This event is aimed at all skill levels, so it is a great opportunity to learn Nix, meet developers and maintainers, and become a contributor.
Complete, up-to-date information: NixOS ZHF Hackathon, Sat, May 27, 2023, 9:00 AM | Meetup
Please register your attendance!
On Sunday, we will continue the UX workshop led by Elton Vecchietti (UX Department lead at Modus Create). See the report on the previous event for details.
This time we will have multiple workshop slots, and expect a variety of developers from all over the Nix ecosystem, including @edolstra @zimbatm @roberth @infinisil @ericson2314 @flokli @davhau and many more.
There will be an opportunity to join remotely through Jitsi. Last time we had some technical issues, and are working towards a better streaming experience. As usual with in-person events, due to practical constraints remote participation will be limited to listening in.
Looking forward to meeting you there!
Preparation meeting 2023-05-02
Attendees: @imincik @fricklerhandwerk @flokli @ners @ericson2314
- Elton confirmed attendance, the Sunday slot will be for UX
@ners will update the GitHub page
@fricklerhandwerk: could split Saturday into two parts:
- depict the ecosystem architecture
- sketch changes and develop a roadmap
@ners: could collect sticky notes with problems to work on
- then cluster and break out into working groups accordingly
@fricklerhandwerk: could do the ecosystem architecture exercise as a fallback if nothing clear comes out of the brainstorming
@fricklerhandwerk: other ideas what to spend a day on? should probably focus on sharing knowledge, making decisions, plans
@iminicik likes the whiteboarding architecture idea
@ners: might still turn out not to provide a lot of value to participants
@fricklerhandwerk: worry that the workgroup process may be too dense and stressful given the short time. whiteboarding is a bit more lightweight
- could do a combination: first map out the architecture, then collect issues and sketch possible projects as a targeted outcome
@imincik: yes, does not require too much commitment yet, but allows for a clear statement of interest
- first slot: map out ecosystem architecture collaboratively
- second slot: collect issues, assess which ones need to be helped with, carve out possible projects to focus on in the future
@imincik: agree that underlying issues should be discovered and a discussion started
- have to determine how to capture the discussion results
@ners: what is the minimal set of things we should all agree on to allow each other to do their work?
@fricklerhandwerk: what kind of outcome do we expect?
@imincik: decisions on improving processes can be worthwhile for an in-person workshop
@ners: people who care about particular issues could ask for help, then review progress at the next workshop (i.e. half a year later)
@ners: the challenge is to balance interests between people intersted in low-level vs. high-level issues
@ericson2314: most people are working on small-scale incremental issues, to scratch their own itches
@fricklerhandwerk: the workshop would be a great opportunity to simply get an overview and coordinate at least a bit, and more clearly delineate responsibilities, exposing dependencies while at it
- next time:
- determine a desired outcome of the workshop
- flesh out ideas for report contents:
- publish the architecture as agreed-upon common knowledge
- publish a set of agreed-upon project ideas that would get moral support
- refine the schedule draft
Preparation meeting 2023-05-09
Attendees: @ners @imincik @fricklerhandwerk @infinisil @roberth
@ners: let’s define deliverables
@fricklerhandwerk: two target audiences of the report and other outcomes:
- developers present
- agree on a world view and common goals
- developers we don’t know yet
- help contributors to do work and form teams
- concrete example: @connorbaker expanding on the CUDA maintainers team
@ners: could exchange what the teams need, e.g. docs team
@imincik: from experience with other developer communities, face-to-face meetings were most valuable for discussing pain points and agreeing on how to proceed
@infinisil: Vision and alignment of people, e.g. what to do with nix.dev vs nixos.org
@fricklerhandwerk: Vision should be the outcome, workshop could be breaking down problems to arrive there
- but should be careful with ambitious goals. we first have to clarify the status quo with that kind of group
- minimal goal should be to agree on important issues to prioritize
@fricklerhandwerk: Everybody picks first a cluster of problems to spend resources on. Then some design/planning work happens for those problems.
- Does drawing a picture of the state of the art improve it?
@infinisil: A bit of planning but then just hacking/working on things together, e.g. I could sync with @roberth on the module system, we could discuss RFC’s
@infinisil: Another idea: Create RFCs for ideas that we haven’t had the time for
@ners: First things that come to mind are usually the most important ones
@infinisil: Personally don’t need a diagram, Nixpkgs Architecture is the biggest problem
@fricklerhandwerk: Everybody will have a different most important problem
- We probably want a big picture and see where we can direct efforts to make the biggest positive impact in finite time
@ners: Would resources, engineers help with that?
@imincik: Event is for wider community, even without much Nix knowledge. We haven’t discussed onboarding new people, very different from what we’ve been discussing.
@ners: We invite the wider community, but we don’t have to serve beginners
- hanging out with developers is already a win for beginners
@fricklerhandwerk: We aren’t ready to support a large influx of beginners anyway, should focus on fundamentals
- With long time developers present, we can dig down into superficial problems and figure out why things are messed up
@ners: Nixpkgs Architecture seems worthwhile to me. Improves experience for all
@infinisil: Idea: Write down most important problems first, discuss those, then work on them
- What about Sunday?
@roberth: Should announce it, even if Elton can’t come we can continue the hackathon
- What is the outcome when various group work on problems?
@ners: Goal is to have people stick to some problem for some months, then evaluate it afterwards
@fricklerhandwerk: Write down what we want to do, also a preliminary to get funding and other resources
- we converged on roughly the following process for the workshop:
- collect large-scale issues
- cluster or categorise
- possibly combine that with a collaboratively developed overview of the ecosystem
- that would involve both the software architecture as well as the development process
- we have to figure out how to implement that technically and how to make time for it in the schedule
- form working groups to do some design work on a small subset of those issues
- the expected outcome is an analysis and a set of possible solutions
- ideally we’d also have an implementation plan and estimate of required resources
- depending on how it goes, the groups could write issues to track their progress and insights
- briefly present results
- publish a report that states the collective understanding gained by the group
- this will probably entail some follow-up work to present everything cleanly and in a way that can be understood by people who were not present
There will be two follow-up meetings same time next week and the week after next, simply to write things down so we have a consistent plan.
Preparation meeting 2023-05-16
Attendees: @fricklerhandwerk @ners
Updated the schedule in the meetup event according to prior discussions.
Looking forward to meeting everyone soon!
Sunday afternoon will have great weather. Bring swimming trunks, there is a lake nearby!
I created a Matrix Channel, in case people want to coordinate social events etc. around the event. It’s at #nixos-zhf-23.05-zurich:matrix.org
Reminder: The live stream will start on Saturday and Sunday at 9:00 CEST respectively.
For everyone following this thread, here’s the report: Zürich 23.05 ZHF Hackathon and Workshop report