to involve it seems.
The best idea could be a sort of lookup table that can be operated by rust foundation etc. which just links names to their hashes etc., iirc nix can definitely do something similar with nix flakes in the sense that they also follow the hash based approach but don't take my word for it as I am not sure and nix is also not the point of this discussion.
If this is something that you are really passionate about, maybe you should give it a go! Have a nice day!
Yeah, "make it my religion" is a bit extreme - these are, as you say, software tools. Nonetheless, I do think it is bad that a lot of digital infrastructure for free/open source software, such as crates.io, is dependent on centralized services run by large corporations such as Microsoft-owned Github. It is important for free software hackers to develop free alternatives to this infrastructure that don't rely on e.g. having to have an account on a service wholly owned and run by Microsoft. Radicle is a really neat software project precisely because it is an attempt at building something that does the same job as Github, but without having to use proprietary software hosted by Microsoft.
Love radicle! Excellent docs, great UI, easy-to-use cli and it's all written in Rust. Have been running a public node recently: https://app.radicle.xyz/nodes/radicle.ext4.xyz
Windows support yet?
Ah well
Windows is increasingly less of a viable platform for, well, anything. Time to pick a distro.
I don't know what it is about git but it really brings out the "Nuh uh, your use-case is wrong" talking points.
I think you can pick from most of them via wsl2.
I would advise using WSL.
I think they're focusing on developers, most of which are either planning a move to something else, wishing they could move or have already left.
50% of devs use windows.
Personally, what dev related motivation I had to move to Linux disappeared when wsl got decent.
That said, 90% of my work is still in windows proper, so any universal tool like git related needs to support it to be any use.
And if not me, anyone else you're collaborating with.
That feels like a contradiction; either WSL is a good solution and you can just run radicle there, or it isn't.
Good enough in a pinch to prevent a switch but not good enough to do all your work in the wsl filesystem.
WSL... filesystem? Either way, I firmly disagree, there are not many cases where I've been unable to do dev work on WSL. Only when I need particularly weird / specific networking or hardware (ie. GPU, which might work now) have I had significant problems.
File IO from windows into the wsl disk and vice-versa is significantly slower so it's not great to, for example, use wsl git on a project living under your windows user directory or visual studio on a directory under your wsl home.
I think they're just using FUSE to make it work but don't quote me on that part.
Last I heard, using native git on Windows was slow anyways? Something about how NT handles files. So even with git it was already best to keep things inside WSL and just use ex. VS Code with a remote in WSL, and at that point it doesn't matter if you use radicle in WSL instead of git.
Oh that use case isn't great yeah. They probably aren't using FUSE, they notably use 9P a networked filesystem protocol from Plan9.
If you operate fully inside of WSL (either via X11 or in a terminal) it's a pretty good experience.
> 50% of devs use windows.
Sure, not doubting that, I'm also a developer, and use Windows. But not because I want to, which I feel is a pretty common position to be in, at least around me.
The motivation you had to move to Linux evaporated when you started regularly using Linux, and it's important to consider Windows developers, for whom it's now easy to use Linux without completely moving to Linux.
Seem a lot like fossil-scm
Issue tracking support is great. I hope they add web login (maybe through ActivityPub and/or OAuth), and a wiki tab. This would put it on par with Fossil, which is similarly decentralized but has a bespoke version control which makes rewriting history and other git workflows difficult.
Fossil is primarily a DVCS, like git, but aimed at a different set of users (e.g. smaller projects) and includes decentralised tickets etc., but in practice it is easiest to have a centralised server. Its easy to run your own, but there is hosting available for Fossil: https://chiselapp.com/
Redicale seems much more decentralised. Rather than making it easy to run your own server it eliminates the server altogether. A more radical approach :).
it would probably be easier to build something like Radicale on top of Fossil because all that is missing is a way of finding to peers, and connecting to them without a server (e.g. from dynamic IPs, behind NAT etc.).
> Fossil is primarily a DVCS, like git, but aimed at a different set of users (e.g. smaller projects)
Is it actually? I never got the impression SQLite was a smaller project, but I'd confess to not knowing much about how it is organized.
Small in terms of team size, not impact.
SQLite is developed by 3 people. They don't accept outside contributions.