With the 2023 crowd-funding in documentation funding, we’ve seen some improvements, but a lot more is needed to make significant progress towards improving the overall experience with Nix documentation, including onboarding and discovery. Especially newcomers frequently struggle with our documentation, get frustrated and notice gaps, and have a hard time adopting Nix.
In addition, in the Nix community survey 2025 we see that most people want excess funding to be allocated to improving documentation. While the NixOS Foundation doesn’t have that much excess funding, @hsjobeki has stepped up to pick up leadership of the documentation team with support by @friedow. While they’re volunteering their own time to lead the team, they’d like to have access to funding for not just ensuring that stuff gets done, but also maintained over time.
(8.000 €) Continuous deployment & Reduce feedback times
(8.000 €) Navigation & redirects
If a company is interested in providing a significant amount of funding and would like a contract and invoice, please reach out to foundation@nixos.org.
Furthermore, if you’d like to support documentation in non-monetary ways, feel free to join the Matrix channel and/or bi-weekly meetings (see “Nix documentation meeting” in the NixOS Calendar), or see where you can help from updates in the discourse category.
Is the filled portion of the horizontal bar the estimated monthly budget, or today’s balance ?
According to the “Estimated annual budget” mouse hover, the recuring monthly donations are €113.00 which seems much less than even the first goal, which is more than overtaken by the filled portion of the bar
The estimated yearly budget however is of €2,412.07 which looks in-line with the progress of the bar if we assume the ammounts are per year and not per month
I’d appreciate if we didn’t have to include inflammatory meme account posts in the official thread here, but:
There’s more details about the goals on the OC page. This one notably also mentions " including making the new content visible in https://docs.nixos.org", which means integrating it into infra, which is arguably going to be the biggest part. Docs team lead @hsjobeki probably thought about this in more detail and can update the goals to be clearer if needed
These are just rough estimates for now so we have something to aim for, estimates are never accurate and won’t be here either. We might be able to make it work with less, in which case we can use the surplus for future goals, or we might need more. In general it’s better to over- than underestimate.
The foundation generally sets hourly rates with a total limit for such work, so we don’t intend to pay the full amount unless it’s actually justified. Yes there’s tradeoffs and it can be gamed, a balance of trust and supervision is necessary.
We can and should also adjust these goals as needed, so if there’s good feedback as to why the goals are too high/low, would love to hear it. This in itself also takes time however, so let’s be mindful of how much of our volunteer time is worth spending refining estimates, vs. actually helping improve docs.
Something that would help as well is a bit of clarification on the ammounts (also given the OC website does not clarify):
Does the “/month” carry over to every line, or are the others one-time amounts ?
I think putting them in the same numbered list might have create some confusion here
Rendering the manuals as separate pages is trivial. Making that not break tens of thousands of links and bookmarks all over the internet is not. LLMs don’t help with that, because the problem is extremely underspecified. The work is in figuring out the constraints and also the workflows that would allow for split-page and single-page rendering as well as checking for consistency and correctness. And then I expect around half of that budget to be reserved for sticking around to fix issues that will keep popping up after the transition — often these are issues outside of the immediate code one has touched, such as in infrastructure.
Also I don’t expect just any rando to get that money, because since knowing what matters is not something you can wish into existence via prompt, this should (and hopefully will) be put into the hands of someone who got theirs dirty enough with the material already.
