I'm going to withdraw my -1 vote and instead abstain. I do feel that
Jubula should ideally be given more time to mature in the areas of
communication, community, and diversity. However, in light of the
fact that there have been at least a couple of previous projects
that were allowed to graduate while in a similar state, it would be
unfair to enforce selectively and punish Jubula.
Having said that, I will ask that there be scheduled some kind
post-graduation review in a few months to check that Jubula is
improving the areas outlined below and elsewhere in this discussion.
Wayne, I agree with you and others that some adjustments to how we
(PMC) prepare projects for graduation is needed. We need to either
change the graduation requirements or change the process, because
right now the guidelines we have aren't being followed consistently.
Eric
On 5/17/11 10:09 AM, Eric Rizzo wrote:
I'm sorry, but I am still unconvinced. I've reviewed the
graduation requirements as documented here:
http://wiki.eclipse.org/Development_Resources/HOWTO/Criteria_for_Graduating_from_Incubation_Phase_to_Mature_Phase
There are still too many of them that I haven't seen evidence of
Jubula satisfying. For example:
Active communities (see [2]).
- An active framework user (plug-in provider) community.
- An active tool user community.
The Jubula forum is almost empty; there are a grand total of 2
messages from people who are not Bredex employees.
- An active multi-organization
committer/contributor/developer community.
It's not clear if any of the committers are non-Bredex-employees.
If not, the diversity requirement has not been met.
Open and transparent project schedules.
There's almost no information about project plan/schedule
published on the web site or via the mailing list. Only a very
small token amount written to the mailing list just this past
Friday.
- The project decision making processes are published, and
all project decisions are being made in public.
This has been discussed as part of this graduation review thread,
but I can't see any evidence that it is in practice.
The underlying theme to me is that, in my eyes, graduation means
maturity, not "starting down the path towards maturity." I commend
the Jubula team for the great work they've done with regards to
many of the graduation requirements. Furthermore, any of these
above items, in isolation, would not be enough for me to be
uncomfortable with the project's graduation. But since there
appears, to me, to be a collection of graduation requirements not
yet fully satisfied, I feel it would be against our own published
guidelines and the spririt of EDP to graduate this project at this
time.
If there is specific evidence that some of the above items have in
fact been more fully satisfied, please point it out and I'll be
glad to reconsider again. As I said, it's not any one of them but
rather the collection.
I also still do not understand the urgency to graduate now. I
don't think it's a requirement to participate in the Indigo
Release Train; is it? I think given a few months Jubula would
easily address these items and be fully ready to graduate, so I
don't understand the resistance to allow it to incubate a while
longer.
Eric
On 5/17/11 9:35 AM, Wayne Beaton wrote:
Eric, have you changed your mind yet? :-)
Wayne
Achim Lörke <Achim.Loerke@xxxxxxxxx> wrote:
Okay, 1) and 2) have been worked on and should now be adequate. We will continue moving stuff from our internal Trac wiki to the Eclipse wiki.
on 3) Actually I think we do have a plan to work in the open. It just needs some time to work. We've decided that for our current community it would be easiest to move them from a product to OSS in tiny steps. But by considering all my mails on this topic and the feedback we've received from our activities (talks, articles, tutorials, etc.) it seems to be working.
- Achim
On 13.05.2011, at 21:02, Chris Aniszczyk wrote:
2011/5/13 Konstantin Komissarchik <konstantin.komissarchik@xxxxxxxxxx>:
Bug 345755 - Consider eliminating 0.x versioning requirement for incubating
projects
Bug 345757 - Consider adding informal graduation readiness review
Thanks for opening these two bugs. I think what we had here was a
misunderstanding of how the EDP works and a failure for mentors to
notice. I would like to see the Jubula project work more in the open,
from mailing lists to offering some support in their forums (and also
put a link on their website to their forums as a link for open source
support, along with the professional support option). I am abstaining
for now until a couple things are done
1) website is fixed to include link to forums or user related mailing
list for support. See the Mylyn page on how to tastefully mix open
source and commercial support options
(http://eclipse.org/mylyn/support/)
2) The contributing section of the website
(http://eclipse.org/jubula/developers.php) needs to be fleshed out
more... please see the EGit Contributor Guide as a reference for a
good contributor guide. You should also potentially consider having
your documentation in the wiki (a future date)
3) have a concrete plan on how to operate more in the open, you can do
this in the graduation review
Once these things are fixed, I will put forward a +1 vote to continue
the release in the spirit of doing what's right. In the future though,
I expect projects to operate in the open as much as possible even
though the source comes from a mature product code base.
--
Cheers,
Chris Aniszczyk
http://aniszczyk.org
+1 512 961 6719
|
