On Friday, 13 January 2012 18:22:33 UTC+8, Emil Widmann wrote: > > > Do you mean that one cannot run Virtual Box without X11? > > I'd imagine if it were possible one can just run the sagenb server > there, > > and > > use the browser on the host system only. > > The discussion goes in circles ;-) > Running the Vm without X and accessing it from the host-browser is the > current solution. >
Really?! I thought I saw the browser running inside the VM window... Can't check this right now, perhaps it was just a coincidence that the host system browser had neatly placed itself over the VM window... > > If networking/port forwarding from VM to host would work reliably or > even if one would have good testers base and docs about what can go > wrong and how to fix it > then headless mode and autostart of the notebook in the host-browser > would be a good solution. > > Problems start when something goes wrong. In headless mode there is > even no visible Linux terminal window. > If if a terminal is brought up then the user is left to the linux > commandline and - probably - get lost. > Example: > http://groups.google.com/group/sage-devel/browse_thread/thread/a4808f2cf5b8b79f/03db2e576b21f42e > > Hmm, reading this again the missing terminal in headless mode is > probably an asset. > > In my personal opinion the work follows 3 tangents: > 1) Docs! > 2) Integration and support from the windows side , e.g. start- > scripts&gui, menu entries, generation of desktop icons, diagnose > scripts - e.g. network connection/firewall, even eventually combined > installer "VirtualBoxOSE plus Sage VM" > 3) Two virtual images, with possibly complete capabilities (e.g. X > +browser, GuestAdditions, graphics in R, JRE, Matplotlib graphic > backend, etc.) which can be started either in server mode (headless) > with autostart of the host browser, or in VirtualBox seamless mode. > One of the sage images is basic, 32/bit, compiled with option FAT > Binaries and 1 processor. The other is 64 Bit and more demanding on > the Hardware. > > I tried Volkers VM, but unfortunately it is to heavy to run on my > machine properly (yes, only 1 GB RAM). I also tried to improve my own > Puppy-VM with the Guest Additions, but I failed to get the mouse- > pointer integration to work - tough luck. Personally I will not have > time to work on this a lot over the next several months, and my > opinion is, that the progress is determined if a group of people on > the windows side is motivated to help, be it testing or even > developing. This could be the hardest part, because most windows-users > expect something working out of the box and give up rather quick if > something behaves unexpected. The very few "bug reports" I got mostly > were as detailed as " A is not working and B is not working and I > wasted 15 min on this crappy software". Still better than no feedback > at all ... > > Since it the whole is technically nontrivial but definitely not rocket > science I wonder if this would be a possible student class project of > some Computer Science / Engineering course? > > just my 2 Cents > emil > -- To post to this group, send an email to sage-devel@googlegroups.com To unsubscribe from this group, send an email to sage-devel+unsubscr...@googlegroups.com For more options, visit this group at http://groups.google.com/group/sage-devel URL: http://www.sagemath.org
