Meeting summary/minutes from today's FESCo meeting (2010-09-14)

Vaclav Misek misek at tnet.cz
Wed Sep 15 13:26:08 UTC 2010


 On 09/15/2010 01:48 PM, Adam Williamson wrote:
> On Tue, 2010-09-14 at 20:48 -0400, Máirín Duffy wrote:
>
>> Is anyone else feeling a little uncomfortable about the voting process,
>> irregardless of its conclusion?
> 'regardless'. 'irregardless' would mean 'not regardless'. =)
>
> Yes, I agree, and I'd like to point up another procedural issue here.
> During the meeting it was generally assumed that this was just a usual
> 'do we approve this feature' vote, in which case the 'default' would be
> 'no', and the onus would be on the 'yes' side to get five votes to have
> the feature approved. It was essentially rejected by default - it was
> rejected because there weren't five people voting in favour, not because
> there were five people voting against.
>
> I think this is an erroneous interpretation, because this wasn't a
> normal 'do we approve this feature' vote. systemd had in fact already
> been voted on as a feature at an earlier fesco meeting and had been
> *provisionally accepted* - that is, it was accepted, with the proviso
> that if fesco was particularly worried about something, it could reverse
> that acceptance any time prior to beta release.
>
> Given the previous provisional acceptance of systemd, I would argue that
> the situation at the meeting should actually have been that *accepting*
> systemd would be the default case, and it should have taken five 'no'
> votes (or five 'yes' votes to the proposal 'do we reverse our earlier
> decision and reject systemd?') to reject it - it shouldn't have been
> rejected just because five yes votes couldn't be found on the day.
From my point of view all this situation is clearly wrong.
I understand that Lennard is pissed off, he followed the tight schedule
and tried to fix as many bugs as possible to deliver working init system
in time.
I understand why systemd was postponed as default init system foro F15,
but this discussion/decision should come much much sooner, with clear
criteria for accepting/rejecting it. It should be stated at the
beginning of process that systemd should be accepted when ... yada yada.
In case it does not meet these criteria it should be rejected. But the
terms for acceptance were IMO drastically changed at the end.
Kind regards

        sHINOBI



More information about the devel mailing list