My Involvement

In the early seventies I applied for a job at the California Institute of Technology (Caltech - link) for the position of Photo Lab Technician. This was due in part to my extensive background in photography that I obtained while in the Air Force and applied at the Pacific Missile Test Range (now The Regan Test Site - link) as well as my interest in science and technology. My boss was Dr Bruce Murray (link) and he envisioned a method of computer storage and retrieval of the vast number of pictures that had been taken by the Astronauts and robotic spacecraft at that time. Such a device would be very useful to researchers as they could then find all the pictures they needed for a research project by using criteria like Planet, Slant Range, Viewing Angle, Longitude, Latitude, etc. The system would then find all pictures that fit the search criteria and present them to the viewer. The viewer would then further refine the selection and when complete, print out all the picture informationbfor the selected pictures. The researcher could then submit the picture ID to JPL and obtain photographic prints of the desired pictures. This method would save weeks or even months of effort by any other means then available.

To retrieve 42,000 or so pictures was something of a challenge in the seventies as the desktop computer had not been invented yet so there was no mass storage like the hard drives we have today. We didn't even have floppy disks then! So one wonders how could it be done? Fortunately that was decided prior to my arrival as I wouldn't have had a clue.

The task was to be accomplished by using a microfiche viewer that could be controlled by the campus computer via an acoustical modem which was made by Image Systems of Culver City, CA. It had a rotary carrousel that stored something like 700 microfiche. The microfiche format that we chose was 5 rows by 12 viewable columns for a total of 60 viewable images per microfiche. As the carrousel held 700 cards the total number of on line images could be as high as 42,000. There was another row available but was not viewable. I used this row to identify the microfiche and to establish photographic controls.

The source pictures were photographed onto 70mm roll film using a step and repeat camera also provided by Image Systems. As there were several sizes for the source pictures (70mm Gemini and Apollo as well 28" x 30" transparencies for Lunar Orbiter), it was necessary to modify the step and repeat camera to handle them. The various source formats were handled by the camera via special modifications designed and constructed by us at Caltech. Once completed, the microfiche negatives were then hand processed. Attempts at printing them using high speed roll to roll copiers like the Exteck failed as the negative to positive contact moved slightly which reduced the quality of the image by blurring them. Consequently it was necessary to hand contact print them. The positives were then machine processed and hand cut.

In order to make the microfiche machine selectable, a metal clip was crimped onto one edge of the microfiche. This clip had two purposes: One was to allow for support of the card in the carrousel and the other was to facilitate the machines ability to randomly select any given card in the carrousel. This was accomplished by cutting a unique pattern of notches in the clip which represented a binary number that was unique to that card. A particular binary pattern was initiated by a set of fingers that were activated by solenoids controlled by the computer. The fingers would then fit into the notches in the clip. Once the fingers were activated, the carrousel would rotate causing all microfiche clips to pass by the fingers. When the selected card came by, the notches in the clip exactly fit the fingers and allowed the card to be pulled out of line via a magnet. The selected card then followed a different path out of the carrousel. and into the read gate. A mechanical device then picked up the selected card and positioned it in the projector as per the computers instructions. Various keyboard commands would then move the card's position as desired or pick up another card if desired.

In order to microfiche the Lunar Orbiter pictures it was necessairy to construct a very evenly illuminated light box as the Lunar Orbiter pictures were in the form of a 28" x 30" transparancy. Further, as the camera was not designed for such a large image source, the back lite Lunar Orbiter transparency was reflected into the lense of the step and repeat camera by two custom mirrors and metal structure. This allowed us to directly copy the 28" x 30" transparency to a 16mm image on the microfiche with a minimum loss of quality.

At the compleation of that project we found that the microfiche were of a very good quality and so they were sent to the National Space Science Data Center (NSSDC - link) for distribution to anyone who wanted a set. They gave me a nice credit here.

At this point the thousands of Lunar Orbiter pictures were only available via a very costly set of prints, or on view at only a few Lunar Libraries or via the afore mentioned microfiche.

In the end it was a very successful project. It was, along with Dr. Murray, featured in a segmant of Nova (my hands had a cameo roll controlling the microfiche viewer). Ultimately it was installed in several places. Around 1977 I was sent to the National Air and Space Museum in Washington DC to install a system for Dr. Farouk El-Baz (link) who was then director. I was invited to meet him but as no one had informed me as to who he was, I declined. I have been sorry about that ever since. But, hay, perhaps I can justify it by saying I was only 32 or so then. Hope that works. Missed seeing Richard Feynman in a similar way. Live and learn, I hope.