Re: [eclipse-pmc] [eclipse-ide-wg] Fwd: [cross-project-issues-dev] IMHO

[Roland Grunberg <rgrunber@xxxxxxxxxx>]

On Thu, 2021-06-03 at 04:58 +0000, Sravan K Lakkimsetti wrote:
> I am see more and more discussion on usefulness of Orbit project. But
> there is no discussion on how we are going to maintain the same. With
> Roland taking other responsibilities in his organisation and not
> having much time for Orbit project(See:
> https://www.eclipse.org/lists/orbit-dev/msg05359.html), with no one
> else stepping in/up I don’t see a point in discussing usefulness of
> Orbit project, if there is no one willing to maintain it.

Looking at what Sravan mentioned in this issue, it does seem like steps
were taken to ensure the impact of this change would be minimal. I
can't really say Orbit inclusion would have prevented this because we
have also had cases where changing Orbit content caused such problems
at the simrel level. Is testing of simrel going to increase to catch
more of these issues ?

The nice part about Orbit, which we sort of miss with this approach is
that as a "common" repo, you're much more likely to have people
reviewing your changes than if you do it off in your own repo. This at
least guaranteed some basic level of quality across all bundles,
particularly when reviewed by those who have more experience in where
problems can arise.

Does it make sense to limit Orbit to just those libraries that are
known to be heavily shared among the various projects ? How about some
triage system where a different group of people handles Orbit reviews
on milestones ? On the other hand, if the process is still seen as a
burden on maintainers, I can't really blame anyone for opting out of
the process.

As for maintenance of Orbit. Yes, I don't have as much time as before
to review / simplify things. At best, I have time to kick off builds as
needed for milestones and maybe in case of serious issues, spend some
time to investigate.

-- 
Roland Grunberg