Re: [Evolution-hackers] Reconsidering our release cycle

[Srinivasa Ragavan <sragavan gnome org>]

Hi Fabiano,

On Wed, Jul 24, 2013 at 6:29 AM, Fabiano Fidêncio <fabiano fidencio org> wrote:
> Srini,
>
>
> > I really wouldn't want EDS to be part of this, if we ever want it to
> > be a proper platform/core material. Just Evolution would be better fit
> > for this model IMHO.
>
> Could I ask you why?
> If you check our git's activity you will see that the most part of
> bugs we are fixing are touching both in Evolution as in EDS.
> I really cannot imagine these two parts not walking together.

I know how tight the development is. Infact, in my watch, we added a
configure check that mandates same minor & micro version. But IMHO
this model should change for good. It kind of brought lot of
flexibility for introducing breakages because we know Evolution would
work with only the right version of EDS well but we kind get into a
model of ignoring the apps that would depend on EDS. All along we
(Evolution/EDS hackers) were blamed for breaking API/ABI. Im not
support this following statement, but there was a time, when you can
just build evolution and not use latest EDS/Gtk/other platform bits on
atleast 2 older GNOME releases. I may not ask for this, but atleast we
should have EDS decoupled from Evolution releases to let it
survive/develop as a platform on its own. Some might argue that having
a yearly releases will give longer API/ABI stability, but again
coupling EDS & Evolution will let us into more breakages.

My point for this was not about the duration, but about decoupling the
EDS & Evolution releases to plan/avoid any API/ABI (re)designs for
EDS. This is my thought, and some may support it but a lot might
oppose it ;-).

Thanks,
-Srini.
> Best Regards,
> --
> Fabiano Fidêncio