Re: Gtk+4.0

[Sébastien Wilmet <swilmet gnome org>]

On Sun, Aug 14, 2016 at 01:05:08PM +0000, Philipp A. wrote:
> I tried just to read and not ask anything but no amount of reading has
> resulted in any enlightenment, so:
>
> Why not do what almost everyone does and have 4.X mean “stable” while
> anything with alpha/beta/pre/rc means unstable?
>
> KDE made the same mistake with the exact same version number, i.e having
> the number look stable to everyone while the software was (as they clearly
> said everywhere!) a pre-release. People used it, distros shipped it, it was
> buggy and incomplete and everybody was confused and angry as a consequence.
> Was it simply lack of historic knowledge that led to the GTK-4.0 decision?
>
> Besides, there's no gain in specifying some arbitrary minor version to be
> suddenly stable (as it was said GTK 4, “somewhere around 4.6” would
> become). There's exclusively a disadvantage, i.e. that you can't rely on
> common sense, convention, or any other kind of rule to know if that's a
> stable version. You have to know our look it up.
>
> Just use http://semver.org and you have something that follows the
> principle of least surprise.

See:
https://wiki.gnome.org/Projects/GTK%2B/Lifecycle

Nothing is decided yet.

There is another proposal with even/odd major versions to distinguish
between stable/unstable.

--
Sébastien