Re: filesel dialogs

[Sean Middleditch <elanthis awesomeplay com>]

On Thu, 2001-12-13 at 04:23, crispin iinet net au wrote:
> 
> On Mon, Dec 10, 2001 at 10:26:26PM -0500, Sean Middleditch wrote:
> > 
> > Damn, that thing rocks, from the screenshots.  I doubt it could end up
> > being considered the "standard" file selector for GNOME apps for 2.0,
> > tho, hrm?
> > 
> > Sean Etc.
> 
> The question should arise if the new file-selector doesn't become the standard for 2.0 as to why? In an open source environment, *theoretically* their are no impediments to the best code bubbling to the surface (darwinian development). If it doesn't bubble, whos stopping it?

I think the reason would be bureaucracy.  In a one man/small team
development project, each member can make major decisions.  In GNOME,
you have to run thru so many hoops, *plus* take into consideration that
you have to worry about compatibility with other applications.  Yes,
with GNOME 2, compatibility may seem a funny term at the moment, but to
add *another* interface would just make life more difficult for
application porters.

Plus, this new file selector doesn't seem to be done.  That never helps
any.  ~,^

> 
> I agree with the sucky nature of the gnome file selector.
> 
> One of gnomes probs as I see it is it doesn't follow the release early release often. Thus the bugs aren't getting squashed. Thus the great code is getting overlooked.

KDE does the release early, release often, and it's debatable how much
its helped them.  The few areas where I have to agree that KDE is ahead
GNOME will easily make up in a year, which is about how much of a head
start KDE has (or am I wrong with the history here?).

Also, GNOME doesn't really happen "in concert."  There are 10 billion
independent project, each writing their own support libraries and so on
(can you say gal?  eel?).  These independent projects do release early,
and often.  But trying to get them all together at once is a pain.

That is why I still maintain that these support libraries, which are
used by major GNOME applications, should be merged with the base
libraries.  All UI bits in gnome-ui, non-ui bits in their respective
places, etc.  OAF, bonobo, all these other libraries that all have to be
installed together anyhow to work properly, should all be merged, and
released in concert.  If the teams worked together, it might actually
work.

Granted, this will never happen.  Every time I or someone else has
brought this up, the respective maintainers have complained about how
they want to release separately, be allowed to do their own thing, etc. 
So, we end up with many many little bits and pieces that barely work
together, and every now and again the developers have to put months of
effort into making one big release.

GNOME 2 I expect to be great and all with all this effort going into it,
but for GNOME 2.1/2.2, I really hope the focus is *integration*.  Merge
everything that has no good reason to not be merged other than
ego-trips, get rid of deprecated/duplicate features that just end up
making apps look/feel different (we'll only need *one* file selector in
the future).  If nothing else, it'll make application developers' lives
a lot easier.

> 
> Ego is ugly and the G in GNOME stands for GNU.

Given my opinions of GNU, I'm not going to comment on this one.  ~,^

> 
> Kind Regards
> 
> Crispin
> _______________________________________________
> gnome-devel-list mailing list
> gnome-devel-list gnome org
> http://mail.gnome.org/mailman/listinfo/gnome-devel-list