On Sat, 2010-12-25 at 11:56 +0100, Josselin Mouette wrote: > Le samedi 25 décembre 2010 à 10:40 +0000, Emmanuele Bassi a écrit : > > > Well, none of the currently available virtualization solution does it > > > correctly. Now what? > > > > incorrect: GL passthrough in QEMU allows to run the MeeGo netbook user > > experience, which is based on Mutter and Clutter, exactly like > > gnome-shell: > > > > http://wiki.meego.com/MeeGo_SDK_with_QEMU#Setup_QEMU-based_MeeGo_Emulator > > Given how GLX is designed (in short: giant root security hole), that > basically means giving root access on the host to any user on the guest.
and, as I said in the rest of my email, virtualization is only interesting for testing/evaluation and development purposes. > On a related but different side of affairs, good luck getting GNOME > Shell to work on a thin client, as well. As soon as you have a bit of > network latency, AIGLX is not usable, and other solutions like VirtualGL > are merely insecure hacks to cope with the absence of accelerated > offscreen rendering support on Linux. I do wonder how Dave manages to work with Compiz on thin clients, then. in any case, for those set-ups the rule of upgrading with the *next* stable release of a long-term support distribution holds. we don't need to cover the entirety of use case scenarios right out of the door of GNOME 3.0: it would be a recipe for disaster and for indefinite delay. the core has already been identified, and the goal is to provide a kick-ass user experience for that core of users. spreading resources we don't have to cover use cases that are not well defined and, more importantly, that are not interested in tracking bleeding edge is a waste of time. ciao, Emmanuele. -- W: http://www.emmanuelebassi.name B: http://blogs.gnome.org/ebassi _______________________________________________ desktop-devel-list mailing list desktop-devel-list@gnome.org http://mail.gnome.org/mailman/listinfo/desktop-devel-list