Well, we want to create a new programming experience not a duplicate of languages that already exist. A language is more than syntax it should consider the entire experience of programming. We are for example doing content addressed code which means each unique expression, (somewhat) disregarding variable names, has its own hash address. Commutative patches like in Pijul is a nice bonus for this.
1 Like
The IPFS/Blockchain model is a bit at odds with a CRDT like Pijul, Git is more suited to that model. But I’m working on something more efficient.
I believe Nixpkgs can already be imported into Pijul. There is one major improvement coming up soon, related to the size of repositories on disk, these could get large with repositories of that size. My current prototypes for baby implementations of that are quite encouraging, but this is a large change touching many parts of Pijul.
1 Like
IPFS works independently of blockchain. It is just a content addressed file sharing network. So it should be just as suitable for CRDTs if not the most suitable. The point is that you address things by their hashes/content and not any particular server or folder structure. That way you get a decentralized distributed system right out of the box.
nixpkgs can be imported, but nixpkgs does not yet support loading content from a pijul server. This shouldn’t be too hard to implement. I think it is much more suitable as a standard VCS than git for nixpkgs because of the optimizations and comutability.
When do you plan to open source nest and pijul servers?
I’m also considering trying to implement Pijul in Yatima/Lean for fun.
I started a bit here Anderssorby/Pijul.lean
Note that Lean4 is unstable and most of this software is experimental.
’ ')/
I’m Alex, but most know me online as ‘Dr. Bluefall’.
I tinker with multiple languages, but my mainstays are C/C++, Rust, and Python.
As of right now, I’m working on a loadout calculation library for Splatoon 2 (and eventually Splatoon 3) called libwoomy, which I intend to use Pijul to track (as soon as I can resonably comprehend it, because a lot of my experience is with git).
Hello!
I’m Victor (or multun), a young software engineer working at SNCF. I make a railway infrastructure editor and simulator as a dayjob.
I love history, programming, algorithms, low level systems and software architecture.
I’d love to fix a few bugs and implement branch merging on pijul!
2 Likes
Hello, Daniel here. Working professionally with Python, C++ and Rust on Windows. Found out about the project from Can We Please Move Past Git? (22 Feb 2021), which is just horrible post content, but it got me here.
I’m interested in and support:
- organization: For example, see RustConf 2022 - Your Open Source Repo Needs A Project Manager by Alice Cecile .
- documentation: Not having it means explaining things over and over, which is a systematic bug that should be fixed upstream by just, well, documenting things.
- pragmatism: Software and processes should be about solving actual problems that people have, going from most frequent ones to less frequent ones. There should be a way to gather such data, for example listening to the users when they complain.
Some questions from my side:
- does this project have RFCs?
- how many projects are there? Pijul itself, Nest, anything else?
- is there a project board for overview of what is currently being worked on?
- is there a backlog?
- where do I open a new issue?
- is there a structure of a team?
- who is responsible for what?
- who is doing triage and when?
- is Pijul oriented towards users?
I really like that Pijul has discourse forum. 
1 Like
Interesting questions. Pijul started with two theorists (Florent Becker and myself), and currently has had one very active contributor (me). I’ve also written almost all of Sanakirja (the database backend for Pijul, but also a project on its own) and almost all of Thrussh (an SSH library used in Pijul and the Nest). And the Nest.
So, no RFCs, no project board, the backlog and new issues are at pijul/pijul - Discussions. Everything has been quite experimental so far, but I do want to change that at some point. There’s a chicken and egg problem: without enough users and contributors, you don’t need structure. Without structure, you won’t get many contributors. Some contributors have found their way in the code without help though, I’m thinking of @Rohan in particular. Others have invented creative ways of testing Pijul and Sanakirja (tankf33der on Zulip/the Nest), without any guidance at all.
So, the “team” isn’t very structured. One thing that is hard to realise for newcomers is that there have been lots of discussions since we started back in 2015, and many decisions were made after discussing issues thoroughly. Many others were just dictated by the theory, we can’t change mathematical truth, some things are impossible to do.
The project has a number of goals we don’t really want to give up on, one of them is to use only rigorous algorithms, meaning we won’t introduce “hacks that work most of the time”. Another one is inclusiveness, as we try to be welcoming to diverse users.
I like what we’ve achieved so far in terms of community. People from all over the world are using Pijul, not a huge crowd but not insignificant either. We’ve successfully avoided haters in the community, at least here and on our Zulip. I don’t know how, they seem to be everywhere on Reddit/Lobsters/Twitter/blogs/… every time the topic Pijul comes up.
Edit: I do have projects for Pijul in the next few months. I’m still undecided between different directions, but I’ll have news about that soon, at the moment my main job is keeping me way too busy.
1 Like
Hi! I’m Ryan, I tinker around with Rust on various personal projects, although lately I’ve been learning OCaml too. I’m interested in exploring the potential use cases for pijul and libpijul in some of my hobby projects. My daily driver is FreeBSD so maybe I can also help out with testing and contributions in that area as well! 
Hello there…
I am a humble hacker who loves to code, and boy, I have coded for almost 30 years. No one really tackled the real problem of source for all that time! Back then, the computers were slow, and screens were just too small…
Now I work for a big tech corp. They killed git recently. They deprecated an API and did not port the library by in the company-flavor-git client. They “officially” replaced it with hg client. A bunch of fussy college graduates! Thankfully, the git client was just a frontend for a Perforce-derived Planetary version control system… so you know… Maybe it’s time to experiment a little.
I knew Darcs back in 2003 or something. It was not a happy hacker using it… Then Git came out, and I invested years into building understanding and experience with it. I call myself a mid-level Git-fu master. No food, no water, no light for 24 hours and I will solve the worst octomerge. But I am happier, but not happy.
I have ideas, but first I should read this forum, issue tracker, and manual…
3 Likes
My name is Blaise Pabon (mais pas mauvais…)
I’m actually not French, but that’s another story.
I stumbled on pijul last year when I happened to be working with zuul-ci and Gerrit.
I like that it facilitates trunk based development without some of the complexity of git.
Ironically, I am badass with git, but I recognize that guy is mysterious for many people.
My favorite environment is Python and I’m a big proponent of open source communities.
My working languages are English, Spanish and French.
Japanese, Portuguese, Sanskrit and Swedish are for recreational purposes.
1 Like
I think if pijul makes it easier to do trunk-based development, then it will get interest from projects influenced by DORA
But we would actually have to show that process in action.
The other killer feature would be to show multi-repo patches, similar to what people do with zuul-ci.
Hello all, my name is Brandon. I am playing with elixir, go and python right now. I have enjoyed learning and applying git professionally and (even more so) personally. I am hoping to immerse myself in pijul by converting my obsidian vault to use it instead of git. Check out my progress on nest
You can find me on gitlab/github I currently work in local municipal government supporting a public transportation department. I work with service planners, marketing folks, and other IT people who do not want to learn git, but marvel at what I can accomplish ( which I attribute in large part to the consistent repetition I get when working/experimenting with changes using git). I am really hopeful that pijul will bring better tooling for domains outside of software development.
Hello there,
My name is Paul, I saw your talk at FOSDEM on sunday @pmeunier, and now I am eager to contribute to pijul or at least its ecosystem 
.
While I would not describe myself as a VCS and datastructure expert, I have a little experience with git high level APIs. I maintain a version manager called cocogitto, worked on an unfinished (and now abandoned) git/activitypub forge project called gill and made a few minor contribution to gitoxide.
I am really enthusiast about tooling in general and I think it would be fun to start an alternative forge projet for pijul. Would that be a ok though ?
1 Like
Hi. I’m currently working on an update to the VSCode extension for pijul.