Adam,

On Fri, 2011-08-19 at 14:06 +1000, Adam Murdoch wrote:
[ . . . ]
> 
> We hope to get a 1.0-milestone-5 out in the next week or two, once we
> have fixes for the worst of the breakages in milestone-4.

Can I ask  when in the M4 beta sequence, the critical change was made.
I am finding that the last released M4-beta,
1.0-milestone-4-20110725000027+0200, seems to work where M4 itself does
not.
 
[ . . . ]
> This is the first time we've had to withdraw a release. I guess we
> need some practise, to get better at it. Normally, we'd just fix the
> issues and do a new release. This time we haven't been able to, given
> the nature of the change.

I think I'd prefer it if you got no further practice at doing it!

[ . . . ]

> Using 'milestone-n' as a versioning scheme was a mistake, because we
> don't have a way to distinguish between low risk releases like
> milestone-3, and the potentially unstable ones such as milestone-4.
> Once we get back to regular version numbering post-1.0, we can label
> the unstable releases as 'rc' or 'beta' or 'experimental' or whatever.

Please, please do not get into the Groovy habit of treating the label RC
as a synonym for beta.  RC is a name for a release candidate.  This
should be a version that can, barring a blocking bug being found, be
release simply by being relabelled.  It worries me that Gradle HEAD has
been assigned an RC label.  This is WRONG, it is nowhere near RC, this
should be labelled M5 with immediate effect.  I suspect the plan should
be for an M6 before even contemplating an RC.

Whatever the faults of the Mx system that was employed, please do not
pervert the RC label.


-- 
Russel.
=============================================================================
Dr Russel Winder      t: +44 20 7585 2200   voip: sip:[email protected]
41 Buckmaster Road    m: +44 7770 465 077   xmpp: [email protected]
London SW11 1EN, UK   w: www.russel.org.uk  skype: russel_winder

Attachment: signature.asc
Description: This is a digitally signed message part

Reply via email to