Krita Sprint 2014, Saturday Meeting

[Sven Langkamp]

On Tue, May 20, 2014 at 2:39 PM, Boudewijn Rempt <boud at valdyas.org> wrote:

> We had a Krita Sprint in Deventer this weekend, and here's the notes from
> the Saturday morning meeting. We discussed a bunch of things, some relevant
> for Calligra as a whole, so I'm sending the notes to two mailing lists.
>
> * Krita Foundation
>
> The Krita Foundation currently has about 7000 euros to sponsor development
> with. We're going to do a big fundraiser really soon, too, the kickstarter
> project page is nearly done.
>
> The new splash screen helps a lot: one-off donations are seriously up
> already.
>
>
> * 2.9 and 3.0
>
> We would like 2.9 to be the last Qt4-based release. A precondition for
> this release is merging the Model-View-Controller branch (because that is a
> huge refactoring that makes porting easier). 2.9 will also have loads of
> interesting stuff for users to be happy while we port to Qt5.
>
> We think that the 3.0 release should be as pure a port to Qt5 as possible.
> No new features, just porting as quickly and purely as possible, release
> and then go on with 3.1 (where we might want to start using the new OpenGL
> framework from Qt5, btw).
>
> From previous experience with the sailfish qt5 branch, it would be a good
> thing to close master for a week while a volunteer does the mechanical
> parts of porting in one go: sanitizing, search and replace and all that
> sort of thing. After that effort is done, we can all pitch in and start
> fixing problems.
>
> It should be doable to port to Qt5 in just 3 months, which leaves us with
> two options:
>
>         * release the pure port as 3.0 when it's done, then go and work on
> new stuff.
>         * after the port is done, go on doing stuff until it's time for
> the regular release.
>
>
> * KF5
>
> For Krita and calligra-libs, limiting our KF5 dependencies to Tier1
> (karchive, k18n, kconfig, extra-cmake-modules) makes a lot of sense.
>
> * MVC branch
>
> We're afraid we're going to run into a brick wall here. The MVC branch is
> a _big_ refactoring, and basically touches every Calligra application
> except for Kexi (because Kexi doesn't use the Calligra platform libraries).
>
> It isn't feasible for the Krita team to port all Calligra applications,
> and many of our applications are unmaintained. As far as we can see, there
> are three options:
>
> * everyone chipping in to fix applications to work with mvc
> * removing broken, unmaintained applications
> * forking the calligra platform applications and make them krita-specific,
> which would sort of put Krita on the same footing as Kexi.
>

I'm in favor of option 3. We make a fork of the required parts and when the
other applications are ready they can be ported to the new stuff. Might be
better to give it a test run in Krita first before pushing it into the
other applications.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.kde.org/pipermail/calligra-devel/attachments/20140530/c9b9740d/attachment.htm>