Media Technology

My brother is finishing up his Masters of Fine Arts at SUNY Buffalo, and just recently debuted his main dissertation project: a 20-minute experimental film about Eadweard Muybridge called tesseract (downloadable here).

Even to my relatively untrained eye it’s a beautiful piece (it just won the award for Best Photography at the Jutro Filmu international film festival in Warsaw), but the part that’s most interesting to me is how he’s applying ideas from Scott McCloud‘s Understanding Comics to film. Take a look and see for yourself.

I’ve run into a few interesting models for information flow-control in the past months:

• The Chronicle of Higher Education gives paid subscribers full access to their archives and lets them email / blog links to those articles which expire after five days. That gives them the advantage of word-of-mouth draw to their articles while still limiting the number of people who feel compelled to cut-and-paste repost their articles on their own sites.
• The online museum site Rhizome.org lets you follow any link from an external website into their site (for example, this one) but any link you type directly into your browser, follow from email or click from their own site redirects you to a “you must become a member” page. (They also open their site to non-members every Friday.) They’re just redirecting any request that either has no referral header or that has Rhizome.org as the referrer.
• I should also mention how the New York Times allows blogs to create non-expiring links to Times articles (that is, they don’t become subscriber-only or pay-per-view after a week). Unfortunately, the link generator over at BlogSpace is down at the moment (as is the rest of BlogSpace), so I can’t show an example…

All three cases are creating two classes of info-users: people who can disseminate their information (paid subscribers, plus people who read the Times when it’s still fresh) and consumers who can read what the first class of people point to but can’t go further without paying. While it’s a little Tom Sawyerish to essentially force people to pay to drive potential new subscribers to your site, I can see the basic model working if the balance is gotten right and fit the content well. At the very least, it’s nice to see models that are more subtle than the all-free / ad-based / subscriber-based triple that’s so common today.

Concerned about the influence Google‘s PageRank algorithm has in determining what information people see? Think that ranking pages by how many people link to them isn’t objective so much as automated mob rule? Want a search engine for people who don’t want to just follow the herd, or just want to see the dominant paradigm get a little more subverted?

If so, then you’ll be interested in Shmoogle, the Google-randomizer developed by Tsila Hassine. Shmoogle forwards your query on to Google and then randomizes the results, presenting them in the same no-nonsense interface you’d expect from Google along with the original rank of each result. Shmoogle is more of an art project than a practical alternative to Google, designed to encourage us to question whether “everyone else thinks this is good, so you should too” is really the best assumption upon which to base the library of first resort. Random order is at least diversifying, but to me it seems so arbitrary — and has me thinking of all sorts of alternatives:

• Dictoogle: rank sites in the order the elite think you should see.
• Diversoogle: give higher rank to those that have the fewest incoming links, in an effort to increase diversity and raise up websites less fortunate than the more popular ones.
• Novegle: give higher rank to sites that you’ve never seen before, in the interest of keeping up with the now. (I was sure I found it using this query just yesterday!)
• Capitoogle: give higher rank to those who pay for it (we’ve got this one already).
• Confugle: Instead of giving results directly related to your search, give results that are only metaphorically related. Include a Wiki for users to write their best guess as to why a page was included. (Actually, this one is probably already commonplace as well…)
• Voyeugle: See the results from the previous person’s query instead of your own.

If you can think of more variations feel free to comment…

Google has indexed around 8 billion web pages, total.

The National Archives in Washington D.C. archives around 100 million paper pages per year.

Nearly 15 trillion copies are produced on copiers, printers, and multi-function machines per year.

Update 4/20/05: fixed Google stat from 8 million to 8 billion (what’s a few orders of magnitude among friends?) and added copier stat. (Thanks to Mort & Beemer for keeping me honest.)

Another thing I wish I knew about long ago: http://printfreegraphpaper.com/. Good source of PDFs for graph paper in a variety of types, measures and pitches. Prints in faint grey. (By way of the Seattle Times technology section.)

Remember the Doodle Writer, the writing-desk toy with the magnetic stylus that lets kids (or you) write without making a mess? Well, Pilot has the same thing in whiteboard size. It’s called the CleanWriter Chalkless Board, and it’s mainly being marketed as a whiteboard replacement for clean rooms. A coworker of mine just picked one up for the new playroom he’s setting up for his two-year-old — I’ll post an update when I find out how she likes it. (I know I’d think it was way cool at age two — or even age 35.)

Nice piece of glue by Paul Rademacher, combining Google Maps with Craigslist to produce a great integrated service. This is exactly what I wish I’d had when I was looking for a house all last year. (Thanks to Adam for the link!)

EFF has posted a short paper on how to blog without getting fired, breaking it down roughly into (1) blog pseudonymously, (2) limit your audience and (3) know your (lack of) legal rights.

It’s unfortunate that (4) come to a reasonable agreement with management about what’s acceptable wasn’t even in the running. That’s a tricky negotiation though, both because once you broach the subject it’s much harder to go back to being anonymous and because your management might feel OK about looking the other way but when pressed might feel the need to say no rather than yes. And when it comes to protecting themselves from upper management or angry stockholders should your blog embarrass the company, they’re probably right.

I’m of two minds when it comes to pseudonymous writing. On the one hand, I still want more choice of soft walls when it comes to managing what I write. Mailing lists and things like LiveJournal‘s friends lists are good starts, but what I really want is a publish-this-to-everyone-except-those-who-would-get-me-in-trouble-for-what-I-wrote button. But on the other hand, I can’t help but see such a button as a kind of cowardly way out. Maybe it just stirs some deep emotion implanted during half-listened to high-school discussions of Thoreau, but isn’t the measure of a writer, at least in some small way, just how much trouble his writing gets him into?

Comments are working after being down for a day, now that I’ve finished installing MT-Keystrokes, the latest weapon in the anti-comment-spam arms race. It’s a plug-in that uses Javascript to detect whether a person has typed something on the form prior to submission — it’ll be pretty simple for spammers to code around, but it may give me a month respite in the mean time. (Note that this also means you’ll need Javascript enabled to post comments… shouldn’t be a problem for the one or two of you who comment every month, and hopefully it’ll be slightly more of a problem for the hundred or so spambots that try to post and clog my filter box.)

Movable Type has so much effort going into trying to block spam — I wish they’d put even half that effort into making a half-decent interface for just deleting comments in bulk…

It’s interesting how the Google satellite pictures differ ever so slightly from the Google map.

In particular, notice how the NW-to-SE roads are skewed about 5 degrees clockwise from true while the NE-to-SW roads are skewed about 5 degrees counterclockwise from true. (Click on the image for higher-resolution.)

On second thought, it could be that the map is correct and the satellite images are skewed to locally fit the Google perspective. Maybe the map is the territory after all!