Bring fuse to the desktop

[Sanel Zukan]

> That's great (having fuse integrated in gnome/kde) but I was more
> thinking of a way to standardize fuse filesystems to be able to use
> them anywhere.

Standardizing OS features is a nightmare, especially some 'relatively' new
stuff like FUSE, because OS developers tends to:

 a) be often agnostic to desktop 'high tech' needs
 b) be slow to incorporate replacements/proposals
 c) tries to make balance between server and desktop needs

For example, just look how long was needed to bring descent notifying
support in linux kernel (via inotify replacing hellish dnotify). And
that was just a story for one OS.

AFAIK now only linux officially supports FUSE (there is also project for
FreeBSD), but what about other unix-es? 

IMHO, first step in standardization would be to let them know that we 
can't live without FUSE, just as we can't live without networking support;
then wait few years things gets stabile and then try the same thing with
desktop environments (althought desktops will addopt it much smoother
and faster).

> I would say any OS implementing FUSE, but the original idea was to be
> able to use fuse filesystems without using the command line on any
> desktop environment that support freedesktop standards.
> ...
> The idea is to be able to use fuse filesystems transparently on any
> desktop environment that whishes to do so. And have the ability to add
> support for others fuse filesystems without having to recompile or
> upgrade my desktop environment in any way.

And roots for this are in above comment. I'm not sure how many users
would like to see desktop supporting one feature only on one specific OS.

Developers struggle with these differences last I-don't-know-how-many
years (sound, notify, etc.), but at least we had same or similar things
implemented on other way (e.g. inotify on linux, kqueue on BSD-s).

Unless there is FUSE-like implementation on other OS-es, IMHO, there is
no point for standardization, at least in desktop land.

Best,
--
Sanel