As I noted in my post over on Between the Lines ("Why doesn't IBM just buy Novell already?"), I've been testing OpenSUSE's Linux for Education Project and Ubuntu 10.04 server beta 1. I have a couple of problems that I was hoping they could solve: 1) a thin client environment that would allow me to redeploy several stacks of colored iMacs; 2) an easy Moodle setup that I could not only use for some pilots but share with colleagues looking to deploy their first LMS; and 3) determine if Ubuntu really should reign among Linux distros in education (and elsewhere).
Why use the early beta of Ubuntu Server? I wanted to evaluate whether it was as cool as the desktop version of 10.04 and I was working in a non-production environment, so why not? Today I did actually leave it behind and began using 9.10 to solve some problems with LTSP and Moodle, by the way, but overall, it's quite solid for the first beta.
At any rate, I was looking for the equivalent of the old Edubuntu distributions that included both a fully functioning and incredibly easy LTSP configuration and lots of great software for kids. LTSP (Linux Terminal Server Project) is now included on the Alternate CDs for Ubuntu and a trial platform is included on the Live DVD of Li-F-E. Actually getting it running in production on OpenSUSE? That requires a bit more configuration. Edubuntu? It's an add-on now that supplies quite a bit of educational software, much like the extra layers in Li-F-E. As with Li-F-E, the cool server-oriented software doesn't work as well out of the box as the desktop applications.
In terms of the "alternate CD" for Ubuntu, a simple F4 at boot is supposed to create a working LTSP server; graphical tools can be installed to manage the server after the fact. In 10.04, the second network card required to run LTSP wasn't recognized (apparently a fairly common problem according to the forums for the beta), so as I left this afternoon, 9.10 was installing. According to the Ubuntu documentation, I should be thin client-ready when I get back in the morning.
In a roundabout way, I'm getting at two big points:
- We need a turnkey Linux solution in education, a big ol' DVD from which installers can pick and choose features that are needed for a given situation.
- Imagine being able to choose if you want a working Koha, Joomla!, or Moodle server at install time.
- What if you just need an open source SIS, ready to begin populating with data?
- Or if you're just creating student desktops and need primary, secondary, or tertiary software?
- Both Edubuntu and Li-F-E are moving in this direction, but neither offers the functionality or customizability it should out of the box.
- This is the perfect example of why an educational appliance makes complete sense, both in developing markets where IT expertise is often lacking and in many developed markets, where staff simply need something to work out of the box. Check back later today to hear about just such a device that is in testing right now.
- In many ways, both Li-F-E and the Edubuntu add-on DVD are far better suited to rural or developing markets where Internet connections are slow or unreliable.
- Until a customizable turnkey solution or appliance emerges, many users with fast Internet connections would be better off simply installing base desktops/servers and using repositories to add the software they need. For example, check out this painfully easy walkthrough for setting up Moodle on an Ubuntu server.
- Users without such connectivity are probably better off tapping the resources of the educational projects where they can.
The deep repositories and incredible wealth of software available to school users means that the real winners (for now, at least) of the prize fight in the title of this post are the base distributions, whether Ubuntu, OpenSUSE, or any number of others. The education projects of Edubuntu and Li-F-E demonstrate the breadth of software choices, but administrators will often have better luck simply building their own educational software stack appropriate to a given situation.