Walt Disney Enterprises was the forerunner to Walt Disney Imagineering - Disney's R&D shop that produced major innovations in cinematography, animation and even robots. As researchers they had contact with Universities and their animatronic work became known to the military.
The 50s and 60s were a period when a lot of crazy engineering schemes were hatched. Schemes ranging from the truly crazy to those that merely weren't possible with current technology, to a few where the technologies were invented on the fly and enormous change came about. Organizations like ARPA were charged not with incremental improvements, but with order of magnitude change. Their batting average wasn't good, but they were swinging for the fence coming up with the the foundations for the Internet in ARPANET, many foundations of modern computing with Project MAC and GPS.
One of the big problems the military faced was how lighten the load of a soldier. There was an obsession with the idea of an exoskeleton - an Ironman-like suit that would give the wearer superhuman capabilities. A Naval Research organization approached Walt Disney Enterprises for help.
The memo reached Walt's desk. He simply scribbled NO on it and that was that. It turns out he made the correct decision. The necessary technology was a long way off - too far off to invent on the fly. General Electric and a few others tried and failed. Only now are we seeing such dreams realized (and then some) by companies like Boston Dynamics er Google.1
Much remains to be done, but at least the necessary technologies have advanced to the point where it is possible to see a few of the most likely directions and begin creating scenarios and timelines.
It is easy to dream up a future technological niche. At the same time it is often impossible to predict the form it will take and on what time scale. As telegraphy spread in the 19th century the bandwidth of communications was exceeding the capacity of the telegraphs. The trick was to multiplex the signals - run several frequencies over the lines. Multiplexing quickly jumped from four to eight to twelve and more signals on a single line. It was noticed sounds in rooms with several multiplexed sets were being reproduced in remote locations along with the morse code signals. It became clear that audio telephony was possible and a race to the telephone began.
About the same time several schemes for sending images over the wire were invented - a precursor to fax. It wasn't long until newspapers could exchange images to other newspapers in under and hour rather than waiting for the post. Remote images, telephony and the movies gave people the idea that television and video telephony was possible. Predictions of video telephones and television were common before the arrival of the 20th century.
Getting to television was more difficult. Numerous important technologies needed to be invented, but by the late 1920s there was a considerable amount of invention underway - most of it was impractical, but it even amateurs were building their own televisions. More serious organizations like RCA had enormous efforts underway.2 Even Bell Labs was involved, but with something else in mind.
I was reminded of this last weekend as I chatted with Uche. She was at home near London and I was in New Jersey. We were using Skype, but could have used any of a number of video chat clients. I don't know what her device was - I was using my iPad, but could have used my desktop, laptop or iPhone. The key, and what AT&T lacked, was we were both connected to the Internet with enough bandwidth.
The early AT&T attempts at video telephony were enormous and only exploratory, but in the 50s the company took the idea seriously and started working towards a real service. Expending an enormous amount of resources on everything from television cameras and receivers to network engineering to human factors work, a design known as the Mod II became part of the service offering. For only $169 a month, plus a per minute fee, you could have a video telephone in your house (note that we've seen five to six times inflation since that period).
Reading the technical journals it becomes clear the most difficult part was making the service work over AT&T's circuit switched network. Each device required a special six wire connection with line modifications required for the local loop as the one megahertz bandwidth pushed the capability pushed its capability. Even hacking the lines a six mile connection was as far as a signal would travel. To go further it was necessary to digitize the signal and put it on a special purpose T-2 digital line.
By 1973 AT&T finally reached one hundred subscribers in the US.3 By 1977 the number had dropped to nine. The labs had an effort to understand how to make color videophones to make the images more acceptable and serious advances were made to color gamuts on CRTs. A bit earlier the CCD array had been invented by Bell Labs and engineering work on it was partly supported by the lure of a small video camera. Many important technologies came about - some useful for AT&T, many for the rest of the world.
AT&T tried again in the late 80s with image compression that tried to push a signal through the regular telephone network. They were sold in pairs and an effort was made to market them to grandparents through an AARP magazine among other efforts. It failed and the extras were made available for little or nothing to AT&T employees. I remember trying to use one and being appalled at how awful the image was.4
The major problem was the AT&T's legacy network which made it almost impossible to do real video telephony at a cost people were willing to accept. The Internet changed all of that. If you can attach whatever you want to a network you can pretty much do what you want within limits. Dumb has its virtues.
Uche and I have corresponded over various matters for years, but we've never met. The video chat was a good enough that I feel as though I have finally met here.
There were many interesting learnings from the Bell Labs research. Here is a bit from a human factors summary paper that might be useful for your video chats
Beyond the quality of the picture itself, other factors played an important part in the design of the display unit. For example, the camera is placed above the picture tube to make the eye contact angle as small as possible. This is significant because, while the camera is looking at the subject, the subject is looking at the picture tube. The apparent "looking away" is annoying to the viewer unless the angle is small. The least annoyance occurs when the subject appears to be looking slightly down, which is frequently the case in normal conversation. Locating the camera just above the picture tube creates this effect.
The best position for the camera has proved to be about 12 inches above the desk top so that its field of view is essentially straight ahead. Here the camera gets the most natural-looking, least distorted view of the subject, keeping him on camera while allowing some movement, and (hopefully) not picking up too many ceiling lights.
Many technologies take decades to evolve to the point where they can leave the laboratory. History tells us that the core inventors and developers of technologies are usually not those who reap the benefits. That, along with the need to get an early return on investment is a serious problem. In the old days of Bell Labs, IBM and a few other places serious industrial research with connections to universities flourished. It still exists, but I have the sense it is not even close to being as dynamic. Outside of universities few research places are doing real research and most that claim R&D and doing very low technical risk development. Attempts are being made to change that and I hope they're successful.
__________
1 Boston Dynamics is deeply in bed with the Department of Defense with the all terrain pack animal robot Big Dog, the running robot Wild Cat, and the more human form Atlas. They were recently purchased by Google for whatever reason. I have my suspicions, but they are only a gut feeling and not worth sharing at this point.
2 The winning scheme by Philo Farnsworth was pretty much stolen by Sarnoff of RCA.
3 In my early years at Bell Labs, talking with some of the videophone pioneers I learned that Hugh Hefner was one of the subscribers. It is also possible that the 100 subscriber number was inflated with units being given to prominent people who were then comped separately.
4 The pioneers would have been horrified by the early digital compression models. A huge amount of work had established the minimum acceptable image level and they couldn't do it. A smartphone, on the other hand, passes with flying colors.
__________
Recipe Corner
Tiny Caramelized Potatoes
Ingredients
° 1 kg small and waxy fingerling potatoes
° 1/2 tbl sea salt
° 125g fine sugar - I use a very fine, but not quite powered, variety.
° 115g unsalted butter (a half stick) cut into four pieces at room temperature
Technique
° cover the potatoes with cold water and bring to a boil over a medium heat in a lidded saucepan. Add the salt and continue to boil with the lid slightly off for another 10 minutes or so until they seem close to done.
° cut the heat and keep them in the water for 10 minutes to soften them. Drain and cover to keep them warm.
° put the sugar in a heavy pan over a medium heat and cook stirring every now and again for until golden colored (about 12 minutes for me). Add butter and stir until they are properly mixed - about three minutes. Now you have your caramel.
° Cut the heat to a simmer and add the little potatoes carefully folding them in. Continue to simmer stirring every now and agin until the potatoes look like they are caramel colored About a half hour on my stove.
"Outside of universities few research places are doing real research and most that claim R&D and doing very low technical risk development. Attempts are being made to change that and I hope they're successful."
Do indulge us.
Posted by: David | 01/12/2014 at 12:26 PM