GAL vs gnome-2.0

[Jody Goldberg <jgoldberg home com>]

On Tue, Sep 18, 2001 at 03:29:12PM -0400, Havoc Pennington wrote:
> 
> eel is just some random chunks of Nautilus that have been factored
> out, and gal is similarly just some random chunks of the office-suite
> type of apps that have been factored out.
> 
> This is not _bad_, it makes lots of sense to do this as we hack on the
> office suite and so forth, but people need to understand the reality
> of the situation. These are "expediency libs" not "supported platform
> libs."
> 
> If this causes problems for people compiling/installing stuff, the
> solution is that people should stop using these libs. As expediency
> libs they should be used only by the target "insider projects" such as
> Nautilus and GNOME Office. Other people who need the features should
> consider cut-and-paste or workarounds in the short term, and getting
> patches into the supported libs in the long term.

I partially agree with you.

gal/eel in their current forms are not suitable for use as platform
libraries.  They have been forced into existence to serve 2 needs.

    - A place to provide application utilities
    - A staging area for new development.

The gnome platform suffers from a lack of coherent shared
application utilities.  Things like some centralised locale
utilities for date/value formating/parsing,  GnomeRecentlyUsed,
plugin systems, etc.  These have been reinvented for every major
application.  IMHO that is due to the stagnation of core.  There was
nothing perfect enough to go into gtk, and noone was adding things
to libgnome.

I believe Michael Meeks is in the process of porting gal to 2.0.
That does not mean it should be part of the platform.  However,
immediately after 2.0 when we start to look at what the targets are
for 2.2 gal/eel would seem to be _very_ fertile hunting grounds.  In
the mean time I would strongly advocate that application developers
make a conscious effort to move some of their non-application
specific utilities into these libraries.  Chasing their requirements
is admittedly difficult, but they patch a necessary hole in our
platform.  If we want to avoid the current problem where libgnome
development only starts when gtk freezes, then gets cut off due to
time constraints we need fluid libraries like gal/eel to work in.
However, these libraries are only truly useful in the long term if
their functionality is fed back into the platform.

These libraries are not strictly for insiders.  They are for anyone
developing a high level application.  Until libgnome (and brethren)
offer a larger set of application utilities.