Re: [linuxdc-dev] VMU filesystem question

[M. R. B. <mr...@0x...>]

* Adrian McMenamin <ad...@mc...> on Sat, Dec 07, 2002:

> (a) Ban all interaction between vmufs and any other filesystem (ie assume=
 that=20
> all an any file is a vmufs file)
>=20
> (b) Ignore the problem because it is probably not too serious in any case
>=20
>=20
> (a) seems overly restrictive and I doubt whether it is practible in any c=
ase.=20
> I won't know about (b) until I've actually done an implementation and tes=
ted=20
> it.
>=20
> Have I missed anything? Any thoughts?
>=20

Heh, well, you already know what I think of (a) :P.

(b) is potentially harmful, but I can't think of a clean way around it.  If
we do find out that the browser will just abort (and hopefully not display)
broken headers then I think that's good enough - or you can create a dummy
header for each file that contains "Linux vmufs - < filename >" or
something.  That way you can still access files in the browser and have
them make sense to other DC software.

I don't think you can ever achieve full compatibility without breaking
access to other non-vmufs filesystems.

M. R.