system and desktop VFS merged

[rosen georgiev]

Diego Calleja wrote:
 >
 >El Mon, 28 Mar 2005 17:14:03 +0300 (EEST),
 >rosen georgiev <dewie at abv.bg> escribiу:
 >
 >> Diego Calleja wrote:
 >> > The idea of /vfs/protocol is just against the idea of userspace filesystems. I mean, the
 >> > whole purpose of a userspace filesystem should be being *completely* transparent
 >> > If you're going to do /vfs/protocol, what's the point in using the filesystem namespace for
 >> > it, except for tyding it? What would be next, require the path to be in valid utf-8?
 >> 
 >> but if its ~/vfs 
 >> (the idea is from)
 >> Lars Hallberg wrote:
 >> > It would be cool if stuff i conect to over VFS turns upp in the 
 >> > filsystem to for old apps and tool like find. But I realy expect them to 
 >> > turn upp ass:
 >> > ~/vfs/protocl/server/path
 >
 >
 >The idea of putting "vfs" in the path name is not good, IMO. The _whole_ purpose of 
 >"userspace filesystems" is to be fully transparent for apps, and
 >"vfs/protocol/server/path" is hardly "transparent". The whole point of a userspace
 >filesystem is having a "~/music", "/randomstuff/" or whatever without caring about
 >what program is behind or who is providing it.
 >
 >What we have now is:
 >kernel-libc <-> "path name" <-> apps
 >
 >What people here is suggesting is:
 >kernel-libc <-> "path name" <-> apps
 >|
 >|---> additional layer
 >
 >And the Right Way (TM) of doing it is:
 >kernel-libc <-> "path name" <-> apps
 >|
 >|---> userspace filesystems
 >
 >IOW: apps get the cool stuff without knowing it.
 >I mean, people is suggesting to create this standard, and then if an app wants to be aware
 >of the functionality provided by this standard we need to modifiy _every_ app on the planet,
 >from gnome/kde to apache. Path name involves everybody, not just desktop.

nooo, why should even one program be rewriten to work with ~/vfs - the
idea of it is to make cooperation between apps using D-VFS and all the
rest transperant to the user. and if you want ~/vfs can be replaced
with a environment variable eg. the user sets $VFS_PATH=/mambo/jambo
and mounts FUSE-DVFS there.

 >
 >What's so bad about doing a _good_ design and implement this behind apps, so they
 >don't need to know anything at all about files being implemented by kernel or userspace
 >programs? Creating a alternative vfs different from the real one, which needs to be
 >implemented by every app is a quite ugly design, as far as I can see....

i'm not getting something - is this about "no URIs for D-VFS and 
everyting under /"?

rosen


-----------------------------------------------------------------
http://gbg.bg/search - Изпробвайте още сега най-добрата българска търсачка!