It would be crucial to have the actual data on the groups representation in the Nix community, e.g. what groups are most/least/not at all represented, and then take action on this data. In the absence of the data it would be difficult to track if the taken measures actually increase the representation of the desired groups or if it’s actually needed. Maybe we already have this data?
We already doing a good job on representing gender minorities, but I don’t have a feeling we doing a good job at representing women for example.
1 Like
A post was merged into an existing topic: Separated disrupted posts from “improving diversity in this community”
I have been a coordinator for LibreOffice in three Outreachy rounds (third one is ongoing at the moment). I’ll write down a few thoughts and I am happy to answer any questions you might have.
Like GSoC, Outreachy comes with the challenge of how to retain the contributors for the long term. For FOSS organisations, this is even more relevant with Outreachy as the org has to use its own funds. It’s always disappointing to see an intern vanish after the project is over, but this is actually typical of internships everywhere, so it helps to adjust one’s attitude and expectations accordingly. Probably the surest way to retain a contributor would be for some company to hire them to work on Nix.
The emphasis of Outreachy is a bit more on learning than GSoC. That doesn’t mean you are forced to select candidates with inadequate skills.
Outreachy is not only about development and you can have projects for docs and design, for example.
Navigating the Outreachy website can be a bit confusing, so you might want to document where to find stuff like the timeline.
The Outreachy applicant contribution period of one month feels short. For GSoC it was 5 weeks this year, but for GSoC there is a certain sense of continuity, so newbies might start preparing really early, like in August of the previous year. You should be prepared for very intense mentoring activity during the contribution period. Declining new mentees due to capacity overload is always an option.
In the busiest week this round I had 44 meetings scheduled (not all of them about Outreachy), with 17 ending up as no-shows. If the thought of constant no-show meetings and ghosting infuriates you, you need to adjust and adapt because that is a fact of life and there is no technical solution to it. Plan your day to minimise disruption and have an alternative task ready to avoid wasted time. Prefer chats over calls as it is rather miserable to sit in an empty call waiting for someone.
Regarding increasing diversity in general, volunteer platforms have been very good for us. I wrote about my mentoring workflow while touching upon the platforms in Sustain OSS forums a couple of years ago.
24 Likes
Thanks for sharing you experience! I’ve been trying to google Outreachy experiences from mentor perspective but wasn’t able to find anything except for hundreds of blog posts of interns which was not what I was looking for. (Google search really is useless these days :/).
2 Likes
8 posts were split to a new topic: Separated disrupted posts from “improving diversity in this community”
Data is nice, but we shouldn’t let that hold up any other efforts. Stuff like Outreachy probably won’t move the needle, as we won’t get like 200 people, but probably more like two.
3 Likes
Thankyou folks. Posts in this thread have been flagged and hidden automatically as a result of community actions, and constructive and specific feedback has been given about what is acceptable and expected in this thread.
These are good examples of community self-moderation, that I want to encourage and endorse, visibly, on behalf of the moderation team. That’s not something we do often.
Unfortunately, it does derail the thread. Official warnings sanctions have been given. I’ve moved them elsewhere and hopefully we can get back on topic from here on.
14 Likes
Sorry I am not familiar about the normal distribution but what is the appropriate target percentage number so that the problem is considered solved?
It may determine how many effort we should take to solve this problem.
I don’t think we should consider a target % to be our goal because it really isn’t. If person X is unwelcoming to person Y of some community Foo, person X’s behaviour is not gonna change simply because the size of Foo increased from 1% to 20%.
What we need here, as @piegames mentioned early on in this thread, is “to put tangible actions behind our words.”
And that’s limited to folks who even say good words upfront (not a lot). We need people to be more professional and considerate when engaging in any conversation with anyone. In an ideal world, we would all practice de-escalation.
This is an issue that goes beyond the nixos community but that shouldn’t diminish our efforts to at least resolve it within. I think if everyone (on all sides of the debates) took this year’s debacle as a wake up call (which we are – this very thread being an example of it) and showed some humility, we’d be on the right path.
I can see how this could’ve sounded general and not pertaining to the “diversity” bit of the thread, but I believe not being professional is the real problem here, either directly or indirectly.
7 Likes
At the risk of veering offtopic (apologies), I’d like to offer a perspective here:
If someone proposes “Improving our CI infrastructure”, nobody asks “what is the appropriate target level of CI infrastructure so that the problem is considered solved?”. We welcome someone improving our CI infrastructure - even/especially if that makes it better than the CI infrastructure in other projects.
While I get that not everyone feels equally strongly about improving diversity, let’s at the very least support those who want to make things better. It’s not a zero-sum game.
(to respect the topic of the thread, I’ll happily talk more about this in DM, but will resist posting more here unless it’s actually about improving things 
)
21 Likes
Sorry I am not working in IT field, but in my opinion there are always improvements in CI infrastructure, because the clear ultimate target/goal for it is to make it faster, more convenient, more automation to save as many developers valuable time on other stuff as possible. The scenario is different because there is no balance between entities in this case.
If I am veering offtopic, sorry about that. I do agree we should focus more on actual improvements than the goal/target.
I will get at least one Summer of Nix participant via Outreachy in 2025. Subscribed to their mailing list, according to their general timeline we should expect them to open the “call for mentoring communities” in January, where I will register the NixOS Foundation and allocate time for the paperwork, and ensure mentoring capacity.
Our own call for applications still needs to be prepared, but if you’re a Nixpkgs maintainer and particularly interested in mentoring, please contact me.
18 Likes