How Radicle Works Under the Hood
radicle.xyzRadicle shows a lot of promise. As far as I understand it, every participant runs a local radicle node in true peer-to-peer fashion. Every node can seed git repositories which then get replicated on the network. From the perspective of a git command line user, radicle acts as a git-remote that is running on your local machine. The repositories that you push to are then pulled by any other radicle nodes who want a copy. These other nodes will then also seed your repository.
I know that there is also a web-frontend (kinda like GitHub) for radicle: https://app.radicle.xyz/nodes/seed.radicle.xyz/rad:z3gqcJUoA.... But I haven't seen a guide how to set that web frontend up locally.
There's a manual way to run the UI locally for now: https://app.radicle.xyz/nodes/ash.radicle.garden/rad:z4V1sjr...
We'll be also working on packaging the js bundle so that it can be distributed and served locally without setting up a dev environment in the future.
One interesting thing you can do is you can actually point the hosted web interface to your local node, as long as you're running rad-httpd.
https://app.radicle.xyz/nodes/127.0.0.1:8080/rad:z3gqcJUoA1n...
Recent HN post about Radicle: https://news.ycombinator.com/item?id=39600810
Looks like 1.0.0-rc.1 was just released yesterday.
Is it something that's aimed at a specific problem I have never encountered or is it something that can improve everyone's workflow? It feels like it's something closer to git + mailing lists than typical PR workflow, which I do find interesting because PRs are... a lot of unnecessary work. But this has to be better than mailing lists, right?
It seems to address data centralization issues. Government activists and pirates will likely find it attractive.
For the average GitHub user, it adds a lot of complexity for undetermined benefit. It basically ensures your repositories can’t get pulled if someone at Microsoft doesn’t like what you’re doing.
Yes, some of us think that entrusting one’s life’s work to platforms that lock in your identity and collaboration data is not a great idea.
I remember chuckling at those dramatic ads before movies where they tried to equate copying of digital artefacts with heavy-duty "piracy", thinking to myself: who could buy that. Robbing a house isn't the same as sharing a copy of an mp3 with a friend.
I'm not certain you're using it in a pejorative manner, or just neutrally, but in any case, it's funny to see on here.
Something like Suyu, or Great Firewall evasion. PRs (http/ssh in general) and mailing lists will end up being a game of whackamole.
I don’t think I understand how this protocol is supposed to solve the problems of the commons that arise in any peer to peer app, such as how to deal with malicious or misbehaving peers, illegal or unsavory content, and so on. Every P2P app needs to put those problems front and center in their design and I never feel the proposed solutions are adequate. But in this case, I simply don’t see them acknowledged at all.
I'm still trying to get a deeper understanding of Radicle, but let me try to answer with what my current understanding is:
> how to deal with malicious or misbehaving peers
Radicle makes heavy use of public keys and signed data. Every user and every repository has an ID that is tied to some cryptographic key.
> illegal or unsavory content
Each peer can choose what to seed/mirror. Sidenote: The docs describe the number of seeders as an alternative metric for GitHub stars. The Radicle project also has a peer that tries to seed all repositories in order to promote the protocol.
@binary132: Are there some specific problems that you want to know how Radicle addresses them?
Ok; maybe it has a good set of problems it solves for and clearly documented solutions to them. I just didn’t see a front-and-center treatment of them, or really any. Maybe that’s just me, but I really think any p2p platform needs to be clear to its users about these things. I know I definitely haven’t thought comprehensively of what I might need to be aware of.
What is it with web3 sites and their obsession with marquees? https://radicle.xyz/faq
It's actually a web1 thing: https://developer.mozilla.org/en-US/docs/Web/HTML/Element/ma...
Right well I meant that they seem to be making a comeback after web2 realized they were a bad idea and shunned them.
Is marquee a design technique on the front end?