You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@maven.apache.org by Jason van Zyl <ja...@tesla.io> on 2012/11/25 20:42:45 UTC
To RC or not RC
I wish RCs were useful, but I don't believe anyone really looks or takes the time to even try anything until it's actually released.
It's not hard to make RCs. So if folks want them, not a problem, I can roll an RC.
Anyone have an opinion?
Thanks,
Jason
----------------------------------------------------------
Jason van Zyl
Founder & CTO, Sonatype
Founder, Apache Maven
http://twitter.com/jvanzyl
---------------------------------------------------------
To do two things at once is to do neither.
-- Publilius Syrus, Roman slave, first century B.C.
Re: To RC or not RC
Posted by Benson Margulies <bi...@gmail.com>.
On Sun, Nov 25, 2012 at 7:22 PM, Olivier Lamy <ol...@apache.org> wrote:
> 2012/11/26 Stephen Connolly <st...@gmail.com>:
>> Well technically we only vote on the sources *not* the binaries *nor* the
>
> You don't test the binaries we will propose to users for download ?
> Do you think users download sources to build maven locally ?
> So for me binaries we produce as important as sources as long as it's
> material we propose to users.
Unfortunately, there's a big legal distinction between the sources and
everything else. If you like, you can think of a vote as being,
really, two votes: the one that creates the official Foundation
release, which is source, and a community vote to also shove the
convenience binaries onto the mirrors.
There are days when I think it would be less of a headache if the
Foundation's practice was to do releases in two phases, with a
testing/voting process for binaries that followed the official release
of the source. I'm not trying to talk anyone into this.
>
>> SCM tag... We just have tge habit if giving the tag for good form... With
>> the move to GIT, I don't see it as Bering so critical... You can always
>> include the tag hash ID in the vote email and reproducibility isn't so
>> important when we have the sources... Build ability of the sources is one
>> if the criteria we are supposed to check before voting
>
> That's not my point !
> Please do not mix release process and the tool we use for sources.
> My point is to have a separate branch for the release manager to not
> prevent others to commit in trunk.
>
>
>>
>> On Monday, 26 November 2012, Olivier Lamy wrote:
>>
>>> 2012/11/25 Jason van Zyl <jason@tesla.io <javascript:;>>:
>>> >
>>> > On Nov 25, 2012, at 2:07 PM, Brett Porter <brett@apache.org<javascript:;>>
>>> wrote:
>>> >
>>> >>
>>> >> On 26/11/2012, at 6:42 AM, Jason van Zyl <jason@tesla.io <javascript:;>>
>>> wrote:
>>> >>
>>> >>> I wish RCs were useful, but I don't believe anyone really looks or
>>> takes the time to even try anything until it's actually released.
>>> >>
>>> >> 3.0.4 had 5 RCs from memory, and so far, 3.1.0 has had one pulled back
>>> - seems like they are being useful?
>>> >>
>>> >
>>> > I doubt a user would have found that. For issues akin to what Anders
>>> noticed we can just leave it in staging for a week or two. I think people
>>> see RC and think "I'll just let everyone else try it."
>>> >
>>> >> I don't mind naming them all 3.1.0 and having the RM delete the tag,
>>> but we should keep enough time/repeats in there to find issues - we don't
>>> want another "don't use 2.2.0, it's broken" type scenario.
>>> >>
>>> >
>>> > That's easy enough. I'll just leave it in staging, but the crux of the
>>> argument is in staging or calling it an RC fields no real use from
>>> non-developers.
>>> >
>>>
>>> Perso what I like with rc mode (or call release branch mode) is the
>>> idea about having a separate branch for the rm to release. As it
>>> others can continue hacking in trunk.
>>> (minor detail: rc naming mode can prevent having to cleanup artifacts
>>> from various repo manager chain we are using)
>>>
>>> Furthermore we usually vote too on sources tarball and some binaries
>>> distributions (zip/tar.gz) and users will download those files from
>>> Apache distribution download sites.
>>> Note there is a process described here
>>> (http://maven.apache.org/developers/release/maven-core-release.html)
>>> for that.
>>> So please include those files in your vote call.
>>>
>>>
>>> >> - Brett
>>> >>
>>> >> --
>>> >> Brett Porter
>>> >> brett@apache.org <javascript:;>
>>> >> http://brettporter.wordpress.com/
>>> >> http://au.linkedin.com/in/brettporter
>>> >> http://twitter.com/brettporter
>>> >>
>>> >>
>>> >>
>>> >>
>>> >>
>>> >>
>>> >> ---------------------------------------------------------------------
>>> >> To unsubscribe, e-mail: dev-unsubscribe@maven.apache.org <javascript:;>
>>> >> For additional commands, e-mail: dev-help@maven.apache.org<javascript:;>
>>> >>
>>> >
>>> > Thanks,
>>> >
>>> > Jason
>>> >
>>> > ----------------------------------------------------------
>>> > Jason van Zyl
>>> > Founder & CTO, Sonatype
>>> > Founder, Apache Maven
>>> > http://twitter.com/jvanzyl
>>> > ---------------------------------------------------------
>>> >
>>> > Selfish deeds are the shortest path to self destruction.
>>> >
>>> > -- The Seven Samuari, Akira Kurosawa
>>> >
>>> >
>>> >
>>> >
>>> >
>>>
>>> ---------------------------------------------------------------------
>>> To unsubscribe, e-mail: dev-unsubscribe@maven.apache.org <javascript:;>
>>> For additional commands, e-mail: dev-help@maven.apache.org <javascript:;>
>>>
>>>
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: dev-unsubscribe@maven.apache.org
> For additional commands, e-mail: dev-help@maven.apache.org
>
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@maven.apache.org
For additional commands, e-mail: dev-help@maven.apache.org
Re: To RC or not RC
Posted by Olivier Lamy <ol...@apache.org>.
2012/11/26 Stephen Connolly <st...@gmail.com>:
> Well technically we only vote on the sources *not* the binaries *nor* the
You don't test the binaries we will propose to users for download ?
Do you think users download sources to build maven locally ?
So for me binaries we produce as important as sources as long as it's
material we propose to users.
> SCM tag... We just have tge habit if giving the tag for good form... With
> the move to GIT, I don't see it as Bering so critical... You can always
> include the tag hash ID in the vote email and reproducibility isn't so
> important when we have the sources... Build ability of the sources is one
> if the criteria we are supposed to check before voting
That's not my point !
Please do not mix release process and the tool we use for sources.
My point is to have a separate branch for the release manager to not
prevent others to commit in trunk.
>
> On Monday, 26 November 2012, Olivier Lamy wrote:
>
>> 2012/11/25 Jason van Zyl <jason@tesla.io <javascript:;>>:
>> >
>> > On Nov 25, 2012, at 2:07 PM, Brett Porter <brett@apache.org<javascript:;>>
>> wrote:
>> >
>> >>
>> >> On 26/11/2012, at 6:42 AM, Jason van Zyl <jason@tesla.io <javascript:;>>
>> wrote:
>> >>
>> >>> I wish RCs were useful, but I don't believe anyone really looks or
>> takes the time to even try anything until it's actually released.
>> >>
>> >> 3.0.4 had 5 RCs from memory, and so far, 3.1.0 has had one pulled back
>> - seems like they are being useful?
>> >>
>> >
>> > I doubt a user would have found that. For issues akin to what Anders
>> noticed we can just leave it in staging for a week or two. I think people
>> see RC and think "I'll just let everyone else try it."
>> >
>> >> I don't mind naming them all 3.1.0 and having the RM delete the tag,
>> but we should keep enough time/repeats in there to find issues - we don't
>> want another "don't use 2.2.0, it's broken" type scenario.
>> >>
>> >
>> > That's easy enough. I'll just leave it in staging, but the crux of the
>> argument is in staging or calling it an RC fields no real use from
>> non-developers.
>> >
>>
>> Perso what I like with rc mode (or call release branch mode) is the
>> idea about having a separate branch for the rm to release. As it
>> others can continue hacking in trunk.
>> (minor detail: rc naming mode can prevent having to cleanup artifacts
>> from various repo manager chain we are using)
>>
>> Furthermore we usually vote too on sources tarball and some binaries
>> distributions (zip/tar.gz) and users will download those files from
>> Apache distribution download sites.
>> Note there is a process described here
>> (http://maven.apache.org/developers/release/maven-core-release.html)
>> for that.
>> So please include those files in your vote call.
>>
>>
>> >> - Brett
>> >>
>> >> --
>> >> Brett Porter
>> >> brett@apache.org <javascript:;>
>> >> http://brettporter.wordpress.com/
>> >> http://au.linkedin.com/in/brettporter
>> >> http://twitter.com/brettporter
>> >>
>> >>
>> >>
>> >>
>> >>
>> >>
>> >> ---------------------------------------------------------------------
>> >> To unsubscribe, e-mail: dev-unsubscribe@maven.apache.org <javascript:;>
>> >> For additional commands, e-mail: dev-help@maven.apache.org<javascript:;>
>> >>
>> >
>> > Thanks,
>> >
>> > Jason
>> >
>> > ----------------------------------------------------------
>> > Jason van Zyl
>> > Founder & CTO, Sonatype
>> > Founder, Apache Maven
>> > http://twitter.com/jvanzyl
>> > ---------------------------------------------------------
>> >
>> > Selfish deeds are the shortest path to self destruction.
>> >
>> > -- The Seven Samuari, Akira Kurosawa
>> >
>> >
>> >
>> >
>> >
>>
>> ---------------------------------------------------------------------
>> To unsubscribe, e-mail: dev-unsubscribe@maven.apache.org <javascript:;>
>> For additional commands, e-mail: dev-help@maven.apache.org <javascript:;>
>>
>>
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@maven.apache.org
For additional commands, e-mail: dev-help@maven.apache.org
Re: To RC or not RC
Posted by Stephen Connolly <st...@gmail.com>.
Well technically we only vote on the sources *not* the binaries *nor* the
SCM tag... We just have tge habit if giving the tag for good form... With
the move to GIT, I don't see it as Bering so critical... You can always
include the tag hash ID in the vote email and reproducibility isn't so
important when we have the sources... Build ability of the sources is one
if the criteria we are supposed to check before voting
On Monday, 26 November 2012, Olivier Lamy wrote:
> 2012/11/25 Jason van Zyl <jason@tesla.io <javascript:;>>:
> >
> > On Nov 25, 2012, at 2:07 PM, Brett Porter <brett@apache.org<javascript:;>>
> wrote:
> >
> >>
> >> On 26/11/2012, at 6:42 AM, Jason van Zyl <jason@tesla.io <javascript:;>>
> wrote:
> >>
> >>> I wish RCs were useful, but I don't believe anyone really looks or
> takes the time to even try anything until it's actually released.
> >>
> >> 3.0.4 had 5 RCs from memory, and so far, 3.1.0 has had one pulled back
> - seems like they are being useful?
> >>
> >
> > I doubt a user would have found that. For issues akin to what Anders
> noticed we can just leave it in staging for a week or two. I think people
> see RC and think "I'll just let everyone else try it."
> >
> >> I don't mind naming them all 3.1.0 and having the RM delete the tag,
> but we should keep enough time/repeats in there to find issues - we don't
> want another "don't use 2.2.0, it's broken" type scenario.
> >>
> >
> > That's easy enough. I'll just leave it in staging, but the crux of the
> argument is in staging or calling it an RC fields no real use from
> non-developers.
> >
>
> Perso what I like with rc mode (or call release branch mode) is the
> idea about having a separate branch for the rm to release. As it
> others can continue hacking in trunk.
> (minor detail: rc naming mode can prevent having to cleanup artifacts
> from various repo manager chain we are using)
>
> Furthermore we usually vote too on sources tarball and some binaries
> distributions (zip/tar.gz) and users will download those files from
> Apache distribution download sites.
> Note there is a process described here
> (http://maven.apache.org/developers/release/maven-core-release.html)
> for that.
> So please include those files in your vote call.
>
>
> >> - Brett
> >>
> >> --
> >> Brett Porter
> >> brett@apache.org <javascript:;>
> >> http://brettporter.wordpress.com/
> >> http://au.linkedin.com/in/brettporter
> >> http://twitter.com/brettporter
> >>
> >>
> >>
> >>
> >>
> >>
> >> ---------------------------------------------------------------------
> >> To unsubscribe, e-mail: dev-unsubscribe@maven.apache.org <javascript:;>
> >> For additional commands, e-mail: dev-help@maven.apache.org<javascript:;>
> >>
> >
> > Thanks,
> >
> > Jason
> >
> > ----------------------------------------------------------
> > Jason van Zyl
> > Founder & CTO, Sonatype
> > Founder, Apache Maven
> > http://twitter.com/jvanzyl
> > ---------------------------------------------------------
> >
> > Selfish deeds are the shortest path to self destruction.
> >
> > -- The Seven Samuari, Akira Kurosawa
> >
> >
> >
> >
> >
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: dev-unsubscribe@maven.apache.org <javascript:;>
> For additional commands, e-mail: dev-help@maven.apache.org <javascript:;>
>
>
Re: To RC or not RC
Posted by Olivier Lamy <ol...@apache.org>.
2012/11/25 Jason van Zyl <ja...@tesla.io>:
>
> On Nov 25, 2012, at 2:07 PM, Brett Porter <br...@apache.org> wrote:
>
>>
>> On 26/11/2012, at 6:42 AM, Jason van Zyl <ja...@tesla.io> wrote:
>>
>>> I wish RCs were useful, but I don't believe anyone really looks or takes the time to even try anything until it's actually released.
>>
>> 3.0.4 had 5 RCs from memory, and so far, 3.1.0 has had one pulled back - seems like they are being useful?
>>
>
> I doubt a user would have found that. For issues akin to what Anders noticed we can just leave it in staging for a week or two. I think people see RC and think "I'll just let everyone else try it."
>
>> I don't mind naming them all 3.1.0 and having the RM delete the tag, but we should keep enough time/repeats in there to find issues - we don't want another "don't use 2.2.0, it's broken" type scenario.
>>
>
> That's easy enough. I'll just leave it in staging, but the crux of the argument is in staging or calling it an RC fields no real use from non-developers.
>
Perso what I like with rc mode (or call release branch mode) is the
idea about having a separate branch for the rm to release. As it
others can continue hacking in trunk.
(minor detail: rc naming mode can prevent having to cleanup artifacts
from various repo manager chain we are using)
Furthermore we usually vote too on sources tarball and some binaries
distributions (zip/tar.gz) and users will download those files from
Apache distribution download sites.
Note there is a process described here
(http://maven.apache.org/developers/release/maven-core-release.html)
for that.
So please include those files in your vote call.
>> - Brett
>>
>> --
>> Brett Porter
>> brett@apache.org
>> http://brettporter.wordpress.com/
>> http://au.linkedin.com/in/brettporter
>> http://twitter.com/brettporter
>>
>>
>>
>>
>>
>>
>> ---------------------------------------------------------------------
>> To unsubscribe, e-mail: dev-unsubscribe@maven.apache.org
>> For additional commands, e-mail: dev-help@maven.apache.org
>>
>
> Thanks,
>
> Jason
>
> ----------------------------------------------------------
> Jason van Zyl
> Founder & CTO, Sonatype
> Founder, Apache Maven
> http://twitter.com/jvanzyl
> ---------------------------------------------------------
>
> Selfish deeds are the shortest path to self destruction.
>
> -- The Seven Samuari, Akira Kurosawa
>
>
>
>
>
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@maven.apache.org
For additional commands, e-mail: dev-help@maven.apache.org
Re: To RC or not RC
Posted by Jason van Zyl <ja...@tesla.io>.
On Nov 25, 2012, at 2:07 PM, Brett Porter <br...@apache.org> wrote:
>
> On 26/11/2012, at 6:42 AM, Jason van Zyl <ja...@tesla.io> wrote:
>
>> I wish RCs were useful, but I don't believe anyone really looks or takes the time to even try anything until it's actually released.
>
> 3.0.4 had 5 RCs from memory, and so far, 3.1.0 has had one pulled back - seems like they are being useful?
>
I doubt a user would have found that. For issues akin to what Anders noticed we can just leave it in staging for a week or two. I think people see RC and think "I'll just let everyone else try it."
> I don't mind naming them all 3.1.0 and having the RM delete the tag, but we should keep enough time/repeats in there to find issues - we don't want another "don't use 2.2.0, it's broken" type scenario.
>
That's easy enough. I'll just leave it in staging, but the crux of the argument is in staging or calling it an RC fields no real use from non-developers.
> - Brett
>
> --
> Brett Porter
> brett@apache.org
> http://brettporter.wordpress.com/
> http://au.linkedin.com/in/brettporter
> http://twitter.com/brettporter
>
>
>
>
>
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: dev-unsubscribe@maven.apache.org
> For additional commands, e-mail: dev-help@maven.apache.org
>
Thanks,
Jason
----------------------------------------------------------
Jason van Zyl
Founder & CTO, Sonatype
Founder, Apache Maven
http://twitter.com/jvanzyl
---------------------------------------------------------
Selfish deeds are the shortest path to self destruction.
-- The Seven Samuari, Akira Kurosawa
Re: To RC or not RC
Posted by Brett Porter <br...@apache.org>.
On 26/11/2012, at 6:42 AM, Jason van Zyl <ja...@tesla.io> wrote:
> I wish RCs were useful, but I don't believe anyone really looks or takes the time to even try anything until it's actually released.
3.0.4 had 5 RCs from memory, and so far, 3.1.0 has had one pulled back - seems like they are being useful?
I don't mind naming them all 3.1.0 and having the RM delete the tag, but we should keep enough time/repeats in there to find issues - we don't want another "don't use 2.2.0, it's broken" type scenario.
- Brett
--
Brett Porter
brett@apache.org
http://brettporter.wordpress.com/
http://au.linkedin.com/in/brettporter
http://twitter.com/brettporter
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@maven.apache.org
For additional commands, e-mail: dev-help@maven.apache.org
Re: To RC or not RC
Posted by Christian Schulte <cs...@schulte.it>.
Am 11/26/12 00:49, schrieb Jason van Zyl:
> It's easy enough to not push the tag, or blow it away.
>
> So I'll just stage it again.
All I am saying is that this
<https://github.com/apache/maven-3/tree/maven-3.1.0> tag will be
incorrect, if you ever need to re-create another 3.1.0 tag in
subsversion from now on. I do not know if this can be corrected. If it
could, then someone could also correct that 3.0.4 tag ?
--
Christian
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@maven.apache.org
For additional commands, e-mail: dev-help@maven.apache.org
Re: To RC or not RC
Posted by Jason van Zyl <ja...@tesla.io>.
It's easy enough to not push the tag, or blow it away.
So I'll just stage it again.
On Nov 25, 2012, at 3:46 PM, Stephen Connolly <st...@gmail.com> wrote:
> +1
>
>
> On 25 November 2012 23:09, Benson Margulies <bi...@gmail.com> wrote:
>
>> On Sun, Nov 25, 2012 at 5:29 PM, Christian Schulte <cs...@schulte.it> wrote:
>>> Am 11/25/12 20:42, schrieb Jason van Zyl:
>>>> I wish RCs were useful, but I don't believe anyone really looks or
>> takes the time to even try anything until it's actually released.
>>>>
>>>> It's not hard to make RCs. So if folks want them, not a problem, I can
>> roll an RC.
>>>>
>>>> Anyone have an opinion?
>>>
>>> Problem is (was?) with tags pushed to github ending up wrong. For
>>> example, the tag 'maven-3.0.4' at <https://github.com/apache/maven-3>
>>> just isn't the final 3.0.4.
>>
>> I've always thought that one of the virtues of git was that the RM
>> could hold off pushing the tag until the release vote passes. If
>> people want to see it, we could push to a temporary vote repo to
>> expose it.
>>
>>
>> It's a pre 3.0.4-RC and there have been
>>> changes between that initial tag and the final tag in subversion, if I
>>> remember correctly. So tagging a 3.1.0 now, removing that tag and then
>>> re-creating another 3.1.0 tag is a bit problematic when the intial tag
>>> already got pushed to github.
>>>
>>> --
>>> Christian
>>>
>>>
>>> ---------------------------------------------------------------------
>>> To unsubscribe, e-mail: dev-unsubscribe@maven.apache.org
>>> For additional commands, e-mail: dev-help@maven.apache.org
>>>
>>
>> ---------------------------------------------------------------------
>> To unsubscribe, e-mail: dev-unsubscribe@maven.apache.org
>> For additional commands, e-mail: dev-help@maven.apache.org
>>
>>
Thanks,
Jason
----------------------------------------------------------
Jason van Zyl
Founder & CTO, Sonatype
Founder, Apache Maven
http://twitter.com/jvanzyl
---------------------------------------------------------
First, the taking in of scattered particulars under one Idea,
so that everyone understands what is being talked about ... Second,
the separation of the Idea into parts, by dividing it at the joints,
as nature directs, not breaking any limb in half as a bad carver might.
-- Plato, Phaedrus (Notes on the Synthesis of Form by C. Alexander)
Re: To RC or not RC
Posted by Stephen Connolly <st...@gmail.com>.
+1
On 25 November 2012 23:09, Benson Margulies <bi...@gmail.com> wrote:
> On Sun, Nov 25, 2012 at 5:29 PM, Christian Schulte <cs...@schulte.it> wrote:
> > Am 11/25/12 20:42, schrieb Jason van Zyl:
> >> I wish RCs were useful, but I don't believe anyone really looks or
> takes the time to even try anything until it's actually released.
> >>
> >> It's not hard to make RCs. So if folks want them, not a problem, I can
> roll an RC.
> >>
> >> Anyone have an opinion?
> >
> > Problem is (was?) with tags pushed to github ending up wrong. For
> > example, the tag 'maven-3.0.4' at <https://github.com/apache/maven-3>
> > just isn't the final 3.0.4.
>
> I've always thought that one of the virtues of git was that the RM
> could hold off pushing the tag until the release vote passes. If
> people want to see it, we could push to a temporary vote repo to
> expose it.
>
>
> It's a pre 3.0.4-RC and there have been
> > changes between that initial tag and the final tag in subversion, if I
> > remember correctly. So tagging a 3.1.0 now, removing that tag and then
> > re-creating another 3.1.0 tag is a bit problematic when the intial tag
> > already got pushed to github.
> >
> > --
> > Christian
> >
> >
> > ---------------------------------------------------------------------
> > To unsubscribe, e-mail: dev-unsubscribe@maven.apache.org
> > For additional commands, e-mail: dev-help@maven.apache.org
> >
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: dev-unsubscribe@maven.apache.org
> For additional commands, e-mail: dev-help@maven.apache.org
>
>
Re: To RC or not RC
Posted by Benson Margulies <bi...@gmail.com>.
On Sun, Nov 25, 2012 at 5:29 PM, Christian Schulte <cs...@schulte.it> wrote:
> Am 11/25/12 20:42, schrieb Jason van Zyl:
>> I wish RCs were useful, but I don't believe anyone really looks or takes the time to even try anything until it's actually released.
>>
>> It's not hard to make RCs. So if folks want them, not a problem, I can roll an RC.
>>
>> Anyone have an opinion?
>
> Problem is (was?) with tags pushed to github ending up wrong. For
> example, the tag 'maven-3.0.4' at <https://github.com/apache/maven-3>
> just isn't the final 3.0.4.
I've always thought that one of the virtues of git was that the RM
could hold off pushing the tag until the release vote passes. If
people want to see it, we could push to a temporary vote repo to
expose it.
It's a pre 3.0.4-RC and there have been
> changes between that initial tag and the final tag in subversion, if I
> remember correctly. So tagging a 3.1.0 now, removing that tag and then
> re-creating another 3.1.0 tag is a bit problematic when the intial tag
> already got pushed to github.
>
> --
> Christian
>
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: dev-unsubscribe@maven.apache.org
> For additional commands, e-mail: dev-help@maven.apache.org
>
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@maven.apache.org
For additional commands, e-mail: dev-help@maven.apache.org
Re: To RC or not RC
Posted by Christian Schulte <cs...@schulte.it>.
Am 11/25/12 20:42, schrieb Jason van Zyl:
> I wish RCs were useful, but I don't believe anyone really looks or takes the time to even try anything until it's actually released.
>
> It's not hard to make RCs. So if folks want them, not a problem, I can roll an RC.
>
> Anyone have an opinion?
Problem is (was?) with tags pushed to github ending up wrong. For
example, the tag 'maven-3.0.4' at <https://github.com/apache/maven-3>
just isn't the final 3.0.4. It's a pre 3.0.4-RC and there have been
changes between that initial tag and the final tag in subversion, if I
remember correctly. So tagging a 3.1.0 now, removing that tag and then
re-creating another 3.1.0 tag is a bit problematic when the intial tag
already got pushed to github.
--
Christian
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@maven.apache.org
For additional commands, e-mail: dev-help@maven.apache.org
Re: To RC or not RC
Posted by Jason van Zyl <ja...@tesla.io>.
I plan to be working on this for the next 8 weeks or so. There will definitely be another release within a couple months. Hopefully in 4-6 weeks.
On Nov 25, 2012, at 11:50 AM, Stephen Connolly <st...@gmail.com> wrote:
> If you are going to help do another release in the next 2 months or so
> (barring unforeseen circumstances) my vote is Jenkins tip style, no RCs,
> early and often.
>
> If we are going to resume the snails pace (I hope not) then RCs
>
> Thst's my €0.02
>
> -Stephen
>
> On Sunday, 25 November 2012, Jason van Zyl wrote:
>
>> I wish RCs were useful, but I don't believe anyone really looks or takes
>> the time to even try anything until it's actually released.
>>
>> It's not hard to make RCs. So if folks want them, not a problem, I can
>> roll an RC.
>>
>> Anyone have an opinion?
>>
>> Thanks,
>>
>> Jason
>>
>> ----------------------------------------------------------
>> Jason van Zyl
>> Founder & CTO, Sonatype
>> Founder, Apache Maven
>> http://twitter.com/jvanzyl
>> ---------------------------------------------------------
>>
>> To do two things at once is to do neither.
>>
>> -- Publilius Syrus, Roman slave, first century B.C.
>>
>>
>>
>>
>>
>>
Thanks,
Jason
----------------------------------------------------------
Jason van Zyl
Founder & CTO, Sonatype
Founder, Apache Maven
http://twitter.com/jvanzyl
---------------------------------------------------------
believe nothing, no matter where you read it,
or who has said it,
not even if i have said it,
unless it agrees with your own reason
and your own common sense.
-- Buddha
Re: To RC or not RC
Posted by Tony Chemit <ch...@codelutin.com>.
On Sun, 25 Nov 2012 15:24:54 -0500
Benson Margulies <bi...@gmail.com> wrote:
> I see no value to RCs around here. To me, RC's are valuable when the
> packaging is unstable, and you want a way for people to kick the tires
> of 'the package'. Our packaging is stable.
+1
>
> Any member of the community, even I, can check out the trunk and build
> the core and try it out. I don't need to wait for an RC for that.
>
> I'm in favor of just making releases and communicating to the user
> community what's in them. We could go as far as the HTTPD community
> and use a numbering convention for things that might not be suitable
> for the faint of heart.
+1
Just a shame to think of it after calling a vote :(
Let's have a 3.1.0 ;)
tony.
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@maven.apache.org
For additional commands, e-mail: dev-help@maven.apache.org
Re: To RC or not RC
Posted by Benson Margulies <bi...@gmail.com>.
I see no value to RCs around here. To me, RC's are valuable when the
packaging is unstable, and you want a way for people to kick the tires
of 'the package'. Our packaging is stable.
Any member of the community, even I, can check out the trunk and build
the core and try it out. I don't need to wait for an RC for that.
I'm in favor of just making releases and communicating to the user
community what's in them. We could go as far as the HTTPD community
and use a numbering convention for things that might not be suitable
for the faint of heart.
On Sun, Nov 25, 2012 at 3:17 PM, Mirko Friedenhagen
<mf...@gmail.com> wrote:
> Hello,
>
> I would support Stephen's suggestion and go with a quick 3.1.0 and
> liked it better to fix-forward and release a 3.1.1 quite soon
> afterwards. In the end you will get more feedback from a release then
> from a RC.
>
> Regards Mirko
>
> On Sun, Nov 25, 2012 at 8:50 PM, Stephen Connolly
> <st...@gmail.com> wrote:
>> If you are going to help do another release in the next 2 months or so
>> (barring unforeseen circumstances) my vote is Jenkins tip style, no RCs,
>> early and often.
>>
>> If we are going to resume the snails pace (I hope not) then RCs
>>
>> Thst's my €0.02
>>
>> -Stephen
>>
>> On Sunday, 25 November 2012, Jason van Zyl wrote:
>>
>>> I wish RCs were useful, but I don't believe anyone really looks or takes
>>> the time to even try anything until it's actually released.
>>>
>>> It's not hard to make RCs. So if folks want them, not a problem, I can
>>> roll an RC.
>>>
>>> Anyone have an opinion?
>>>
>>> Thanks,
>>>
>>> Jason
>>>
>>> ----------------------------------------------------------
>>> Jason van Zyl
>>> Founder & CTO, Sonatype
>>> Founder, Apache Maven
>>> http://twitter.com/jvanzyl
>>> ---------------------------------------------------------
>>>
>>> To do two things at once is to do neither.
>>>
>>> -- Publilius Syrus, Roman slave, first century B.C.
>>>
>>>
>>>
>>>
>>>
>>>
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: dev-unsubscribe@maven.apache.org
> For additional commands, e-mail: dev-help@maven.apache.org
>
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@maven.apache.org
For additional commands, e-mail: dev-help@maven.apache.org
Re: To RC or not RC
Posted by Mirko Friedenhagen <mf...@gmail.com>.
Hello,
I would support Stephen's suggestion and go with a quick 3.1.0 and
liked it better to fix-forward and release a 3.1.1 quite soon
afterwards. In the end you will get more feedback from a release then
from a RC.
Regards Mirko
On Sun, Nov 25, 2012 at 8:50 PM, Stephen Connolly
<st...@gmail.com> wrote:
> If you are going to help do another release in the next 2 months or so
> (barring unforeseen circumstances) my vote is Jenkins tip style, no RCs,
> early and often.
>
> If we are going to resume the snails pace (I hope not) then RCs
>
> Thst's my €0.02
>
> -Stephen
>
> On Sunday, 25 November 2012, Jason van Zyl wrote:
>
>> I wish RCs were useful, but I don't believe anyone really looks or takes
>> the time to even try anything until it's actually released.
>>
>> It's not hard to make RCs. So if folks want them, not a problem, I can
>> roll an RC.
>>
>> Anyone have an opinion?
>>
>> Thanks,
>>
>> Jason
>>
>> ----------------------------------------------------------
>> Jason van Zyl
>> Founder & CTO, Sonatype
>> Founder, Apache Maven
>> http://twitter.com/jvanzyl
>> ---------------------------------------------------------
>>
>> To do two things at once is to do neither.
>>
>> -- Publilius Syrus, Roman slave, first century B.C.
>>
>>
>>
>>
>>
>>
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@maven.apache.org
For additional commands, e-mail: dev-help@maven.apache.org
Re: To RC or not RC
Posted by Stephen Connolly <st...@gmail.com>.
If you are going to help do another release in the next 2 months or so
(barring unforeseen circumstances) my vote is Jenkins tip style, no RCs,
early and often.
If we are going to resume the snails pace (I hope not) then RCs
Thst's my €0.02
-Stephen
On Sunday, 25 November 2012, Jason van Zyl wrote:
> I wish RCs were useful, but I don't believe anyone really looks or takes
> the time to even try anything until it's actually released.
>
> It's not hard to make RCs. So if folks want them, not a problem, I can
> roll an RC.
>
> Anyone have an opinion?
>
> Thanks,
>
> Jason
>
> ----------------------------------------------------------
> Jason van Zyl
> Founder & CTO, Sonatype
> Founder, Apache Maven
> http://twitter.com/jvanzyl
> ---------------------------------------------------------
>
> To do two things at once is to do neither.
>
> -- Publilius Syrus, Roman slave, first century B.C.
>
>
>
>
>
>