[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Re: Ready to vote on 2004-003?



On Fri, May 21, 2004 at 12:57:45AM +1000, Anthony Towns wrote:
> Does it matter who said it, or if it's true?
> 
> Is it true?

I do think there's still a bit too much ambiguity floating around,
if that's what you're asking.

> >    We resolve to release Sarge as expediently as possible based on
> >    the current set of packages which have been designated as release
> >    candidates for Sarge, regardless of the state of DFSG compliance of
> >    any of those packages.  
> 
> That would be in direct violation of the social contract. As a project,
> do we want to wilfully and deliberately violate the social contract? I
> wouldn't have thought so. If we're going to do that, why bother having
> a social contract in the first place?

Call it a rider on the social contract.  It didn't occur to me that
anyone would take this as anything but -- I'll make sure to reword the
proposal so it is clear.

I would not expect something like this to be relevant if it doesn't get
the 3:1 supermajority which is required of social contract changes.

If it doesn't pass with a 3:1 supermajority requirement, that says
something rather significant.

If the idea is so unpopular that it doesn't get any seconds, that says
somthing rather significant, too.

> If we are going to do that, there's no need for a GR; the technical ctte
> can just use that as the reasoning for setting the release policy back
> to what it used to be right now.

What reasoning?  That was an expression of intent, not an expression of
the reasoning leading to that intent.

The reasoning leading to the intent fits in the same category as "common
sense" and "historical precident" and so on.  I'll write up a more formal
rationale this evening.

> If the tech ctte's not confident in doing that -- and you're a member,
> so you should have a pretty good feel for that -- then you should ask
> yourself why you think it'd be better if there was a GR backing it up.

Because the tech committee probably shouldn't be overruling explicit
statements from the social contract.  It might be that more general
statements in the social contract allow for this, but at least some of
the committee members seem to feel that making a formal announcement
would damage what the rest of the developers are doing.

> I'd have thought good ideas could be justified purely on their merits, and
> the weight of numbers is only useful for distributing the blame for bad
> ideas so no one feels too responsible them. Should we be implementing
> good ideas, or bad ones? Obviously, if you think it's a good idea,
> and are confident about it, you've got an easy path to resolving this
> issue. If anyone else has a problem with it, they can work out what to
> do, and try to overrule the tech ctte by GR themselves.

I don't know what you are trying to address, here.

> Having a contract like that helps us be consistent, even when we might be
> tempted to make exceptions, which lets our users easily decide "Should
> we use Debian or something else?", because they know exactly how we'll
> handle things, rather than having to guess at what mood the DPL, RM,
> or debian-admin is in on the day.

And here it's pretty obvious that we (for some value of "we") want
an exception.  Why not just call this exception an exception?

Look at it this way: we are working towards a goal and here we're tripping
over our own feet.  In this context, a boulder between us and our goal
seems like it might be useful to lean on, for getting back up on our feet.

Or, if metaphor isn't your bag, come up with your own way of looking
at it...

> Various folks seem to think the appropriate group to interpret the
> social contract is subscribers to the debian-legal list. That question
> should probably be considered, but I've done everything I can to make it
> possible for the tech ctte to make that decision. If they choose not to,
> or we decide they're not able to, someone else will have to make the
> decision. I will not. There are quite a few possibilities, each with
> varying degrees of effectiveness and exploitability.
>
> Interpreting the social contract as it stands to allow GFDL'ed docs
> into main for sarge seems to me to make a mockery of the entire purpose
> of the social contract, but, again, if folks think that's a good idea,
> then you don't need to convince me to do it.

If need be, we could pass a "revert the social contract" and if you
didn't like that the tech committee could ratify a "revert the release
policy to the way it was before the social contract changed".  But that
seems unnecessarily indirect.

What I'm proposing, instead, is that we explicitly override the social
contract [temporarily, within some limits, based on what we've been
doing for years] and just be done with it.

-- 
Raul



Reply to: