Backend Information-Technology Working Group

The primary focus of this Backend-IT Working Group is to care for the hardware and software infrastructure that comprises really the main physical incarnation of a "virtual" university. This group is specifically charged with running the website, mailing lists, databases, etc. In keeping with the University's strong ties to Free Software, this working group also makes sure that all software is compliant with the DFSG (Debian Free Software Guidelines).

Our current implementation is as follows:

Operating System: Debian GNU/Linux 3.0 (DFSG)
Web Server: Apache 1.3.26 (Apache/X11/BSD)
URL Link Checker: LinkLint 2.3.4 (GPL)
Mail Transport Agent (MTA): Postfix 1.1.4 (IPL)
Mail List Manager: GNU Mailman 2.0.11 (GPL)
Internet Relay Chat (IRC) server: Dancer IRCd 1.0.31
Structured Document Markup: LaTeX2e (LPPL), TeX 3.14159 (PD), and Docbook XML (BSDish)

Looking ahead

I thought I'd throw some ideas down here that were circling around in my head for technical additions in the very-distant horizon that might be worth thinking about (sooner even, if anyone is really interested and decides to take the ball and run with it). They could have potential use to this University Project, and I would of course encourage their author(s) to release them to the general Free Software community, under an appropriate DFSG license such that we could use them. Some of them aren't even software, but instead just coordinating with other technologies. Here goes nothing (except a little daydreaming):

Obviously, we plan on providing all our course material online for free under an appropriate license, to be determined. We will not only be providing HTML-formatted course materials, but also a number of camera-ready output formats. These are intended for students to be able to print out on their own. However, I think it would be wonderful if we could somehow coordinate with appropriate companies to make available very-low cost professionally-bound versions of these works.
Barnes and Noble might have something available with their new print-on-demand machines appearing in some urban stores. It might be worthwhile also to stay abreast of the Print on Demand Journals. Obviously, they are are professionally-bound versions of Free Software manuals, published by New Riders, Coriolis Open Publishing, and others. However, they're still quite costly in objective terms. I would hope that we might be able to make this happen at the cost of the publishing, maybe something like the cost of a loaf of bread per book (purposefully choosing a commodity to base comparison as it enables appropriate ideas about cost, regardless of inflation and national origin; the $1 we might think is negligible would be wildly expensive in some foreign nations).
We could further the reach of our course materials beyond professionally-bound versions by enabling free audio renditions where appropriate (probably much more so for non science or engineering texts), using some current or to-be-developed text-to-speech synthesis software. We could then master ISO 9660 (CD-ROM filesystem) images for students to download and easily burn onqto CD-ROM to have the equivalent of audio books. We also might have to produce some script to convert LaTeX (or DocBook, etc.) formatting into the SABLE speech synthesis markup language recently proposed (should it end up in software we use for this). Some meaningful references for this (and I don't think it should at all distract us from a real job, which is to teach and develop course material) are:
- Carnegie Mellon University Speech Group
- Festival Speech Synthesis System, developed at the University of Edinburgh
- The Festvox project builds upon the Festival system to allow a more systematic and better documented creation of new synthetic voices.
- MBROLA project, initiated by the TCTS Lab of the Faculte Polytechnique de Mons in Belgium, is useful for links and background only (they produce a no-cost, but non-DFSG-compliant synthesizer).
- Sound and MIDI Software For Linux Speech Synthesis Links.

Obviously, we'd encourage you to suggest further forward-looking technologies that could improve the educational process. We tend to think that the idea of online-video lectures is not the best way (scalability, cost, nationalization, etc.) to go about providing educational resources and collaboration, even though the corporate learning companies seem to love it.

Current members:

Daniel Freedman