Re: 2.14 release, or GOP now?

[Graham Percival]

On Sun, Oct 24, 2010 at 1:56 PM, Valentin Villenave
<address@hidden> wrote:
> On Sun, Oct 24, 2010 at 5:22 AM, Graham Percival
> <address@hidden> wrote:
>> We're about 10-20 hours of work away from having 0 Critical
>> issues.  On one hand, that sounds great; we're almost there!  On
>> the other hand, we've been in this state for the past month.  I'm
>> not seeing a lot of excitement and work towards 2.14.
>
> I'm afraid that the longer you wait, the less people will be
> "excited". As I told you before, I think you should get 2.14 out now,
> or at least a release candidate.

Who's these "people"?  Patches for the 5 current Critical defects will
almost certainly come from Carl, Joe, Patrick, Neil, Reinhold, or
David.  (apologies if I missed anybody -- if there's no other
emergencies next week, I'll tackle 1336)
I don't think that any of those persons would work any harder if I
made a release candidate.  Many of them are involved the academic Fall
term (either as students or teachers), so we should expect them to be
busy.

If we _did_ have a 2.14.0 release now, what would happen?  Most people
would try running it on their old scores.  Many would encounter one of
the "music running off the end of the page" bugs.  Those are serious.
I mean, if I was trying to use a new music typesetting program, and it
tried to print music outside the page, I'd switch to Sibelius pretty
quickly.
But wait, it gets worse!  Suppose I decide to be helpful, so I send a
bug report about the issue -- and get told "yeah, we knew about that 3
months ago, but we made a new stable release anyway".

I really don't think that we can ignore the spacing bugs, especially
since "improved vertical spacing" is one of the main new features of
2.14.


Granted, we could probably get away with ignoring the segfault issues
-- they only happen in pretty specific circumstances.  But this is
really a cultural issue between windows+osx programs and unix
programs.  In unix-land, a segfault is a serious, serious problem.  In
windows+osx-land, a program crashing isn't particularly worth
remarking on.

Believe me, I've spent time looking at these issues, trying to find
any excuse to downgrade the priority.  Other than 1355 (for which we
could argue "it's impossible to ensure that scheme code doesn't die in
bad ways, so that segfault isn't important), and 1288 (which will be
fixed soon anyway), I can't in good conscience describe them as "ok
for a stable release".  I mean, if they were *unknown* bugs that were
only reported after 2.14.0 was out, I wouldn't lose any sleep over
them.  But since we know about them already, I don't think we can
honestly claim that 2.13 is ready for stable productive use.

Cheers,
- Graham