Re: Sourceware / GNU Toolchain at Cauldron

[Mark Wielaard <mark@klomp.org>]

And here are some (my client connected and disconnected a lot, so I
might have missed some) relevant #overseers irc logs with comments
from people participating remotely:

bkuhn:
As part of answer to the question of “is SFC's capable of
fundraising for its projects?”
https://sfconservancy.org/docs/software-freedom-conservancy_Form-990_fy-2020.pdf
 and
https://sfconservancy.org/docs/software-freedom-conservancy_independent-audit_fy-2020.pdf#page=11

karen:
I didn't include it in my estimation of our grant funding but I
am currently at the signing paperwork stage for a $450k grant from a
major grantmaker

bkuhn:
I agree with codonell that you need a plan, but the *plan*
(including costs) has to come before doing any fundraising, and for a
FOSS initiative that plan should be publicly available.  That's one of
the things that  SFC helps its projects do.

serhei:
so there was an entire presentation on infrastructure that
got skipped over :) the infrastructure is being implemented as
frugally as possible according to unix philosophy, but there are
potential ideas to develop it further with either fixed or ongoing
costs.

fche:
https://gnu.wildebeest.org/~mark/sourceware/presentation.html << the
talk, as prepared

bkuhn:
I do want to note that dje mentioned Yocto as an example of what LF
can provide.  I note that Yocto's infrastructure is proprietary
software, including even for the mailing lists.
https://lists.yoctoproject.org/g/yocto

https://lwn.net/Articles/700479/

serhei:
it is going to take time to lower the temperature from the way
this has been announced & incorporate those into a more credible
proposal

karen:
I was just noting that SFC has close relationships with RH and
have been talking to them about this (and other things) in a productive
way

bkuhn:
We are too … we at SFC advise all our projects to write up a very
clear proposal for their project plan, ideally on that is for two
years out.

We then encourage the projects to attach specific currency amounts and
lists of hours of staff/contractor time needed for each thing during
those two years. and then get a bottom line number, and THAT plus
15-20% (for safety, in case you under budgeted) is your fundraising
target you show that plan to the public and your potential funders,
and say: "Do you want to fund this?"

The problem with seeking funding FIRST before the plan is clear is
that it gives funders the opportunities to lobby you to do it
differently than the community wants. We see it all the time, it
happens to every organization that relies on contributions (as opposed
to selling a product/service) to do its work.

serhei:
there's a term by Jane Jacobs 'catastrophic money' that encapsulates
the risks of someone coming in and suddenly going "here is a pile of
money"

you end up building something completely different from what you would
build when you have to be cautious with resources and you risk
building something that is far less resilient

bkuhn:
SFC is absolutely willing to work with Overseers both on budget
and making the plan a bit more ambitious.  I do think it's unlikely
400k/year this soon can be used effectively, but I also think (and the
Overseers will be annoyed I've said this) that they are not quite
ambitious enough in the current plan.

The clearly written plan allows you to stick to your plan when you're
tempted by 💰💰💰 to just do it "slightly different" … because, slowly
it becomes it's not so slightly, and then next thing you know, you're
a year out and are like "Um, why are we doing what the funders want
rather than what we want?"

Note that the mailing list thread on overseers@sourceware is a good
place to further this discussion.  I'd really love to see someone post
a summary of the session to that thread.