Remaining issues for the multi desktop DVD

[Toshio Kuratomi]

On Wed, Jan 12, 2011 at 11:49:39AM -0500, Jared K. Smith wrote:
> 2011/1/12 "Jóhann B. Guðmundsson" <johannbg at gmail.com>:
> > The showcase ( in this case Multi Live *DE DVD ) should reflect the
> > final product ( tried and install Chosen Live *DE  ) as much as possible.
> >
> > Each *DE has it's own display manager thus the user experience would
> > differ ( and cause disappointment from the end users ) from "tried and
> > tested liked" to "chosen to install" ( since he would expect the same
> > display manager he was presented with to begin with ) however this would
> > make sense to me to do if all the *DE shared the same display manager.
> 
> There are definitely pros and cons to each method.
> 
> First, the current method that Christoph is proposing, where we put
> the existing LiveCD ISOs on a DVD and use a syslinux menu to select
> the desktop environment.  It has the advantage that it's already been
> done and tested, it uses the existing LiveCD ISOs so there's not a lot
> of additional QA work needed ("If it boots and presents the menu, it's
> going to work!").  The disadvantage is that we end up with multiple
> copies of the kernel, glibc, etc. on the DVD -- one for each desktop
> environment.
> 
> A second alternative that was discussed earlier was to have one LiveCD
> image on the DVD which had pre-configured users -- one user for each
> desktop environment.  This has the advantages of having much more room
> on the DVD for additional applications.  One drawback here, however,
> is that we'll have to choose *one* display manager (GDM, KDM), which
> obviously isn't ideal.  Another drawback is that since this is
> essentially a completely new image, it'll require more testing from
> QA.
> 
> Taking a step back and looking at the problem that Christoph is trying
> to address (having *one* disc that can be handed out at events for new
> users to be able to sample the various desktop environments and
> spins), I think either of those two alternatives will do fine in the
> short term.  Method number one is already basically done and requires
> less preparation, so I propose that we do that for F14 media, as soon
> as possible.  This will hopefully allow our European ambassadors to
> have media produced in time for FOSDEM.  For F15 and beyond, I'd love
> to investigate a more elegant solution, but I don't want perfection to
> get in the way of us making progress now.
> 
+1 to just getting things rolling.

Changes to the basic approach, I think, should look at who the spin is
targetting  If it's people who want something like a desktop-as-an-appliance
continuing the whole image route may make more sense than creating an
integrated live image.

-Toshio
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 198 bytes
Desc: not available
Url : http://lists.fedoraproject.org/pipermail/advisory-board/attachments/20110112/b00e9a3e/attachment.bin