stéfan sinclair :: scribblings & musings

Matt Kirschenbaum has done many of us a favour by posting an entry on doing a Ph.D. in digital humanities (or an akin field). This isn’t anything like a Maclean’s ranking of universities, thankfully, it’s more of a compilation of schools to consider where a critical mass of researchers in the area makes it more likely that a graduate student might appropriate academic and intellectual support for pursuing their studies (and though more targeted programmes do exist at the Masters level, the post is useful for those considering an M.A. as well). There aren’t really any institutions that offer cleanly packaged doctoral programmes in digital humanities, but with a bit of resourcefulness, hybrid programmes can usually be designed. A lot of factors go into the choice of graduate programmes, but I think it’s a good idea for students to find researchers they might be interested in working with and contact them to see about the feasibility of a computing-oriented programme (it’s probably not a good idea to choose an institution solely on the basis of only one researcher).

The schools mentioned in Matt’s post and in the comments that follow are especially useful for those with a primary interest in English literary studies (and who are likely to study in North America). For a slightly different focus, it may also be worth considering an information studies type programme, such as UIUC, UofT’s FIS, or UCL’s SLAIS (among many others).

Writely is an online word processor that allows users to collaboratively create, edit and share documents. There are several (mostly Javascript-based) web-based editors that represent a new generation of web-based applications (see for example the FCKeditor). Writerly takes things a step further by managing the documents centrally, much like a wiki (whereas a wiki tries to simplify the syntax for styling documents – a tool like Writely hides the styling syntax altogether). It even allows you to upload an MS Word document to begin with. Writely looks fantastic, though I have to wonder what happens when it comes out of beta and starts charging money for use. It will also be interesting to see if they have any patentable technologies or if we’ll start seeing a lot of variants of this model.

Wired has a short but incisive article on the ongoing debacle of Google Print, the initiative to digitize and index books. Lots has been said about this, but I liked the point that indexing the web has already changed traditional notions of copyright:

Think about Google’s core business: It copies whatever content it finds on the Web and puts that content in an index. It doesn’t ask the copyright owner first, though it does exclude content if asked. Thus, Google wants to do for books exactly what it has always done for the Web. Why should one be illegal and the other different? Google creates value – a lot of it – by indexing existing content.

In any case, as the article concludes, this is unlikely to be resolved before lengthy and expensive litigation.

Wired has compiled an interesting (and admittedly subjective) list of the ten worst computer bugs in history. Part comic, part tragic… worth a look.

And an update with comments sent to the author.

A couple of more links of interest from the perpetual geyser of goodies at Infosthetics, with particular attention to “infoscapes”, or panoramic views of information that suggest different characteristics than when the information is viewed up-close:

• art collection @ de Young museum, which helps to see stylistic trends in art by showing showing some 100,000 works chrononlogically in a relatively small space
• bible exegesis which shows an overview of the most often cited passages of the bible; you can also search for particular words (see also Gospel Spectrum it’s remarkable how many cool visualizations are being created for biblical study, though of course theologians have always been pioneers in humanities computing)

See also the conceptually cool (though perhaps somewhat aesthetically underdeveloped) NewsQuakes.

Yesterday I posted an article about how easy it was to migrate from TextPattern to Drupal while keeping the same theme. Turns out I spoke a bit too quickly, as I hadn’t bothered to check how things look in Internet Explorer (why would I? oh yeah, market share…). In particular the thumbnail on this post shows how the navigational links cascade downward – not exactly the cascading style I was hoping for…

Students in my Web Programming course have been asked to develop their own assessment schemes for their projects; all groups chose a mix of marks assigned by me (according to criteria they specified) and self-assigned marks. The self-assessment aspect presented some challenges that we discussed in class, challenges that are to be expected given how rarely students get to assess themselves in a meaningful way.

After class I happened to be reading an article that referenced another article entitled Unskilled and Unaware of It: How Difficulties in Recognizing One’s Own Incompetence Lead to Inflated Self-Assessments. I was struck by the coincidence, though the article talks about unskilled individuals, which doesn’t really apply to my students, judging by the quality of the projects they submitted in a difficult Problem-Based Learning context.

In my Electronic Textuality and Programming class we’re doing an optional programming module using Ruby and Ruby on Rails. In thinking about which language to present I was torn between several possibilities, including:

• PHP: wide-spread, relatively easy to learn, some object-oriented syntax
• Java: wide-spread, relatively difficulty to learn (especially for people with no coding background), object-oriented syntax, better support for graphical interfaces beyond the browser
• Ruby: not very wide-spread, relatively easy to learn, object-oriented syntax, very good web framework with Ruby on Rails

I chose Ruby for three primary reasons:

1. Ruby seemed like the quickest way to get from learning the coding to developing the code
2. I’ve coded in a lot of languages and Ruby just feels right
3. the new version of my HyperPo text-analysis tool (not yet publicly available) is built on Rails and provides a good mechanism for students to write self-contained modules (or HyperPoets)

My main hesitation was that Ruby is not (yet?) very common, so it might not be as generally useful to students as, say, PHP (several of my students from last year’s web programming class have told me that they’ve already made extensive use of the PHP and MySQL that we covered). For that reason, I was heartened to read an article in C|Net (a widely-read tech news source) about the ever increasing popularity of Rails. In particular it quotes David Geary, author of several Java books:

Will Rails hit a tipping point and become widely adopted in the near future? I am certain of it; in fact, I think the tipping has already begun.

One point of view, but certainly very encouraging.

The content of this site is driven through a Content Management System (CMS) and I spent some of the weekend migrating from TextPattern to Durpal. It’s not that I was particularly dissatisfied with TextPattern nor was there a pressing need for additional functionality (as I mentioned before), but I’m an incorrigeable geek who likes experimenting with software and I’m also an accomplished procrastinator.

It’s nice to be able to switch systems and still retain the essential appearance of the site; this is thanks to strong theme mechanisms in both TextPattern and Drupal (the site is based on the Leaf theme, you can compare the appearance of this site through drupal and textpattern – though the last link may not stay available for very long).

Every CMS has its strengths and weaknesses: TextPattern has a much nicer editing environment whereas Drupal has a flexible module architecture that allows finer grain control over content access, organization and appearance. I actually played around with TikiWiki as well on the weekend, but became frustrated by various aspects of it (though it too has its particular strengths). As with many things, a blend of features from the different systems would be ideal: TextPattern’s administration interface, Drupal’s extensibility, and Moodle course management facilities. Having a mix of features means that each one is probably not as good as a specialized version (TWiki for wiki editing, for instance), but harmonizing the appearance and administration of a site makes the compromise worthwhile for me.

Dogpile is a meta search engine that compiles results from other engines (like Google, Yahoo!, and MSN).

I must admit that I haven’t used a meta search engine for a while, and I was surprised by how quickly it produced results. I thought that metasearch engines worked by querying other engines in real time (they receive a request, then send queries to each engine, then compile the results). But given its speed, Dogpile seems to also have its own index to draw on, despite the description on its site: “Every time you enter a term to search the Web, Arfie chases down the best results from the Internet’s top search engines.”

Anyway, I think the most interesting aspect of the site is a Venn diagram visualization that compares the results of different search engines. Fairly simple visualization, but visually effective.

Thanks to Andrew Vande Moere for mentioning Dogpile on his blog.