stéfan sinclair :: scribblings & musings

BumpTop is a project, based at the University of Toronto, to experiment with metaphors of the electronic desktop. The underlying principle seems to be to make the electronic desktop resemble as much as possible the physical desktop, with principles of physics respected where appropriate (like the interaction between items), and organization strategies available where possible (like turning items at an angle to reveal a prominent feature). The features demonstrated in the video show extensive sophistication in the interface, arguably too much so for many users. I also found myself several times thinking “yeah, but it should also…”, though on the whole the video did eventually anticipate many of my would-be suggestions (like better representation of the contents of items, rather than their format). I wonder too about how Apple legal feels about the use of the word Piles.

But the only real question of substance I was left with in the end is whether or not the top of a desk is really the most useful/efficient metaphor for working with computers. I’d like to see an operating system make available a set of different metaphors for users to choose from (and not just themes – or superficial variants – of dominant metaphors). In any case, at the end of a stimulating week talking with colleagues about ebooks at the Digital Humanities Institute – with particular attention to the differences between models and simulations – I find BumpTop an exciting project.

Thanks to Stan Ruecker for sending me a link to BumpTop.

ZDNet has published a list of zingers by Scott McNealy, the infamously feisty CEO of Sun Microsystems, who is stepping down after 22 years. There are definitely some gems in the list of McNealyisms, including many pointed jabs at Microsoft, but the one that cracked me up is this one: Open Source is free like a puppy is free. I’m a strong open source proponent and a dog owner, and somehow I think this beautifully captures some of the oft-overlooked economic nuances of using open source software.

Though multilingualism in both natural and artificial languages is always a good thing, one of the drawbacks of regularly using multiple programming languages is the confusion between the syntaxes of each one. Now was that substr(string, start, end), substring(string, offset) or sub_string(start, length, string)? Even if one’s not confused, what better way to better learn a language than through browsing its API? gotAPI has provided a useful API reference service for multiple languages, including HTML, JavasScript, CSS, XML, MySQL, PHP, Java, Perl and Ruby. All of the content seems to come from elsewhere (such as W3Schools), but this is a convenient aggregation of information, with a slick AJAX-based search mechanism.

WordHorad bills itself as an application for the close reading and scholarly analysis of deeply tagged literary texts. It’s a Java-based Web Start application (that can run just about any platform). This tool is “modern” (XML-aware, unicode data, etc.) and provides an impressive array of functions (searching, counting, concording, comparing, annotating, etc.). It works with a large number of texts that are pre-indexed (early Greek epics, Chaucer, Spenser and Shakespeare), and allows developers to add new texts as well. Texts are richly tagged to allow very flexible analysis (tags for morphology, narration, prosody, metre, dates, etc.). And it appears to be extremely well-documented. I think this text analysis tool is exemplary in many ways.

Worldmapper is a collection of world maps, where territories are re-sized on each map according to the subject of interest. This is an excellent example of data visualization that plays on our existing knowledge by warping representations of the world (though of course every two-dimensional representation of the world – like the Mercator project is already a warping (in fact, any representation of anything at all is already a warping)). The thumbnail map shown here is excerpted from a map of the total elderly by region – note especially the proportions of Europe and Africa.

Thanks to infosthetics.com for the link.

Generally speaking, I’m not terribly fond of jokes forwarded by email, but I caught myself laughing several times at a recent message sent to me by Claire Campbell. I thought in this end-of-term period (at least at McMaster), with deadlines looming for final projects and imminent exams, a bit of comic relief might be welcome. Rather than forward the email, here’s a link (if you haven’t seen it already) to a page that hosts Technology for Country Folk.

The New York Times reports on Amazon’s Text Stats tool in an article entitled Book, How Do I Love Thee? Let Me Count the Words (requires registration, try Bugmenot). I’m always delighted to see mention of literary text analysis in mainstream media, though it’s of some concern (in terms of public perception) when the computer tools discussed are so simplistic – even if the article does recognize this simplicity: “... the crudest sort of computer analysis of a book: how many big words there are, and how long the sentences run.” Also of note is Amazon’s interest in such tools which stems from the fact that it seems to keep “readers at the site longer comparing and contrasting books.”

Richard Gilles has a very interesting post on corporate blogging and its apparent benefits. There are tons of articles out there on the hype surrounding blogs and some of their potential roles in the corporate world, but this post has substantial historical and humanistic dimensions. Gilles succinctly describes the development of recent waves of technologies, arguing that there’s been a movement toward “sharing, networking, arguing, collaborating, and the search for greater freedom.” Though this starts to sound a bit like the over-zealous enthusiasm of hypertext theorists early on, I do think he makes some valid points about the promise of blogging. More importantly, though his post is focused on corporate blogging, most of what he says is transferable to other types of blogs, such as academic research blogs or even personal blogs. Here are the main points that he makes (but it’s well worth reading the article for more details on each point):

• Some Reasons For Companies To Blog.
  • Because the internet is humanising, and blogs help humanise a company.
  • Because the blogosphere helps keep your finger on the pulse of the industry, and being a part of that sphere adds credibility.
  • Because consumers are smarter today. E.g. – We’re sick of corporate speak. We can see right through it.
  • Because it means you can communicate with your communities unfiltered, and it promotes instant feedback.
  • Because it builds a network.
  • Google [ranking].
  • Because you become the expert.
• Some Major Points On How to blog.
  • Create a Policy.
  • Post often, post fast.
  • Be truthful.
  • Don’t tell secrets.
  • Be human.

Thanks to Willard McCarty for pointing this out on Humanist.

Danny Sullivan has posted an interesting article entitled 25 Things I Hate About Google. In the wake of a continued flurry of acquisitions, including Writely. and SketchUp, the question is whether or not the venerable search engine can continue managing so many tools (and offering most for “free”), particularly when there are long-standing issues (or bugs) with several of them. The dizzying breadth of activity is also in disaccord with Google’s corporate philosophy that “it’s best to do one thing really, really well,” as pointed out in the article. Most of Sullivan’s points are of course debatable, and he does offer in counterpoint 25 Things I Love About Google, but it’s valuable to see a reasoned criticism of (arguably) the world’s dominant online knowledge broker.

Oliver Steele has a blog post about his reAnimator tool, a finite-state automaton (FSA) visualizer for regular expressions. You don’t really need to care about FSA to find this useful, it can be a good way of debugging a regular expression to determine where a match is unexpectedly failing. The reAnimator interface is a bit clunky and the visualizations are fairly simplistic, but this is a great pedagogical tool for regular expressions.