Monday, November 26, 2007
Surf to it, click on it, wait for it to appear on your doorstep
You survived Black Friday—welcome to Cyber Monday!
Monday, November 19, 2007
Taking the left turn at Uncanny Valley
This weekend, I went with some friends to see Beowulf in IMAX 3-D. It was a very satisfying visual experience, but there was quite a bit of talk afterward about the failure of the computer-generated graphics to appear completely lifelike.
There was quite a bit of talk on this subject in the press, as well. The same conversation seems to happen every time someone makes a CGI movie that aspires to photorealism rather than falling back on the graphical tropes of the cartoon—the 2001 Final Fantasy movie got the same treatment.
It’s dangerous territory. Between depictions of life that we accept because we find them clearly artificial, and therefore harmless, and those that we accept because they fool us into thinking they are real, lie those depictions which we reject because they seem real, but somehow off. Robotics researchers use the term “uncanny valley” to describe this hypothetical space, whose occupants inspire in us a sense of mild to extreme revulsion.
We haven’t been able to cross the uncanny valley yet, with either our robots or our computer graphics. There will be a lot to think about when we succeed—science fictions authors have been preparing us since the mid-twentieth century. But I’m more interested, at least for the moment, about the prospects for computer-generated art once crossing the valley ceases to be an interesting challenge.
Consider, in broad strokes, the development of painting. The pursuit of realism occupied painters during the Renaissance. But once the techniques for realistic depiction had been developed and became widespread, realist painting ceased to be a fine art and became a craft—or the favored style of conservative and anti-intellectual regimes. The arrival of the camera freed painting from the requirements of a practical art, and modern painting became abstract or non-representational, a fitting vehicle for theory.
I think that the recent explosion of visually innovative video games is a good sign that a similar maturation of computer graphics is just around the corner.
There was quite a bit of talk on this subject in the press, as well. The same conversation seems to happen every time someone makes a CGI movie that aspires to photorealism rather than falling back on the graphical tropes of the cartoon—the 2001 Final Fantasy movie got the same treatment.
It’s dangerous territory. Between depictions of life that we accept because we find them clearly artificial, and therefore harmless, and those that we accept because they fool us into thinking they are real, lie those depictions which we reject because they seem real, but somehow off. Robotics researchers use the term “uncanny valley” to describe this hypothetical space, whose occupants inspire in us a sense of mild to extreme revulsion.
We haven’t been able to cross the uncanny valley yet, with either our robots or our computer graphics. There will be a lot to think about when we succeed—science fictions authors have been preparing us since the mid-twentieth century. But I’m more interested, at least for the moment, about the prospects for computer-generated art once crossing the valley ceases to be an interesting challenge.
Consider, in broad strokes, the development of painting. The pursuit of realism occupied painters during the Renaissance. But once the techniques for realistic depiction had been developed and became widespread, realist painting ceased to be a fine art and became a craft—or the favored style of conservative and anti-intellectual regimes. The arrival of the camera freed painting from the requirements of a practical art, and modern painting became abstract or non-representational, a fitting vehicle for theory.
I think that the recent explosion of visually innovative video games is a good sign that a similar maturation of computer graphics is just around the corner.
Tuesday, November 13, 2007
Link roundup: surface computing
In the interest of posting more frequently, I've decided to start posting link roundups. On any given day I run through 50+ rss feeds and news sites, and compile lists of links that I think might be of use in writing future columns or blog posts.
Here’s the first batch: links dealing with the recent explosion in surface computing and multi-touch interfaces. I think there’s a widespread sense that this represents the first viable update to the basic computing interface since the popularization of the mouse.
Like that device, multi-touch interfaces allow us to interact with simulated space in the same way we interact with real space: by moving our bodies. Surface computing, however, represents a much more elegant way of doing so. By combining the point of physical contact and the point of visual representation, simulated objects become more phenomenologically material.
Jeff Han’s early (Feb. 2006) multi-touch table is now available in expanded form at Neiman Marcus for a cool $100,000 (found at Techdirt). Throughout his demo, Han emphasizes the intuitiveness of the interface, going so far as to claim that the interface has disappeared. Multi-touch computing is the rare example of a technology that seems to have arrived fully naturalized: it has appeared in our world as a new and fascinating species of cultural object, instantly graspable.
This effect may help explain why it’s already showing up in the toolkits of the barkeeper and the musician.
Invocations of the computing interface from “Minority Report” are becoming cliché, as in this examination of Microsoft's Surface table.
A couple more: Sharp combines multi-touch with optical scanning and, of course, Apple's iPhone—the most commercially successful implementation of multi-touch to date.
Edit: Somehow I missed this massive history of multi-touch computing from Microsoft researcher Bill Buxton.
Here’s the first batch: links dealing with the recent explosion in surface computing and multi-touch interfaces. I think there’s a widespread sense that this represents the first viable update to the basic computing interface since the popularization of the mouse.
Like that device, multi-touch interfaces allow us to interact with simulated space in the same way we interact with real space: by moving our bodies. Surface computing, however, represents a much more elegant way of doing so. By combining the point of physical contact and the point of visual representation, simulated objects become more phenomenologically material.
Jeff Han’s early (Feb. 2006) multi-touch table is now available in expanded form at Neiman Marcus for a cool $100,000 (found at Techdirt). Throughout his demo, Han emphasizes the intuitiveness of the interface, going so far as to claim that the interface has disappeared. Multi-touch computing is the rare example of a technology that seems to have arrived fully naturalized: it has appeared in our world as a new and fascinating species of cultural object, instantly graspable.
This effect may help explain why it’s already showing up in the toolkits of the barkeeper and the musician.
Invocations of the computing interface from “Minority Report” are becoming cliché, as in this examination of Microsoft's Surface table.
A couple more: Sharp combines multi-touch with optical scanning and, of course, Apple's iPhone—the most commercially successful implementation of multi-touch to date.
Edit: Somehow I missed this massive history of multi-touch computing from Microsoft researcher Bill Buxton.
Wednesday, November 7, 2007
This space intentionally left blank
I’ve just got too much going on. Blogging will resume next week.
Subscribe to:
Posts (Atom)
