cordova-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Roy T. Fielding" <field...@gbiv.com>
Subject Re: [Vote] Cordova 3.4.0 release
Date Wed, 19 Feb 2014 22:05:53 GMT
On Feb 19, 2014, at 12:43 PM, Ross Gardler wrote:

> A note on best practice. Prior to calling a release (i.e. when an RM is
> stepping up to create a release) there should be a "[DISCUSS] release x.y.z"

Some projects do that.  Others use an issue tracker (like Jira or a
simple STATUS file).  It is also reasonable to schedule one regularly.
Just understand that schedule != do.

> This provides a place for people to air concerns without affecting the VOTE
> thread. Ideally the VOTE thread is just a series of +1s and, in exceptional
> circumstances, some -1's See my other thread for clarity on what -1's mean.

Missed it, but I can explain.  -1s are no more important than +1s.
They express an opinion.  Nothing more than that.

> On 19 February 2014 10:37, Brian LeRoux <b@brian.io> wrote:
> 
>> This well illustrates my concern with tying a release to a vote thread.
>> There is always going to be a reason to -1, there will always be important
>> fixes, and there's always going to be a delays.

A -1 on a release is not a veto.  It just means that person would prefer
not to release this package.  Why?  They will probably explain, though
it isn't necessary for them to do so (unlike a veto on a patch/commit).

>> The security issues are definitely worthy of a patch release. I don't see
>> these windows issues as blocker for this release. Indeed this release
>> ideally happened a month ago.

This is why release votes are a majority decision with minimum 3 +1s.
Each person on the PMC has an independent right to make that call.
Formal votes have a tendency to invite opinions that might otherwise be
left unheard.

My usual criteria starts with "is it better than the last release?"
(note that there are a lot of things that could make it not better)
and ends with "does an imperfect release now prevent shipping a better
one next week?".  There are no perfect releases.

Nevertheless, this is a group decision in which each PMC member has an
equal vote.  Non-PMC member opinions are also important and might
influence the RM or PMC.  However, majority rules, and the release can
go forward as soon as the PMC majority says it can.

Earlier, Joe Bowser said:

> This is only useful if your goal is to never release anything.

Our goal is to be inclusive to community input on decision making.
We respect the opinions of the community even when we disagree
with them.  Feel free to explain why your personal opinion differs.
Do not dissuade others from expressing their own opinion.

Do not assume that you speak for Apache Cordova.
Only the result of voting speaks for the project as a whole.


Cheers,

Roy T. Fielding                     <http://roy.gbiv.com/>
Senior Principal Scientist, Adobe   <http://www.adobe.com/>


Mime
View raw message