[olug] VirtualBox?

[Rob Townley]

On Tue, Apr 1, 2008 at 4:04 PM, Adam Lassek <adam at doubleprime.net> wrote:

> That sounds pretty paranoid to me. GPL code can be forked, period. That's
> a
> basic feature of the license. It's also a lot of work, which is a
> perfectly
> good explanation why it hasn't been done.
>
> I've used every virtualization solution available in Linux, and would rank
> Qemu a distant third in usability, stability and performance. YMMV.
>
> On Tue, Apr 1, 2008 at 8:03 AM, Luke -Jr <luke at dashjr.org> wrote:
>
> > On Tuesday 01 April 2008, Adam Lassek wrote:
> > > On Wed, Mar 5, 2008 at 11:33 AM, Luke -Jr <luke at dashjr.org> wrote:
> > > > On Tuesday 04 March 2008, Brady Cox wrote:
> > > > > Anyone had any experience with this?  I'm installing Windows right
> > > > > now.  Wish me luck.
> > > > VirtualBox is anti-free software. (k)qemu works great and is free!
> > >
> > > This is not true. Virtualbox OSE is GPL2 licensed. Qemu can't hold a
> > candle
> > > to VB -- especially seamless mode.
> >
> > VirtualBox "OSE" is just a legal form of artificial restrictions.
> > VirtualBox,
> > complete, is not GPL2. The company also acts to prevent* any kind of
> > developer community from growing outside of their employment, so they
> can
> > effectively close even the GPL2 code at any time without much risk of a
> > fork.
> >
> > Basically, that GPL2 "edition" serves the purpose of killing off
> > competition
> > from free software like qemu.
> >
> > As far as a technical comparison between qemu and VirtualBox... qemu
> > supports
> > x86, ARM, SPARC, PowerPC, MIPS, and m68k-- including running an x86
> system
> > on
> > PowerPC or almost any other combination. qemu also supports user-mode
> > emulation, meaning you can run just one application instead of booting
> an
> > entire OS. VirtualBox has USB support too, but it is one of the features
> > that
> > is kept out of the GPL2 edition.
> >
> > On the other hand, VirtualBox does have "seamless" Windows support and a
> > nice
> > GUI. I don't use Windows, nor care overly much about GUIs, so qemu is an
> > obvious win to me. Maybe those things are more important for you. But in
> > the
> > long run, I think it'd be far easier to add "seamless" + GUI to qemu
> that
> > it
> > would be to rewrite VirtualBox's architecture to support the things qemu
> > does.
> >
> > Luke
> >
> > * No, I don't have actual proof they are actively trying to do this.
> > However,
> > I have heard from developers that in effect this is the case. Also note
> > the
> > list of external contributors has a total of ONE person:
> >        http://www.virtualbox.org/wiki/List%20of%20Contributors
> > _______________________________________________
> > OLUG mailing list
> > OLUG at olug.org
> > http://lists.olug.org/mailman/listinfo/olug
> >
> _______________________________________________
> OLUG mailing list
> OLUG at olug.org
> http://lists.olug.org/mailman/listinfo/olug
>



Are you really saying that QEMU with the kqemu driver enabled takes a
distant third when it comes to performance.  Have you verified kqemu is
enabled with "info kqemu"?

Tons of people have compiled qemu because it was made to be easy to port and
compile.  If VirtualBox is hard to fork, i would guess only because
impediments and lack of docs have been put up to make it hard.