Re: The new Social Contract and releasing Sarge
Anthony Towns <aj@azure.humbug.org.au> writes:
> Well, that's fine, and is now clearly the case. I can't see why you'd do
> anything but expect the procedure to be dropped once the interpretation
> it was based on was rendered implausible.
I haven't said anything to the contrary that I can recall; if I did,
it was wrong.
> > Because I'm willing to tolerate slowness and areas of ambiguity, I did
> > not expect that this would mean that sarge would have to be delayed
> > until it gets brought into compliance.
>
> Well, there's no ambiguity anymore to tolerate, and if slowness is the
> only remaining choice then that implies sarge gets delayed until it's
> in compliance.
No, slowness is no problem. I don't mind slowness in the release of
sarge, but I also don't mind slowness in adapting to changing
expectations for our work, including slowness in adapting to changes
in the Social Contract.
Similarly, if we had passed the GR to drop non-free, I would not have
insisted that it happen overnight (and regardless, it could not
*possibly* happen instantaneously). Things take time, and that's fine
with me. If it's not fine with you, then we disagree about that. It
doesn't make either of us unprincipled, we just disagree about how
best to implement changes in our standards.
There's no particular merit fighting about such a disagreement,
provided the terms of it are clear. I would much prefer to settle
such things by straightforward votes than by protracted argument.
Just as new policy changes do not necessarily produce release-critical
bugs, I don't view new SC changes as doing so either. But you can
decide what you think is best, though I would ask you to be as
up-front as you can, hopefully giving advance descriptions of how you
judge things. How fast to adapt to a change is a policy decision.
Then, if the developers disagree with your policy decisions, they
could overrule them by GR. At the present, I am really quite
undecided about the matter.
> > But it's really Anthony's decision,
>
> I know it's fun to say things like that so you don't have to take
> any responsibility yourself, but it's really _not_ my decision. It's
> the project's. If the project decides that the social contract is
> less important than I think it is, or that there are other possible
> consequences of the changes to the social contract than the ones I see,
> well, that's fine, and that's the decision, no matter what I say or think.
The project can certainly overrule your decision. I am undecided
whether it should do so, or what the best means would be if it were to
do so. I'm saying that, absent any overruling, it's your decision.
> If the new rules are better than the old rules, or even just better
> than the wrong interpretation of the old rules Debian had been using,
> why wouldn't we want to apply the new rules?
Oh, that's an easy question to answer. Because we want the new rules
to go into place, but we don't insist they do so instantaneously.
Similarly, it is common in the US to pass new emissions standards for
automobiles that affect all new cars, while allowing old cars to
remain on the road adhering to weaker standards.
Thomas
Reply to: