Re: Hurd translators on FUSE

[olafBuddenhagen]

Hi,

On Fri, Sep 25, 2009 at 08:23:57AM +0300, Sergiu Ivanov wrote:

> And I think I can remember somebody from the Hurd community having
> tried to show the advantage of the Hurd way and ran into an attitude
> like: ``Why do we need a Hurd, if we already have FUSE?'' :-(

Indeed this is a problem -- translators were the most visible advantage
of the Hurd archtecture; and now that FUSE covers most of the obvious
use cases, it's much harder to sell then before...

In my latest Hurd talk, I didn't even try presenting translators as a
feature on its own. Rather, I just explained them as an element of the
extensible VFS concept -- which is part of the Hurd's fundamental idea
of a user-extensible system environment. (And the one I covered in most
detail of course, as it's the most evident part, with the most examples
available.)

I mentioned FUSE only as one example of how things that require
extensible kernel support on other systems, can be implemented as simple
unprivileged userspace components with Hurd's extensible VFS -- along
with other examples like LVM, loopback devices, unionmount, and
namespace-based translator selection.

While one or two people told me that the presentation was interesting, I
have no idea whether I was able to convince anyone about the advantages
of the Hurd architecture... Unfortunately most of it is very abstract --
we definitely need more tangible examples :-(

-antrik-