Volume 20, Number 6
Nov 1996

LC Asks IPI to Advise on Vendor Standards for Digitizing Work

The quality of the first digitization of historical documents and photographs will affect all future copies and re-generations, say scientists at the Image Permanence Institute at Rochester Institute of Technology's School of Photographic Arts and Sciences.

IPI has been a pioneer of image permanence preservation since 1985, focused on numerous not-for-profit research projects. Alongside its traditional film and microfilm preservation work, IPI is in the middle of a two-year investigation funded by the National Endowment for the Humanities into "technical issues and problems of digital imaging for use in library and archive photographic collections" which will result in a printed piece and a CD-ROM in 1997.

This academic year, the Library of Congress called upon IPI to advise on standards for vendors who will perform electronic scanning of invaluable historical materials for the new National Digital Library Program. The library intends to have 5 million items-still only 5 percent of its holdings-scanned by the year 2000. Archival items include handwritten letters, maps, documents, microfilm and photographs.

The NDLP contracted IPI for technical guidance in areas such as "capture resolution and tonality after scanning," says Franziska Frey, Ph.D., IPI research scientist. IPI recently presented its findings to the library; the written report goes up soon on the library's Web site.

"Our job was to advise on test targets and quality standards-for example, what's on the market, what equipment others are using on digitized projects and the quality of scanning required for various materials," explains Frey. "We looked at the extreme range of materials to go into NDLP's electronic library and then analyzed how to set quality controls for vendors."

Setting global digital standards will take time, advises James Reilly, director of IPI. "It took more than 50 years for traditional photography to develop its current standards, and those are not always applicable to digital work. The technology is evolving so fast that the standards lag behind."

• Reilly's concerns on the fast jump to electronic storage, not only for the Library of Congress, but any archival activities, include:
• Little or no standardization of software or hardware, therefore, each new electronic imaging system reproduces the pictorial information slightly differently.
• Risk of relying on electronic storage to the point of not bothering to save originals, though he notes that, for now, some institutions including Yale and Harvard are building off-site storage facilities-"cultural bunkers"-to house unneeded items, some relying on robot staffing.
• Forgetting that anything organic deteriorates-even electronic storage devices; such environmental decay factors must still be considered.
• Color reproduction still not "true" to original.

"But the obvious advantage remains," says Reilly. That digital storage should be lossless with each new generation (providing the original was excellently digitized), whereas "with analog storage you're guaranteed to lose information with each copy, each generation.

"With the technology for accurate digitization at our doorstep, we must assure its quality with clear standards." (From an IPI press release.)