[Qgis-psc] PSC Meeting log from 8/4?

[Nathan Woodrow]

Hey all,

Email below from Regis, who doesn't seem to be able to post here.

--------------------------------------------

Hi all,

Sorry for posting in psc list when I’m not part of it.

We funders have many on-going contracts for features we hope to see
landing in 2.10 or 2.12.  We also fund and maintain plugins -
sometimes linked to core new features – and I fear that we are
currently funding things that will need refactoring because they will
never land in a 2.x release.



My point is that we expect at least a 2.10, or better a 2.12 LTR. QGIS
3.0 with new PyQt5 API , Python 3 and composer refactoring  + 400
plugins to maintain is not a light task. There is a new learning curve
for python dev’s, and we may expect a longer fixing period in core and
a full year for upgrading most plugins.



That said, I strongly think you should advertise funders and non-psv
developers to stop working or funding any new feature on a clear and
well advertised roadmap, with explicit deadlines.



Advertising the benefits of this migration and calls for funding that
migration could help to stop the “add feature” fever. That would help
us sharing efforts to jump on 3.0 without having a great part of the
community sticking with python plugins for 2.x branch to “replace”
missing core features.



My 2 cents

Cheers,

Régis


On Mon, 13 Apr 2015 at 23:43 Richard Duivenvoorde <richard at duif.net> wrote:

> On 13-04-15 15:17, Luigi Pirelli wrote:
> > On 13 April 2015 at 13:49, Nyall Dawson <nyall.dawson at gmail.com> wrote:
> >> On 13 April 2015 at 18:44, Luigi Pirelli <luipir at gmail.com> wrote:
> >>> hi,
> >>>
> >>> may we plane a dedicated timespace during next Hackmeeting?... and
> >>> probably, when goals will be more fixed, this discussion would be
> >>> moved (technically speaking) to qgis-developer.
> >>
> >> Do you mean during the hackfest? That's a bit late if 2.10 is going to
> >> be 3.0. It only gives a couple of days between decision and release...
> >>
> > next in Denmark? it will be in 2 month!
>
> As I understand we have some pretty exciting stuff in line willing to be
> incorporated in a 3.0 version (Nyall's stuff, Marco's geom stuff,
> qml-stuff, ??? ...). Will putting all this nice stuff into master not
> making it harder to backport other smaller fixes into 2.8 again? Feature
> freeze is just 38d away: enough to put all this stuff into master, but
> also easy to put a lot of side issues in it.
>
> In my view our 'commitment' to universities etc to have a LTR would also
> mean that we have a clear planning. Like:
> - 2.10 will be the last one in the 2.0 line, so we will try to backport
> issues untill 2.10 (and maybe 2.10.x) is released
> - then we start 3.0 with all the new nice thingies.. and we make sure
> that the next LTR will be as stable one too (while 3.0 probably will
> have some toothing issues...)
>
> My gutt feeling now is that I'm not in favour of going to a 3.0 version
> one version after 2.8 L(!)tr. I see the Juergen's argument 'it is just a
> number', but I think that for non community members this numbering AND
> planning of it is more important (then it actually is for us)?
>
> Regards,
>
> Richard Duivenvoorde
>
> ps there will always a tension about the fast-moving devs and the maybe
> more conservative large corporation/governmental/university user-base...
> So whatever is decided, I'm in :-)
>
>
> _______________________________________________
> Qgis-psc mailing list
> Qgis-psc at lists.osgeo.org
> http://lists.osgeo.org/mailman/listinfo/qgis-psc
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.osgeo.org/pipermail/qgis-psc/attachments/20150414/b667f67a/attachment.html>