On Fri, 11 Dec 2009 13:08:02 -0800
Brian Harring <ferringb@...> wrote:
> On Fri, Dec 11, 2009 at 08:57:22PM +0000, Ciaran McCreesh wrote:
> > > Also, unless I'm on crack, the person leading PMS is fauli- I'd
> > > expect he's the one who can pull the veto trick, not you.
> > 
> > If *anyone* has any objections to patches, we resolve those
> > objections before proceeding.
> 
> Historically that has been a "do as I say, not as I do.".  Via
> ability to directly commit to pms, bits have gone in that would've
> been argued- or, bits have been left out that would've made the
> change in general a no go.

Not since the Council agreed on PMS as a draft standard for EAPI 0 they
haven't.

> Unfortunately because of the way the rules are ran, once it's in all 
> it takes is one person stonewalling to keep from getting it fixed- 
> catch 22, if they can push it in then they get it via pulling a veto.

Please point to any patch on the subject that has been sent to this list
that has been rejected, by veto or any other means.

> Further, frankly it provides a way for you to stonewall fixing known 
> flaws- the entire life of PMS you've been trying to force extended ~ 
> atom support and no one can get that bit removed because *you* 
> stonewall it.  You wrote the original bits, now we can't fix the 
> things you forced in via this idiotic veto rule.

Again, point to patches please.

> I digress.  Take it to the council as said, it would be interesting
> to see the slap down on this one, and frankly PMS does need to be far 
> more democratic.  Pointing at academic issues (1^23 chance is 
> academic, although yes, sorting it out for the academic case where 
> the FS supports NS is useful) as a claim that the majority cannot 
> overrule is plain political idiocy.

The problem with writing code that sometimes doesn't work is that it
sometimes doesn't work. And I've no idea what "1^23" is, but it doesn't
look like the actual odds of it going wrong, which are around one in
ten million per file, or one in a thousand on any given system, or a
hundred affected systems once people start using newer filesystems.
Hardly academic.

If a democracy votes that it's ok to construct bridges out of materials
that are known to cause random collapse one time in a thousand, would
you consider those bridges to be well designed?

> Either that or we just back off and let you get your way per the
> norm. This I consider an untenuable solution if PMS is to have any 
> relevance long term.

You still haven't explained what's wrong with doing a carefully phased
withdrawal, rather than running around with an axe lopping bits out
just because you can.

-- 
Ciaran McCreesh