apex-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Thomas Weise <tho...@datatorrent.com>
Subject Re: master branch in place of devel-3
Date Thu, 21 Jan 2016 23:18:37 GMT
It my be possible but I would question why. The master branch does not
serve any other purpose, so why not use it for development?

On Thu, Jan 21, 2016 at 2:58 PM, Vlad Rozov <v.rozov@datatorrent.com> wrote:

> I guess not, I think that we don't have admin rights to manage apex core
> or malhar mirrors on github.
> Thank you,
> Vlad
> On 1/21/16 14:12, Pramod Immaneni wrote:
>> Can't we set a default branch in the repo to be different from master?
>> On Thu, Jan 21, 2016 at 2:04 PM, David Yan <david@datatorrent.com> wrote:
>> Hi all,
>>> We have been using the devel-3 branch for development in both Apex Core
>>> and
>>> Apex Malhar.  The reason was that we wanted to have the master branch to
>>> point to the latest release so that when a user checks out from our git
>>> repo, it's always the latest source release and it always works.
>>> But on the other hand, from what I see, that is not what most active
>>> apache
>>> projects do.  I checked Flink, Spark, Storm, Samza, Pig, Hive, and
>>> Hadoop,
>>> and ALL of these projects have commits on the master branch that are at
>>> most one day old.  Apex Core on the other hand, the last commit on the
>>> master branch was Nov, 2015, and that was when we released Apex Core
>>> 3.2.0.
>>> Because of our stale master branch, it's easy for someone outside of the
>>> Apex community to conclude that Apex is not very active compared to other
>>> Apache projects.
>>> To me, the benefits of using the devel-3 branch are simply not worth the
>>> potential cost.  I would like to propose that we get rid of the devel-3
>>> branch and use the master branch for development, instead of having the
>>> master branch always reflecting the latest release.  We use tags for that
>>> purpose.
>>> Please share your thoughts.
>>> Thanks!
>>> David

  • Unnamed multipart/alternative (inline, None, 0 bytes)
View raw message