gtk+-2.8.14 version bump

Couannette couannette at
Thu Mar 9 20:12:28 UTC 2006


*taking a "absolute" point of view*.

If galculator depends on glib 2.x if depends on 2.x and not on 2.y.
It was written with the glib 2.x API in mind and can't, by construction, be
compatible with glib 2.y that propose a different API.
That's why major distributions continue to deliver very old libraries: to
support old application that are usefull but no more actively developped.

We can't install 2 versions of the same library into Lunar (not as Gentoo for
example, that can have a slot for each "usefull" version of a given library).

I don't say that we should by the way. Keep Lunar Simple :o)...

We just have to say, as Chan said, that upgrading this library for the sake of a
majority of application will break a small number of other.

Those application that are no more actively developped are doomed in a source
based distro like ours.

You have to "faire une croix dessus" as we say in french: you have to give up
with it, get rid of it.

Anyway if you want to use those application you have the choice to install Debian.


Terry Chan a écrit :
> So by your logic no updates should ever be allowed because they break ONE
> poorly written application?
>>From the lack of cvs commits to the galculator sf page it would appear that
> galculator is NOT an actively maintained project.
> Terry Chan
> -------------------------------------------------------------
> On Thu, Mar 09, 2006 at 11:08:59AM -0800, Eric Sandall wrote:
>>On Wed, 8 Mar 2006, Terry Chan wrote:
>>>Wed Mar  8 14:22:59 2006  Tim Janik  <timj at>
>>>       * NEWS: announce change and glib-2.10.1 requirement.
>>>       * gtk/gtkobject.c: fixed up messing with non-GtkObject floating flags.
>>>       this requires glib-2.10.1.
>>>I've bumped gtk+-2 on my lunar box and recompiled many gtk+-2 apps with no
>>>problems.  I'm already at glib-2.10.1.
>>>glib-2.10.1 works just fine with all of xfce4-profile, including xfce4-panel.
>>>Unless I hear otherwise I WILL be bumping gtk+-2 to 2.8.14 and at the same
>>>time I'll also bump glib-2.10.1.
>>>I really don't care that galculator segfaults with the new glib-2.  That's
>>>a problem that the galculator devs MUST fix, not the lunar devs.
>>Shouldn't updates /not/ break currently working packages? Do the
>>galculator devs know that the new glib-2 breaks their package?
>>- -sandalle
> _______________________________________________
> Lunar mailing list
> Lunar at

More information about the Lunar mailing list