Re: gnome (gconf) saga - please read
- From: Havoc Pennington <hp redhat com>
- To: "Timothy H. Keitt" <tklistaddr keittlab bio sunysb edu>
- Cc: gnome-devel-list gnome org
- Subject: Re: gnome (gconf) saga - please read
- Date: 03 Mar 2002 21:14:10 -0500
"Timothy H. Keitt" <tklistaddr keittlab bio sunysb edu> writes:
>
> Obscure, to say the least. Why can't gconf figure this out itself?
>
GConf has no more information than the kernel does - if the kernel
says the file is locked, GConf has no basis to second-guess. If it
did try to second guess, then it would make mistakes. And that leads
to silent config corruption, rather than the obvious "config doesn't
work" situation that happens now.
So I think the best we can do is ask the user to intervene manually,
and fix the OS bug.
> In this case perhaps, but averaged over many such experiences, I would
> still have to conclude that gconf (or whatever mechanisms are being
> employed) is not very robust. I also get concerned when I see things
> like the galeon developers saying 'we have no idea why this happens
> (gconf looses settings), but try "kill something-or-other" and hope it
> works'. Doesn't speak to a transparent or robust design.
I don't think the Galeon FAQ is very up-to-date or accurate on this
point.
Havoc
[
Date Prev][
Date Next] [
Thread Prev][
Thread Next]
[
Thread Index]
[
Date Index]
[
Author Index]