It took me a while to thumbnail-ify it, but this is the poster that I submitted as the major assignment in last term’s SOCI209: Society, Technology, and Values (PDF) course.

Click to see the full-size ~380KB version. I don’t think I’ll link to the 8MB PNG original.

My goal in creating the poster was to show that the value we place on privacy is in a steady state of decline since the advent of digitized persistent storage — the Network — and increases only transiently through public shocks of violation.

I was never entirely satisfied that the poster’s narrative was sufficiently persuasive, and visually it’s too wordy, but I still think it looks pretty sweet. The glossy 2′ x 3′ version hangs over my desk at the College and never fails to impress…me.

I created it using Macromedia Adobe Fireworks, the only graphics program I’ve ever even partly glommed. Probably the trickiest bit was drawing the green circuit-board-ey border.

Of course, the day after I dropped it off at Staples to be printed, Google announced their acquisition of DoubleClick. One more slight upward blip in our value of privacy before the bottom drops out.

The Computer Studies department at the College set its second-year program of courses for the upcoming academic year. The normal slate would consist of five courses each term, but only three will be offered in the fall, and just two come winter. The reason? Only two of the sixteen students that started in September remain in the program.

The first-year program will again run full-bore, assuming someone registers.

The second-year of the program ran partially this past year, and not at all the year before (due to an alternate-year student intake). As someone who teaches second-year material exclusively — more an artefact of scheduling than advanced content — my full-time employment prospects are dim.

The problem stems from a simple lack of students. My first year at the College, I was spoiled by an exceptionally strong group of twelve. The number dwindled to five the following year, and just a couple since then. It does make for a superb student-teacher ratio, one that betters Oxford’s tutorial system.

There was, and still is, great hope for the recent accreditation awarded the program by the Canadian Information Processing Society (CIPS): graduates can transfer their two-year diplomas toward a computer science degree down south.

So where are the students? I can think of a few explanations (none substantiated):

• Alberta. Although our students can fetch about $46K upon graduating, that’s peanuts compared to someone with a pipe trades ticket in the oil patch.
• Whitehorse. There aren’t all that many IT employers in town, and the College may have simply saturated the market, particularly for network positions.
• The Tech Bust/Offshoring. Ever since 2001, fewer and fewer students have enrolled in IT programs, fearing that the jobs are gone and won’t come back.
• University Computer Science. High-school seniors with geek tendencies are very likely to enrol in a university program — go Outside and see the world a bit.
• Admission Requirements. Those that didn’t have the prerequisites for university, or returning mature students, tend to be put off by the math component.
• CIPS Accreditation. To meet their standards, the program had to incorporate more programming courses and eliminate electives. Traditionally, more students have pursued the networking stream. Programming increasingly seems to be an “either you have it or you don’t” type of endeavour.

What I can’t think of are solutions, so my imminent career plans don’t include much teaching. Oddly enough, I don’t seem to have any trouble finding programming work. What I can’t seem to do is train anyone else for these jobs.

Eugene Wallingford of the University of Northern Iowa writes about teaching computer science and — yawn — marathon running. A posting of his dating from June crystallizes my every thought on textbooks, especially those intended for CS courses. In Picking a Textbook for Fall, Wallingford writes:

I don’t like textbooks.

That’s what some people call a “sweeping generalization”, but the exceptions are so few that I’m happy to make one.

For one thing, textbooks these days are expensive…

…[one-semester textbooks are] written specifically not to serve as a useful reference book for later…

…I want to teach my course, and more and more the books just seem to get in the way…

…By and large, these books aren’t about anything.

[emphasis in the original]

Wallingford concludes by announcing that he has chosen a multimedia-oriented trade paperback for his Java-based CS1 course. For one thing, it should make the programming exercises more meaningful to the students, and for another, it will stay out of the way when he fills in the theory gaps with his own notes.

For my own courses, I either eschew textbooks entirely, or use them selectively for portions of the curriculum. And “textbook” isn’t the correct term, as I tend to choose trade books from either O’Reilly or Peachpit Press. These are much less expensive — I once inherited a database course with a $180 hardcover text that was all but useless — and will serve as references for students who continue in the field after graduation (we’re always going to lose a few to Alberta’s tar sands).

Teaching sans textbook does add to the instructor’s responsibilities: not only to provide sufficiently detailed lecture notes, but also to assign readings that address the topics in an alternate voice. And there is more effort required in developing a course without an accompanying text; we can assume that the writers of these tomes have expended a great deal of effort into the incremental organization of the material — that, and writing out the answers to the odd-numbered questions.

The textbook publishing industry plays an expected role: distributing free samples of the latest glossy hectodollar opus. As of the present, I haven’t been swayed by such enticements, but then with titles like Spacial Databases, Fundamentals of Database Systems, and Administrator’s Guide to SQL Server 2005, they may yet bore me into submission before the next academic year.

We’ve all enrolled in these: 3-day training courses on some fad software package or newfangled programming paradigm. I’ve taken ‘em. I’ve taught ‘em.

I’d like to create and teach more of these in the future, but only under the condition that I define what such a course should be in the first place.

Paying for a typical training course entitles one to periods of enforced playtime, snacks in the morning, and a nap in the darkened lab following a chicken luncheon. In other words, your $2,000 has bought you three days of kindergarten.

The true benefit of these courses is that you’re made to think about the subject material for a period of time in an environment with fewer distractions than the office. And you may even learn something, most likely disjoint facts without any organizing narrative or context.

And let’s not forget the laser-printed graduation certificate!

Leaving aside this ready-for-framing certificate, you would derive the identical benefit by reading a trade softcover on the subject and actually trying out the exercises listed therein.

My definition of a course is one that starts at that point: students have read the book and experimented with the concept before enrolling. The course is now less about spoon-feeding the basics and covering the installation two-step, and more about answering the questions that arose during the students’ experimentation and demonstrating the practical avenues for future experimentation. The instructor’s role is then to impart experience (something that can’t be learned from a book) rather than knowledge (something that can).

Naturally, my perfect course is nigh impossible: people shelling out for training do so precisely because they don’t want to have to read or learn on their own, organizations running such courses don’t want to forgo income by enforcing prerequisites, and the instructors must command a wealth of experience — instead of having simply read the book and been given advance possession of the curriculum materials — and therefore command higher rates.

Of the many folks that exit a traditional training course muttering, “That’s nothing I couldn’t have learned from a book,” there must be a niche demographic willing to pony up for something better.

We were exploring abstract data types one day in class, when a student looked up from his notes and asked, “How do you spell ‘queue‘?”

“Oh that’s easy,” I answered, “F-I-F-O.”

He failed to see the humour.