Re: [RFC] Detailing a service transition plan for glibc to CTI for August 2026.

Hi Zoë,

On Tue, 2026-07-28 at 00:43 +0200, Zoë Kooyman via Overseers wrote:
> August is coming up soon, and given the push of yet another draft
> proposal to move some glibc services, we are compelled to
> respond to the discussions around the hosting of glibc once
> again to get critical about the current proposal:
> https://sourceware.org/glibc/wiki/service-transition-plan.
> 
> The FSF continues to hear objections and questions, and though
> the mailing list doesn't show much response, the discussions
> reaching us make it seem like this is more about topic fatigue
> than any community agreement. The FSF also still has unanswered
> questions of its own.

Thanks for laying out the FSF stance. Even though this August
transition has been canceled, it would still be good to see if we can
get some discussion going around these open unanswered questions so we
can work towards something that has community consensus instead of just
keep pushing for having a "transition" without clear goals.

Let me at least point out how Sourceware can help.

> The plan presented now still shows many TODO's and very limited
> focus (Git and mailing lists). Part of the confusion and
> frustration around this project is the lack of understanding we
> have about goals. After years of work and discussion, a move of
> some mailing lists and a git repository cannot possibly be a
> satisfactory result for anyone?

It does indeed seem like unnecessary make work. But as several people
have pointed out, if this would be presented as setting up mirrors by
the LF for some of the services hosted on Sourceware that would be
genuinely useful. And we would be happy to make that easier.

> At this time it is unclear to me
> even who is actively participating in the TAC, who supports this
> project, or who has simply tuned out. The discussion really should be
> about what the best way forward is, where "the TAC decided" is
> not an argument.

It has indeed been unclear at times which people are supporting what
exactly and how the glibc community as a whole is making these kind of
decisions. I am happy the FSF is trying to help steer the discussion.
That will also help Sourceware set priorities.

> The serious concerns around transparency on funding and lack of
> clarity on what is gained in return continue to be problematic.
> The website now states: "The Core Toolchain Infrastructure (CTI)
> project is hosted and supported by the Open Source Security
> Foundation (OpenSSF), the Linux Foundation (LF) and its members.
> The primary sponsor is the OpenSSF. "  What is the longterm
> plan? It is unclear what exactly "The Core Toolchain
> Infrastructure (CTI) project" is. A good proposal would outline
> clear goals, a budget, and identification of any sponsors who
> are willing to make multi-year commitments to satisfy the major
> increase in expenses that would come with this move to the
> OpenSSF.

For me personally a big concern, and a blocker for cooperation (because
it is unclear who exactly we are cooperating with), is the very vague
proposed "charter". It would be good to clear up the organizational and
governance structure outlined in the charter so it puts Free Software
and community, instead of corporate interests, first.

For Sourceware we tried to do that through the Sourceware Fiscal
Sponsorship Agreement https://sourceware.org/mission.html#organization

> There are other ways to achieve improvements. The Sourceware team
> has taken major steps towards  professionalization over the
> years and made good faith efforts in cooperation and
> coordination. They deserve serious consideration of their
> proposals. It is also worth explicitly stating that the FSF's
> fiscally sponsored project fund is underused. The funds have
> grown substantially over the years and expenses have been
> minimal. I am sure donors to this fund would appreciate to see
> their money being spent towards improvements of the project's
> hosting.

For Sourceware we have a financials statement (for 2026/2027)
https://sourceware.org/financials.html where the budget currently comes
from individual sponsors https://sourceware.org/donate.html and a new
corporate sponsorship program https://sourceware.org/sponsor.html
(which seems to be on track to at least double our financial budget)
plus in-kind support for hardware/services.

> I appreciate the efforts of the current project leads of the
> changeover, I believe the motivation has been to bring positive
> effects for the infrastructure and the discussions have
> jumpstarted a range of improvements to the project's hosting, as
> well as necessary discussions, but the current proposal lacks
> consideration and fundamental support documentation.
> 
> Regardless of what comes next, it has always been the case that
> FSF and GNU want to place trust in the maintainers if the
> community does too, but it is also our responsibility to make
> sure that decisions are made responsibly. The FSF has officially
> asked Sourceware (via its Project Leadership Committee) to make
> sure we have fallbacks in place and to support those community
> members who want to use the services they know and which have
> shown guaranteed support for the FSF's mission.

Thanks. Sourceware of course stands ready to help.

> The current situation is making us question whether the proposed
> steps are indeed the right next  move for glibc. We're concerned
> about the time and energy being spent on this proposal while
> there are other genuine improvements we could achieve if we
> would all put our heads together.

As always we are happy to put our heads together. One opportunity would
be the Sourceware Open Office hour tomorrow. I'll sent out a reminder.

Cheers,

Mark