F9 Alpha spinning
Tom "spot" Callaway
tcallawa at redhat.com
Mon Jan 28 01:32:51 UTC 2008
On Sun, 2008-01-27 at 03:18 +0000, Kevin Kofler wrote:
> Horst H. von Brand <vonbrand <at> inf.utfsm.cl> writes:
> > > Well, your arguments are pretty unconvincing, and apply to a lot of
> > > application-library relationships, still usually library users won't fork a
> > > library just for that, or when they do, Fedora won't ship their fork.
> >
> > AFAIU this is an /upstream/ decision, not a Fedora one.
>
> It is an upstream decision to ship that fuse-lite fork in the tarball and to
> support it. It is a Fedora decision to actually use this fork rather than
> using --with-fuse=external. (In many cases, we even _patch_ upstream projects
> to use system libraries where they don't support it at all, in this case, it
> would just be a configure option.) I'm complaining about both.
The ntfs-3g spec file is currently conditionalized to use fuse-lite by
default, because:
A) I'm concerned that future ntfs-3g specific functionality in fuse-lite
won't go into FUSE.
B) The upstream FUSE lead developer thinks that fuse-lite is a good idea
in ntfs-3g.
C) I'm comaintaining FUSE in Fedora, so I should be able to handle any
security issues that might need to be doublechecked.
With that said, I'm not entirely convinced that fuse-lite is the right
decision. I might change my mind, and make the default be
--with-fuse=external, but it will be still conditionalized so that the
SRPM can be easily rebuilt to the opposite.
~spot
More information about the fedora-devel-list
mailing list