Resolution 2009-03-16.jrk.1: OpenWRT as associated project [revised]

Ian Jackson ijackson at chiark.greenend.org.uk
Tue Mar 17 19:08:37 UTC 2009


Jimmy Kaplowitz writes ("Re: Resolution 2009-03-16.jrk.1: OpenWRT as associated project 	[revised]"):
> I thought other people were explicitly saying they didn't want SPI
> to have to monitor internal OpenWRT operations, and that they
> considered the arrangement with Debian to be a mistake which we
> shouldn't repeat. Can additional people give their thoughts?

This has been an ongoing topic of, shall we say, controversy.

My view is that SPI should of course normally take the liason's word
for it.  After all anything else would be impractical.  But if there
is a serious suggestion that there is some kind of dispute, we should
do what the project wants us to.

If an associated project's decisionmaking structures are sufficiently
incomprehensible or fractious that this is impossible then we may have
to revisit this but in practice this hasn't been a problem even with
respect to Debian (which is large, has fairly complex structures, and
seems to attract kookery).  We have had the odd allegation of
wrongdoing or dispute but nothing that wasn't easily dealt with.

Some other people have said that we should not ever look into a
project's processes because (as I understand the arguments),

1. It's impractical to look into a project's processes in every case.
This is not, I think a valid argument - since I only propose to do so
in cases of dispute.

2. Looking into a project's decisionmaking processes would somehow be
an infringement on the project's right to self-determination.  To that
I would respond with the opposite view - that requiring the project to
delegate complete control over financial resources to specific
individuals is precisely an interference with the project's right to
decide for itself how spending decisions should be made.

If I have mischaracterised these positions please forgive me.  I'm
doing so to try to clarify and assist the discussion, not to attack
straw men.

Josh Berkus's `Associated Project HOWTO' (of which I only became aware
today) seems to take this latter view.

Ian.


More information about the Spi-general mailing list