A couple of weeks have already passed since the WebSci'12 WebSci-Teach workshop. This was run by Stéphane Bazan, Su White and Hugh Davis, and concerned how we go about teaching Web Science in practice. Reasons to talk about this include the youth of the WebSci curriculum, the barriers of teaching across disciplines, and issues such as people confusing Web Science with Web Technology.

Non-Web Science Web Science?

In the first two talks, David Molik described his experiences as a WebSci student at RPI, while Frank McCown described teaching CompSci undergraduates an introduction to WebSci. Topics taught included properties of the web, how search engines work, collective intelligence, recommender systems and Python. What really struck me at the end of these two talks was that neither courses seemed, to me, to be Web Science so much as Slightly Modified Computer Science. No one mentioned sociology, law, politics, geography, digital humanities, economics... A web-intensive CompSci course does not Web Science make!

This isn't to say I don't empathise with the difficulties of launching a Web Science course. Thus far, the majority of Web Science courses come from Computer Science departments, and CompSci is indeed foundational to WebSci -- but at the same time, you need that breadth of disciplines to achieve true Web Science. No Computer Science department can provide that alone. At the same time, it is no small chore to reach out to other departments and find technology-friendly researchers who are willing to spend time and money being involved.

(The flip side of this coin is also true, of course; for instance, Stéphane Bazan also presented in this workshop, on teaching at USJ Beirut. There, they have no trouble covering such areas as politics, trends and industrial geography -- the struggle is to find people to teach technical topics.)

Regardless, it strikes me as inaccurate to claim you're teaching Web Science when really, it's a mash-up of webby CompSci with a bit of Network Science thrown in for good measure.

Challenges in teaching WebSci (and how to overcome them)

Problem 1: That issue with disciplinary barriers. If you have WebSci PhD students, and are following best practice -- that is, ensuring that each student has a couple of supervisors from different disciplines -- then those students and their supervisors are one route by which you can get new knowledge to contribute to your undergraduate teaching. Link up your teaching and research.

Problem 2: The web and its complex ecosystems changes with each passing moment. How do you teach a topic that develops faster than you can teach it? You teach people how to teach themselves, of course. This reminds me of my CompSci undergrad, where we were taught a few programming languages (most notably, Java) in depth, but also taught such a range (and diversity!) of other languages (from scripting languages to Prolog to Lisp) that CS undergrads walked out from the degree with the confidence to pick new languages up and run with them.

Relatedly, Leif Isaksen remarked upon parallels with the field of Digital Humanities: there, you can reportedly find people who will strongly argue that you are -- or are not! -- a Digital Humanist if you can -- or cannot! -- code. Apparently the Methodological Commons from KCL helped DigiHumanities coalesce (or, as Leif put it, 'congeal'!) into a more coherent form. Clearly, we need to talk to Leif...

Problem 3: It's tough to teach Web Science when your students come (as you would hope) from diverse backgrounds -- how do you teach programming when some students have studied Computer Science, and others have not? The analogy I'd use is that of CompSci departments who offer broader courses, such as Southampton's IT in Organisations offering. ITO incorporates broader management aspects to the degree course, at cost to some technical elements from the straight CompSci course -- thus, ITO students have somewhat different needs to CompSci students. Perhaps Web Science leaders can pick up lessons from this area in the future.

In closing

The WebSci curriculum is much better defined now than it was at the WebSci-Teach workshop last year -- which is no surprise -- but there's still work to do. We need to pull together best practices in WebSci -- what are the tools, techniques, heuristics and frameworks? We need to talk to industry and find out what skills they want in WebSci students. We need to keep talking as a community, disseminate our knowledge, and promote best practice in Web Science education.

And we need to stop saying our Web Technology courses are Web Science.