Saturday, December 15, 2007

Electromusic Chamber Orchestra Thrills Crowd

Suffering from Bach fatigue and intrigued by electronic multimedia made flesh, neither snow, nor rain, nor gloom of night intimidated nearly 50 people from attending the David Sarnoff Library's concert this evening. And what a concert it was, alternately soothing and stimulating electronic music and images synchronized on Sarnoff Corporation's stage in two fascinating sets. Brain Statik and Xeroid Entity opened while Hong Waltzer mixed a DVD and some software to project a shifting montage of imagery, patterns, and colors through an LCD projector, and Brain Statik's Ken Palmer provided arrays of fixed and fluid lighting on-stage. There was a little bit of everything--after all, you can do a lot with guitar, saxophone, drum, and piano synthesizers. When we have photos or links to the music, we'll post them, but take it on good authority that they were worthy heirs to the legacy of RCA's Harry Olson and the Mark I Music Synthesizer built at the RCA Labs in the early 1950s.

After an intermission where the audience adjourned to Grover's Mill Coffee and McCaffrey's cookies as well as the Library's museum, the chamber players also known as the Martian Radio Orchestra--Howard Moscovitz, thereminist Kip Rosser, and Greg Waltzer--joined Brain Statik, Xeroid Entity, and Hong Waltzer onstage for another fascinating set of improvisations that gained sustained applause and enthusiastic endorsements by the happy crowd. Once again, the Library provided a space where people from a wide array of backgrounds and interests could bridge C. P. Snow's two cultures of art and science, and learn something about both as well as their combination.

Will they be back? You bet; save Saturday night, March 15, for our spring equinox show! If that's too far away, make plans to visit during our open house and NJARC radio repair clinic on Saturday, January 19. Details will follow, but Brain Statik will be back, and Kip Rosser is a good bet to play the theremin. For now, let's thank them all for coming out, along with Library volunteers Sharon Chapman and Vrinda Kaimal for running the ticket booth and gift shop, and Dr. Rebecca Mercuri of the Princeton ACM/IEEE-CS group for making it all happen!

Friday, December 07, 2007

Kaimal Completes Broadcast Manual Index

It was a dull job, but someone had to do it, and that someone was Vrinda Kaimal. Like other volunteers she started with one day a week, and then as the pages turned and she kept typing into a spreadsheet the dot-matrixed printed entries from the RCA Broadcast Manuals Index (1930-1984), something magical happened. She began coming in two days, and then three, caught up in the lines of print, patiently tapping away at MI and IB numbers, and esoteric entries like this:


where 31535 equals the IB (Instruction Book) number, B1 the file cabinet and drawer that no longer exist, ES-561485/6/7 the MI (Master Index) number, and BTG-5AL/10AL/20AL/20AR the model numbers for a device related to Broadcast Transmitters.

Wouldn't that make your eyeballs glaze over before they fell on to the keyboard? Not Vrinda's. After all, this 20-year-old binder and its 61 pages of single-spaced entries represent the only control to the 25 file cabinets of manuals donated to the Library through the good offices of Stu Cooke. While the flood last spring dissolved all the marked archival boxes holding the manuals, this spreadsheet will enable us to inventory what's in the freeze-dried boxes. Then we can try to fill in for the water-ruined copies with the duplicates sorted by Phil Vourtsis in his quiet times in the basement in 2006:

So hats to Vrinda and to Phil for their part in the unending labors of establishing intellectual control over the ever-growing holdings of our archives!

Monday, December 03, 2007

David Sarnoff, Vladimir Zworykin, and The Farnsworth Invention

Tonight, Aaron Sorkin's play finally hit the Broadway stage, and we expect the reviews will be as glowing as the buzz from the many people who saw the previews. Hank Azaria won over even those adamantly opposed to the scripted depiction of David Sarnoff through impeccable preparation and his ability to capture and play off of Sarnoff's vision of a better world united by electronic communications systems.

That said, it is well worth noting, as Paul Schatzkin has written at length in trying to redeem Phil Farnsworth's role in the invention of television, that Sorkin's play is a fiction. He writes in the broadest parody of history. He names characters after historical figures, places them in relationships vaguely relevant to their known character and behaviors, and abandons technical, legal, or commercial details for a fast-paced and engaging story in which no one comes off an innocent or a criminal.

Still, the audience leaves thinking that Vladimir Zworykin, director of electronics research at RCA from 1930-54, stole a patented idea from Farnsworth in order to make his television camera work during a visit to the younger man's San Francisco laboratory in April 1930. Is this true? No. What did Zworykin, and his patron Sarnoff, do that has resulted in them getting such bad press for the last 20 years?

They did what Thomas Edison did with electric light: take an invented, impractical technology and--with a great deal of support from engineers, scientists, technicians, marketers, investors, and customers--turn it into a commercial system that we and the rest of the world have used ever since. Alexander Magoun has written the best synthesis of the how television was innovated in the 1920s and 1930s in the relevant sections of Television: The Life Story of a Teachnology (2007). Based on available scholarship, you can ask for it at your library, borrow it through inter-library loan, or read the briefer history of relevant events below.

Patented and published proposals for television had existed since the 1880s, and an electronic solution, using no moving parts, since Alan Archibald Campbell Swinton posed the concept in 1908. Zworykin began work on electronic TV with his Russian supervisor Boris Rosing soon after, before World War I and the Russian Revolution interrupted his studies. In 1921, after Swinton elaborated on the challenge of making a camera that converted photons of light from an object into a succession of electronic images, Zworykin wrote a fellow émigré, Joseph Tykociner, about his ideas for a TV camera.

He convinced his boss at Westinghouse to invest in the project in 1923, when Zworykin applied for his first TV patent. Zworykin's initial "iconoscope," built and demonstrated in the corporate lab in East Pittsburgh, worked but poorly during a demonstration in 1925 to executives. During the process he and other inventors filed new patents on electronic imagers; Zworykin's 1925 application explained the concept of insulating the imager "pixels" for storage. To get an acceptable picture, the camera needs to store the light from an image, as electrons, between scans, to increase the dynamic range of the image's brightness. After he moved to RCA in April 1930, Zworykin spent three years recruiting skilled engineers, scientists, and technicians to overcome the bottleneck to practical, commercial, electronic TV: a camera that stored light electronically rather than requiring immense amounts of light to operate in a studio or outdoors.

Phil Farnsworth, as he preferred to be called, was the first to publicly demonstrate an electronic TV system, in 1928; unlike Westinghouse he needed the media attention to attract investors. He also won priority on a basic patent that he could not reduce to commercial practice for physical reasons that remain true today. The failed struggle against the laws of physics represents a pattern in all three of the inventions that his fans promote, and it's indicative of the real tragedy of Farnsworth's genius. What might he have done with university training, the mentorship or guidance that might have led in legitimately memorable directions?

We'll never know. The image dissector, the camera imager in which he received priority on the concept of an electrical image, had no storage capacity for the photo-electrons landing on its imager between scans. That is, it converted into electrons only the photons emitted from an object at the moment of scanning, which occurred 30 times a second. Since those 30 moments represent an infinitesimal fraction of a second, very little light information is captured for transmission. As a result the image dissector requires extremely high light levels, from bright sunlight or carbon arc lamps (used in movie projectors), to give a good picture. It was used only briefly in broadcasting movies and in certain industrial environments. For live action its primary advantage over the electromechanical systems that John Logie Baird and Charles Jenkins pioneered in the mid-1920s was the prospect of a larger if equally dim picture.

Zworykin was one of a series of corporate visitors to Farnsworth's lab, for the latter's backers were increasingly desperate as the Depression deepened to retrieve some of their investment. He spent three days in friendly and diplomatic engagement with the younger inventor and his amateur staff, and sent instructions on how to make a better image dissector back east for use until his team could reduce the iconoscope to practice. Shortly after returning himself, Zworykin filed another patent application specifying the storage principle in his imager.

How good was the image dissector? Consider these photos from Donald Godfrey's solid if flawed biography, Philo T. Farnsworth: The Inventor of Television (2001). In the first, showing a photo and a screenshot around 1929, Pem Farnsworth closes her eyes against the light needed to capture her image, possibly outdoors. In the second, Farnsworth stages some football dummies on Philadelphia's Benjamin Franklin Parkway during his 1934 demonstrations. It appears to be overcast, so note the two stage lights set up to illuminate the target for the camera.

The fact that Farnsworth had a valid and well-written patent on the next big thing, however, forced Sarnoff and RCA to look further into his work. In the spring of 1931, Sarnoff visited the lab during a trip to California on sound-on-film business. As luck would have it, Farnsworth was closing a research deal with RCA's leading radio set rival, Philco, in Philadelphia because his San Francisco investors had abandoned him. In his absence, Sarnoff offered $100,000 for his patents as an opening bid in what should have been a negotiated settlement. But Farnsworth relayed a flat rejection to bring his family to north Philadelphia, where his boss Albert Murray was an ex-RCA manager who had previously given the image dissector a negative evaluation.

Late in 1932, one of Zworykin's team, Sanford Essig, overbaked an image plate and realized serendipitously that the mosaicked pattern of silver globules gave the world's researchers (besides Farnsworth) what they were looking for: a technique for insulating the image sensor's "pixels." Six months later, Zworykin began publishing the article on the iconoscope in five countries; within two years it had become the standard camera imager and with further improvements the U.S. and German militaries used it successfully in TV-guided missiles during World War II as well as in studio and outdoor cameras.

How good was the iconoscope? Here are two photos of Zworykin in the Camden lab and on-screen from around 1932-33, from the Leslie Flory Collection at the David Sarnoff Library, in an early internal demonstration of its sensitivity.

Meanwhile, Farnsworth attempted to work around the lack of storage and prior work by adding an electron multiplier to the image dissector. To avoid others' patents and disclosures, his "multipactor" differed from the multipliers developed by German researchers and improved by Zworykin and his team. Theirs kicked off additional electrons from a pinwheel or cascade design that triggered more electrons in proportion to those generated by the initial signal. The sequence could be repeated as desired to amplify the signal without stressing each of the successive paddles or plates. Farnsworth's multipactor simply bounced the growing number of electrons between only two plates with the result that it burned out very quickly and was never commercialized.

The problem for RCA was not in its approach to a TV camera, but in the priority of patents. Remember that patent systems are designed to assign a single inventor as a commercial incentive to invention. Patent conflicts obscure the fact that multiple people operating within the same domain of knowledge--like previous patents, publications, and usually educational programs--conceive regularly and independently similar solutions to problems. Thus Farnsworth and Max Dieckmann in Germany conceived of the dead end of the image dissector; Zworykin and at least four other inventors thought of ways to store electronic images for television.

Farnsworth's patent involving the concept of an electrical image in a TV camera issued in 1930 after Zworykin's visit. Two years later the Patent Office put it in "interference" with Zworykin's 1923 application. Naturally RCA defended its claims, and the court battle took four years to settle. Farnsworth won, largely on the basis of his public demonstration, and it remains unknown why Westinghouse wasn't more helpful in documenting Zworykin's work, which is now preserved on microfilm at the Historical Society of Western Pennsylvania. RCA had to license Farnsworth's patent and did so in 1939. Farnsworth and his company needed RCA's support since it had Zworykin's camera and the market power to innovate the commercial system. As a result, Farnsworth joined the enemy; he became part of the patent pool, exchanging his set of inventions for a million dollars and access to some 3,000 RCA patents.

The money offset ten years of Farnsworth's expenses while representing ten percent of RCA's investment in creating, testing, and making a complete system. Between 1939 and 1941, the two rivals became allies. Farnsworth recovered from one of his alcoholic periods to file for several more patents. Led by Sarnoff, RCA and Farnsworth's new company fought the rest of the radio and broadcast industries--Philco, Zenith, CBS, DuMont, etc.--and the Federal Communications Commission for approval of a commercial television standard. Sarnoff described Farnsworth in 1940 to a congressional committee as the man who had done more for television than anyone else outside RCA. In theory a standard would allow Farnsworth to make some money on the cameras using Zworykin's iconoscope, although the royalties on, say, the first five years of camera sales were hardly likely to cover current expenses.

The FCC's delay, consumer resistance to the available programming, and World War II all meant that commercial innovation would have to wait. Farnsworth's basic patents expired in 1947, when RCA was still two years from netting the first profits on its $50 million investment. Meanwhile ITT bought Farnsworth's struggling company and retained him as a vice president in research, allowing him to work on what he liked while enjoying stock options and a handsome salary.

Farnsworth's third memorable invention was electrostatic containment of nuclear fusion for the home heating and power. This proved even more impractical than the first two; there is a possibility that the technique could be made to work with a hydrogen isotope found in abundance on the dark side of the moon. Whether or not Farnsworth knew that, he spent a lot of ITT's money and then all of his own savings tilting at his final windmill in the 1960s. That, not RCA's patent strategy or the FCC's caution in setting a television standard, is why his family had nothing to show financially for a brilliant but mentor-less career.

RCA's staff moved beyond Zworykin's inventions to create better TV cameras during and after World War II. The image orthicon and the vidicon drew on Farnsworth's patent for low-velocity electron-beam scanning, along with numerous other RCA-invented techniques, materials, and components, to become the namesake for the Emmy award and video imagers used throughout the world and space into the 1980s. These tubes also used the storage principle, electron multipliers, and thin-film techniques for the image target that also helped Paul Weimer go on to make the first thin-film transistors (TFTs).

In sum, go to The Farnsworth Invention for the snappy dialog, the great acting, the comedy within the wrong tragedy. Respect the inventor Farnsworth for demonstrating an electronic TV signal at the age of 21, and pity him for the lost opportunities educationally and professionally, had he chosen to work in RCA's labs in the 1930s instead of ITT's in the 1950s. Admire the scientist, Zworykin, for recognizing his own limitations and assembling the team of technologists necessary to make television work as a system. And remain in awe of Sarnoff, the innovator, in whose absence from the historical record one wonders who would have led the free-market United States into the video age.