Linux Foundation spends just 3.4% of its money on Linux
lunduke.substack.com
lunduke.substack.com
Project support at over 50% of expenditures (separate from 3.4% of kernel-specific support) is pretty impressive, the rest going to infrastructure (a bit over 12%) and community training and events, with barely over 7% costs.
Also, all the other projects are also pretty much based on top of Linux and thus driving linux-related things, as even Zephyr RTOS is heavily driven as something to use in cooperation with Linux based systems.
> “Linux Kernel Support” is the lowest priority for The Linux Foundation.
First, this implies that dollars spent equals priority. A lower priority project may need more dollars. The test of priority is which things take priority when dollar needs can't be met. It also assumes large donors have no say at all where there donations go and a million other things.
Second, it ignores the fact that when someone says "Linux" they don't automatically mean only the kernel. It's true that some and perhaps most do and it is maybe even the correct view, but that's demonstrably not the view of the Linux Foundation. Even if it were, there's lots of argument to be made for supporting people who may support or work on the kernel now or in the future.
Third, even though he admits he doesn't know how the financials break down, he somehow knows every project receives more than direct Linux kernel support. That's the best interpretation of his claim. The worst is that he's disingenuously lumping all projects together.
Fourth, he ignores the fact that some of these supported Projects themselves support the Linux kernel.
skimming the toc for the annual report, i see loads of relevant linux support efforts that go beyond just the kernel.
i'd buy this argument if actual kernel devs were complaining, not someone trying to build readership.
"The Linux Foundation (LF) is a non-profit technology consortium founded in 2000 as a merger between Open Source Development Labs and the Free Standards Group to standardize Linux, support its growth, and promote its commercial adoption. It also hosts and promotes the collaborative development of open source software projects. It is a major force in promoting diversity and inclusion in both Linux and the wider open source software community." [1]
About 60% of expenditure is "Project Support" (I included Linux project support here) which seems to be all kinds of non-code help to open source projects. I see Legal, Marketing, Community management, Training, Infrastructure, Creative/Design. Here is a description from the website: https://linuxfoundation.org/projects-support-programs/ I grant you that it is vague, but I doubt there is much politics involved.
A big open source project like Kubernetes probably needs all of that help. As a project grows bigger lots of non-code issues grow in importance. I think maintainers and code contributors would like not having to spend too much time dealing with WCAG on the flashy website, checking everyone's contributor agreements, vulnerability scans, paying for CI/CD, writing articles about the project, dealing with trolls in the community and moderation in their forums.
Next is community infrastructure with 12%. No idea what that is but it might be LFX which looks neat: https://lfx.linuxfoundation.org It seems to be a project management tool.
I will lump together Community Tooling, Training and Events in their 20% category. Especially the Training and Events seem to also bring revenue - so students and attendees pay for that. Not sure what Tooling is exactly - could be LFX again.
Then you have the last 8% which is Corporate operations, or their own Legal, Finance, etc. departments.
While I also have issues with the "Diversity Olympics" that most American companies seem to play on, I am re-assured after reading the full report. I also understand that DIE initiatives are "intellectually fashionable" even if the fundamental morality of DIE is questionable, but I can't have too many complaints if the core mission is not put in jeopardy - which I really can't conclude from the report.
I am leaving the original comment up for reference
[0] https://www.linuxfoundation.org/wp-content/uploads/2021_LF_A...
Original comment:
What a travesty. Honestly, I don't even know which organization I can support if even the Linux Foundation isn't contributing its money open source technical excellence.
I guess directly supporting open source developers is one way, but now there is the question of who to fund. Is funding for open source and/or charitable software a lost cause? Do we have to just accept that open-source software is fundamentally a volunteer effort, and will never allow a person to eat or pay rent?
Personally I sponsor projects which I consider somewhat niche/non-mainstream, but which I myself reap great benefits from every day.
It’s probably nowhere near enough to put food on anyone’s table, but at least it’s directed at people I find deserving, for things I care about, and for better or worse, it’s better than nothing.
Project support and community infrastructure is pretty nice, though. And that's 68.8% of the total. But if it's true that it's support for projects like Open3D that undermine Linux, that's not cool.
3.4% of its money goes to Linux kernel development.
This is because they really don't have to hire kernel developers. Other companies do that for them. There are a handful of kernel developers employed by the Linux Foundation, but there doesn't need to be a lot, because there are already enough kernel developers hired by every other major corporation on the planet.
GKH at one point said the entire reason the book Linux Device Drivers hasn't been updated since Linux 2.6 is because they already have enough kernel developers.
If you look at the graphs, 68.8% of their expenditures are on project support and infrastructure, separately from the 3.4% going to the kernel. The LF is responsible for a lot. It's not just the kernel; there are thousands of things in Linux's ecosystem.
I think there's an underlying reason why people are hesitant to donate money to a cause because if even a software foundation predicated on open source and transparency can't get money into the hands of open source projects then publicly funded donations are a really bad bet for making the world a better place.
Linux is a kernel. If they mean the "Linux-Adjacent Open Source Software Foundation", then call it that.
The philosophy of building free and open source software and merit shall grant you monetisation is completely dead to me. There's no way human group dynamic would sustain this if the Linux foundation with their 177 million is enough to cause such a divergence from any semblance of this core idea.
The article starts about how only 3.4% is spent on kernel development, which is a fair complaint. Then the author picks some other stuff totally unrelated to the spending breakout to complain about, including diversity efforts, for which there are no spending breakouts.
Spending on these may be 1/100th of a percent for all we know, but the author claims at the end that despite not having spending breakdowns, that this is a fair description of the Linux Foundation, based on their annual report:
“The Linux Foundation is dedicated to social justice activism, climate change, and vaccine passports… while occasionally using Open Source software.”
Might be. But it still gets 5-10x more attention in the report itself which hints at the Linux Foundation’s priorities, and most likely spending.
I say it’s a fair complaint.
I agree with the fact that it should be explained, but I definitely do not think the conspirational/rational-right-wingism/omgwokes tone is so pertinent.
How can you continue to call yourself "The linux foundation" and contribute so little?