Intended Audience: The people from the Jujutsu Discord and random bypassers.

What happened in the project.

In this year the project moved from martinvonz/jj to the Alphabet suborganization jj-vcs/jj which also led to the logo change away from the Jujube, the previous name of the project, to the two blue jays which were drawn by J.J and got its governance with the first set of Maintainers. We also choose to standardize the change-id: header in the Git object with both GitButler and Gerrit. Which Martin then brought up in the Git mailing list. The standardization of the change-id header also brought many discussions about the UX of divergence in the project which has improved a bit. These discussions aren't done yet since it requires the users to understand that divergence is not a "failure" state and isn't as harmful as conflicts.

The CLA also was a really tiring hot issue this year, in which better communication definitely would've helped.

Also East River Source Control came out of hiding which makes the project have three interested parties, Google, ERSC and Open-Source users.

The project also had a new community leader step up as another one stepped down.

There also was the first convention for the project in Sunnyvale California named "JJ-Con", which was colocated with Git Merge 2025.

My year

A bad start until the summer

Honestly my year wasn't great, since I choose to add my name to the potential "Maintainer" list where I personally already thought that it was unlikely for me to get added (like a 0.01% chance). Since I was the only person who wasn't chosen it didn't feel great to hear about it on the Jan 1st post without receiving a message beforehand. I also stepped into a discussion with Drew Devault because it was eerily similar to what I'd do or already did in the past. While I also had some plans to give the project docs a better structure, it never happened because I lost my job in February. At the same time I also asked someone for their prototype of integrating a scripting language with jj so I could work on it. I also continued my work on the Topics design doc before burning out on it. At various points early in the year, I've had the feeling of my opinion not being seen which resulted in the horrible message I sent in Spring. I think this could've been entirely prevented by the project having active moderation, not just people sitting on the sideline. While the conflict resolution was quite fast, it affected me negatively for longer which obviously isn't great¹. It almost resulted in me quitting the project entirely. The whole situation forced me to recognize what I achieved in the project to get myself back on my feet, it also took some time until I was comfortable expressing myself in the Discord again². This year I lost multiple nights of sleep to the project, by just having my name mentioned in certain ways and I had various moments where I was questioning my motivations in the project, since it had such negative effects on me. In July I was back to "normal", if there's such a thing.

JJ-Con and Git Merge

And when the CFP for JJ-Con went up in July, I just decided to write a cool title for a side-project I wanted to discuss anyway. I was very baffled to see that it was the most liked talked during the first three days of voting and it was even stranger to me that it held its top five position which allowed me to hold the talk³. JJ-Con and the all the events around it were a bunch of firsts for me, I gave a technical talk, held a discussion and upon the suggestion of Manish held a unconference session which was related to the current governance. Also Git Merge was smaller than I expected and quite nice. I also got to talk to the Sapling developers which were interested in the scripting language I proposed for jj. There I also was asked if I were interested in working in Source Control which was quite interesting⁴. It also was the first time I've heard that I stand up for the community, which never crossed my mind since I only represent myself and try my best to consider what the community wants.

To my surprise I even could attend the Git contributor summit which was an interesting experience of its own where I tried to push the change-id header discussion a bit further.

My pain points with the current Governance

The current governance is doing its job poorly with these points for me:

• It fails to be any kind of organization, instead it's a bunch of individuals.
• There's still not any kind of leadership, be it in a social or technical manner. Deferring to the community is not leadership for people which should deem themselves as the "core project".
• It is not transparent at all with its decision making with all the leading steps upon its creation.
• It also immediately failed to write down the actual job of a Maintainer presenting some vague ideas which only the people involved in the temporary Governance understand (this excludes a high amount of the community).
• It on one hand wants to steer the technical and organizational direction of the project while it hasn't proven it being capable of any of these.

I also don't find it good at all that a majority of the Maintainers can't take the time necessary for the project only retaining their voting position, if it comes to that. I mean a Maintainer should be active in the project and concerned about its future and not constantly complain about the length of the PR queue, since they're responsible for the development speed of the project. I also find it bad capacity planning to make 2/N maintainers responsible for a vast majority of code-reviews, since it will create a single point of failure when one of them has to disappear for internal needs at times.

Most of the above make my opinion of the Governance lacking, I mean it is OK but it definitely could improve and make their position clearer. If I were Google I'd wait another year to see how it improves and what pans out.

Closing

At this point it has been three years since I joined the project wanting to write jj run. During this time I've answered many questions about the project, tried to implement some processes for it and even had a plan on how a native future could look. I even recognized that the project has no leadership figure and thus drove some decisions even though I was asked to step back from this. At this point I've worn many hats, some of them making time for own code contributions harder, this by sometimes shielding people from questions and otherwise. I want to write code too.

It also seems that multiple people in the project think that I'm a Googler working on this full-time as a lead, which is both an insult and a compliment. I have my own reasons to be here, such as making the project a success and having an actual Jujutsu-native future, not just as a better Git client.

I also have thought about enabling GitHub sponsors to give the community a way of sponsoring my presence on the project but I find that soliticing money for something I already do extremely hard. Maybe someone could convince me that it is a good idea, since my involvement currently doesn't pay for my rent.

I also could try to get into the Maintainer position next year but I doubt it will happen even though it would validate all my efforts across the years for the project.

At this point I've also have become very interested in making Meta a third party involved in the project since someone talked about their use-case in the Discord and I always said that I want to write a jj on EdenFS. But that will be in a separate article and which I promised to at least one person.

My last such post ended with my plans for next year, but since I was unable to implement them this year, I'll just keep on working on my publicly known projects, which means at some point pulling out the Topics design doc out of the freezer again.