John R. Stokes passed away this weekend. This caused me to reflect on both his career and mine.
When I started at the Library of Congress in 1985, I was an early entrant into the library imaging scene, but John Stokes was already there. He captured some of LC’s huge photo collections, at that time putting them on videodisk, as part of the Library’s Optical Disk Pilot Project. Anticipating that LC would ultimately want digital images, he saved the digital intermediates. As CD-ROMs became the preferred medium, he was able to deliver those digital images to LC for a tiny fraction of the cost of recapturing them. He didn’t shy away from any original formats, whether slides, large glass plate negatives, or ungainly panoramic photos (for which he built an amazing transport system that captured and stitched together 8 foot long panoramas).
When I came to RLG in
1986 1995, John was already at work there, too, on the Digital Image Access project [19.1 MB PDF file] — where he was helping RLG members with the human side of imaging. He developed software to manage description of images and to provide access to them. He’s done work for NYPL, National Geographic, The Smithsonian, National Library of Medicine, and other museums, universities, historical societies, and cultural heritage organizations.
In the last couple of years, he and I talked many times about ways to increase the scale of digitization of special collections. I wondered if devices could be made to increase throughput for special formats in the way that the Internet Archive and Google had increased throughput for books. Once again, John was already most the way there. He had developed a capture station that could be used with a variety of robotic materials handling devices [PDF] for various formats: manuscripts, large reflective materials, transparent materials of all sizes (including film reels), post cards, and so forth.
His physics background and in-depth knowledge of color, lighting, and photographic processes allowed him to push the envelope in designing capture equipment. As happened with high-end digital cameras, if it didn’t exist and he couldn’t build or adapt it, he’d go to the manufacturer and get them to improve their equipment until it met his high standards. He devoted a lot of attention to the process, too. Software to keep track of the workflow, allow metadata input, perform image correction, facilitate quality control, and track technical data were a key part of any system he put together. He knew that while he could automate the capture, the workflow software would help to improve the human factor.
John’s concession to my plea for faster production of access images was to make the process quickly down-sample images to derive smaller images for web access, while making it possible to save an archival-quality image to storage. He learned long ago that while people may ask for a quick access-quality image, eventually they’ll want more.
John was open and honest with his customers, admitting when he was out of his depth (not often) and pointing out ever so gently when the customer was out of their depth (in my case, more often than I like to admit). His innovative approach and his commitment to quality put him squarely at the top of my list when I was asked for advice on imaging equipment or for a service provider. He was also a kind, genuine, and gentle man, always happy to talk, whether it was about “bidness” or his and his wife Bettye’s latest adventure.
I make it sound as if John ran a one-man show. He had the support of many others over the years, including several of his family members. His good work will be continued by them and other good people at JJT, Inc. under the expert eye of his son, John T. Stokes. Already they reassure us that, within a couple of months, the Stokes Imaging System for special formats will be in place for pilots at two RLG partner institutions.
Ricky Erway, Senior Program Officer at OCLC Research, worked with staff from the OCLC Research Library Partnership on projects ranging from managing born digital archives to research data curation. Ricky left OCLC in 2015.