Voice and gesture interface from 1979!


#1

[Read the post]


#2

“Put that there”


#3

The screen image looks to be a chroma key, so it’s hard to tell how much of this demo is real.


#4

What part of “LARGE GREEN CIRCLE” do you not understand?

(apparently green.)


#5

Hopefully the patent office can find this video to disprove any generic patents for this.


#6

The demo was quite real. There was an old fashioned GE eidophore (oil based TV projector) behind the ground glass screen hooked up to special board originally designed for driving a color Xerox copier as a color printer. (We used it to make tee shirts with the first xray images of Cygnus X1 on them.) The panning and zooming were done with hardware assistance with a digitally controlled mixer for zooming in to individual applications. There was even some ERDOS satellite data for exploring. The Eames chair had a joystick and a touchpad with gesture recognition, and its purchase triggered an MIT internal audit. There was a hacked up four channel sound and an early speech recognition box. The whole thing was programmed in bastard PL/I and assembly language on Interdata minicomputers. It was sort of a wonder that it worked as well as it did.

For a poor scanned copy of the ARPA report: http://kaleberg.com/sdms/sdmsbook.pdf

Wow, that brings back the memories. It would take another 15-25 years for the real world to start catching up. As we old timers say, most modern computer innovation involves trying to remember how it worked on the prototype way back when.

Another lost project at the Architecture Machine was a laser disk based street view “map” of central Aspen in multiple seasons. There were also a lot of interior views and miscellaneous ephemera like menus. I still have the old laser disks in analog and digitized form if anyone cares to do a time travel version of Aspen for Google maps.


#7

Someone should point out that the PI for this project was Nicholas Negroponte who convinced ARPA to pay for it by producing a great pate and serving great wine at the site visits. He loaded up his lab with all sorts of prototype everything and set an army of professors, researchers and undergraduates to try and get something to work. They did. Then everyone forgot about it for 20-30 years and slowly reinvented it.

A lot of the great ideas have yet to be reinvented. I’d be curious of Madeira’s thesis work on adaptive gesture recognition is ready for widespread implementation now that everyone has multi-touch touchpads. She had to work with a rather tetchy sound wave based device.


#8

Even so, the video looks keyed. Note the artifacts around the fingers, as you would expect from a bad key and not from a rear projection:


Also, the black level in the projection is way too dark, even for rear projection. The room lighting would spill onto the screen. The presentation may well have been live, and projected, but I’d say the video has been altered.


#9

There’s also an obvious keying artifact in the back of the chair. It’s possible this was a fake video used to present an actual interface that simply couldn’t be videoed well, but in any case, the video itself seems clearly altered/keyed.


#10

I suspect that very few truly fake presentations end with “oooh, shit. restart.”


#11

At the moment I mostly think the visuals have been keyed, not necessarily that the whole demo is faked. However, adding an “accident” is a pretty common trick in fake videos, because people will say exactly what you did. But, I’m not saying that goofs are proof of fakeness, only that they aren’t proof of reality.


#12

Oh it was real all right. My boss at the time worked on it.


#13

Is he sitting in an EAMES Lounge Chair !?


#14

This topic was automatically closed after 5 days. New replies are no longer allowed.