Follow

As I'm working on fedi.builders I was wondering what is more important to you for picking a project to potential contribute to:

Wow, already 100 votes 😮🤯

Thank you!
I didn't expect so much response.

The projects listed on the website are derived from a delightful list (check delightful.club if you haven't already).

As such, not all feedback can make it into the final form as it stands now.

I'll discuss this with the maintainer to see how to handle it going forward.

@RyunoKi It must do something I consider important.

If it’s just for fun: first language (Scheme), then license (AGPL or GPL).

@RyunoKi (considering that it’s a free license — if your poll allows for proprietary licenses: I have no interest in contributing to those projects and the license is the first priority. Got burned once too often)

@RyunoKi "Will my PR be considered?"

Some projects just ignore new contributors PRs and only work on those that are submitted by other maintainers. If there's a good track record of getting new contributors PRs merged and I do want to contribute there won't be a lot of factors that could stop me.

@Gusted @RyunoKi Exactly this. A good way to find out is often to look at open PRs. If there are hundreds, that's a big red flag. If many of them have been open for months or years, I assume the maintainers probably won't be interested in my contribution either.

@RyunoKi If it has a CLA then I will only contribute fixes to problems I personally face, to make my life easier. Only selfish contributions to those projects.
Rare exceptions apply.

@RyunoKi The climate and culture around the project is always the deciding factor. Gate keeping, arrogance, and cliques are a hard pass.

Good projects are spring boards that share, teach, and support their contributors and communities

@RyunoKi there is a range of aspects that I look at.

Programming language and license are the important selection criteria to start with. But then follows a whole range of other project metrics as well as cultural aspects.

It is a software selection process. Take some library with a concrete function. Select on lang + license you get a shortlist of candidates. How you drill-down to one, depends on your objectives. They determine criteria.

No. of contributors, activity, qualtiy, use, etc. etc.

@RyunoKi Language of course plays a part, in that it's less of a barrier if I know the language in question. Documentation and code structure is also important. The less time I have to spend to understand the codebase the better. Lastly is how much time I have and how much I use and depend on or enjoy the thing in question.

@RyunoKi @humanetech Cultural climate of the project, willingness to accept PRs, how new/naive folks are treated, how design/accessibility/security/privacy, etc. are treated. And is it a true community collab., or tied to a single individual? Is it a toy/PoC or an intentional attempt to scratch a real itch?

Programming language is a narrow litmus test for “can I contribute?” capabilities, but almost doesn’t matter though when asking the more important “is it worth contributing?” question.

@matro @RyunoKi @humanetech

This.

I care about license only in that if it keeps me from using the product the way I want to (or think it should be used), I'll skip the project. And if it looks to corporate or amateurish, that's a red flag as well.

As for language, if it's one I don't know well, that'll be a barrier. If it's a Cool New Language that isn't really appropriate for the task (e.g. a webapp in Rust), that's a red flag.

@RyunoKi
I'm mostly itch-driven so being able to use it and hack on it is the major requirement.
Language can be an excluding factor, but it's not a feature or a something you look for, those can be learned if needed. (Rust/Haskell/JS/PHP is a turn-off for me. zig, java, python and lisps I would also avoid)
Copyright to me is FLOSS-only licences and no CLA.

@lanodan
Itch as in itch.io?
Or as in scratching your own one?

@RyunoKi one of my better experiences contributing got me pondering exactly this. My conclusion is that you can usually tell by looking at the big tracker and seeing how new contributors have been treated in the past. Are they encouraged to contribute even if they aren't experts in the domain, perhaps even mentored? That's a good project. The other end of the spectrum would be projects which have open pull requests that sit around without ever being looked at by the author.

@RyunoKi Many people have already mentioned that the community around a project is a major factor. So some ideas (a bit hard to implement but still): there could be a filter "by types of contributions needed" (requires tracking which projects have i.e. open issues with labels "for begginers", "first contribution" etc), which ones have translation platforms (my first contribution ever was translating diaspora* strings).

@RyunoKi Maybe also Fedi projects can submit to you themselves what they need via a form or an issue to your repo. Some need frontend help like Friendica, others - community development, text, documentation, security audits, design contributions, etc.

@lightone
I want fedi.builders to be a view on fedi projects. It's better to channel ideas into delightful lists or similar.

My vision is something like whatcanidoformozilla.com but for Fediverse.

So I'm stumbling into this direction.

Right now, I am waiting for a PR to be merged that will clean up some erroneous formatting.

@lightone
I was thinking about it.

If it won't be part of that delightful list I'm consuming here, I could query GitHub (which hasn't been given up on by most projects yet) for additional information.

However, this being a static build I'd need to have a cronjob to have somewhat up-to-date information.

Client-side queries would result in rate limits, I suppose.

Might be an idea to build shields.io badges 🤔

@lightone
Especially soft criteria like culture are hard to determine programmatically.

I could look for presence of certain files like CONTRIBUTE.md.

But mentoring etc is next to impossible if not done manually.

@RyunoKi Yes, my idea is hard to automate. Alas, many things in Fedi need manual labour. Even delightful lists are curated manually ;) I think a Fedi board where Fedi maintainers can post about their current needs (logo, translations, text, code contributions) is a nice thing to have. Focused calls for help would be appealing to those who might want to help but hesitate where to start. Can be published via Fedi, using special tag, then aggregated same way as search.flockingbird.social does.

@lightone @RyunoKi

> Even delightful lists are curated manually

I have this long-standing idea to essentially reverse the burden of manual aggregation and only have a Review workflow. A Push rather than Pull Publish/Subscribe like process.

Mentioned the idea in many places, but will transfer to Social Coding forum. It lends itself for federation and has numerous use cases.

This is one of the places: socialhub.activitypub.rocks/t/

(Note: Murmurations only demonstrates the concept)

@RyunoKi Recently, most importantly it should not use discriminatory or non-free development platforms.

Discord? No contribution.
GitHub? Yeah, not so strict, but basically no contribution.

Oh, and: CLA? No contribution. My code is governed by its licence, no more no less.

@Natureshadow
I had an argument with Khan Academy about CLA in their tota11y repo years ago 🙃

@RyunoKi "Does it scratch my itch"

I don't go out looking for new projects to contribute to, but just look at what annoys me most in the software I'm using, and try to change that.

Sign in to participate in the conversation
Layer8 in Space

Welcome to the 8th Layer of Madness

Most topics are related to Linux, Anime, Music, Software and maaaany more