Post by Marla Misunas
This year MCN celebrates its 50th anniversary. Just as MCN has established a network of established and emerging professionals, #MCN50 Voices brings members together, old and new, near and far.
I recently had the pleasure of interviewing Howard Besser, who has been working in musetech since the late 1970s and is known internationally for his work on many topics, including digital stewardship, copyright, and archiving, to name a few. I first encountered Howard at MCN in the 1990s where he presented about his digital image database of T-shirts, a project taken on by his students as proof of concept, when digital image databases were still a novelty.
Howard is currently Professor of Cinema Studies and Associate Director of New York University’s Moving Image Archiving & Preservation Program (MIAP), as well as Senior Scientist for Digital Library Initiatives for NYU’s Library. I’ve been reading and thinking about MCN history quite a lot lately, so an opportunity to hear about it first hand was irresistible.
Below is an excerpt from our interview.
MM: What are your earliest memories of MCN?
HB: My earliest memories are really 1977–1979, when I worked at the Pacific Film Archive (PFA) at the University of California, Berkeley. We were developing a cataloging system, and John Gartenberg was at MOMA (Museum of Modern Art) working on their GRIPHOS implementation.
Note: GRIPHOS (General Retrieval and Information Processor for Humanities Oriented Studies) was a system of computer programs that aided museum workers in cataloging, indexing, and disseminating data about works in museum collections.
We evaluated GRIPHOS and decided not to go with it, since we were part of a university that had access to mini-computers with remote access, dial-in. GRIPHOS still used punch cards, and clearly what Berkeley could do for us with their Berkeley UNIX system was more advanced.
Nancy Goldman at UC Berkeley helped me on some projects. We had these bibliographies that we did on the computer. We would dial in and do the workfrom our homes. We created indexes of film journals and books in California libraries—a union list of film holdings, books, and serials. We did another index on moving image-related equipment on the Berkeley campus. There were 13 different media centers, and none of them really talked to each other.
We made an index of everything they had, so if you needed a 16mm splicer, you’d know where to get it… all computerized.
I became active in MCN in 1986, when I went to my first conference.
The way I ended up there is kind of interesting. In 1986, at AAM (American Association of Museums) and at ALA (American Library Association) conferences, both in San Francisco, I had premiered our Berkeley image database system. We would scan works of art, catalog them, and put it all up on a large, sexy-looking, Sun Microsystems screen, and allow people to do retrieval remotely.
We got Sun Microsystems to rent us a booth. It was a madhouse! No one had ever seen an artwork on a screen before.
MM: How was the system accessed remotely?
HB: We were connecting through the phones to the database in Berkeley.
HB: Yeah, it was a modem, but the whole system was built on X-Windows, which you can kind of see as a precursor to a Web browser.
If you’re designing a database, all of your navigation and display can be handled at your user’s workstation, so it’s not taking up a lot of bandwidth. You’re moving your works back and forth, but you’re not moving your whole X-Windows system.
Note: The X Windows system was a platform-independent graphics protocol, developed by Stanford, MIT, and IBM, in the early 1980s.
MM: Because you’re using local resources.
HB: Right. We chose X-Windows because it was a client-server based system, and this was 5, 6, or 7 years before the first Web browser. That was the way to do it and make it OK for different types of computing systems.
MM: Weren’t there still lots of issues preventing interfacing between systems? In the 80s, things were very platform-specific. You had a system that was married to your hardware.
HB: Right. We wanted to be multi-platform, we wanted it to be easy to use—but people still had to download X-Windows. We really believed that things were going to go in a very different direction, that’s why we designed it to be independent.
Showing the Berkeley system got a lot of attention in the library world and the museum world, and it got back to one of the collection management system vendors, Willoughby Associates, and Lenore Sarasan.
Lenore called me up, came to Berkeley to meet with me, and even offered to pay for me to go to MCN. I think she was on the MCN Board then and she wanted to push them in new directions, and into looking at new things.
This was my first MCN meeting, and I really thought it—the New Orleans conference in 1986—was a transformational break with MCN’s past, because there were possibilities other than the GRIPHOS system.
The conference was so forward-looking. We were considering new types of computing environments, new things we could do with computers in museums.
There were so many vendors—4 or 5 different vendors selling collection management systems, commercial products. People were using microcomputers. And people were talking about other things, not just managing your collection. They were talking about exhibits, exhibition design, using computers to track the movement of objects that go in and out—not just the strict registrarial use. But much broader. So I felt that was really transformational.
I thought it was wonderful, you know, I thought it was one of the best conferences I had been to.
I got to meet people whose stuff I’d read about, but whom I’d never met; like some of the Art & Architecture Thesaurus people, people who dealt with ontology questions—it was really stimulating for me.
Some of the biggest successes of MCN were much less on the hardware and software side, but with vocabulary, consistent naming, and authority lists.
Those were absolutely huge successes that could not have been done any other way. At that time, the hardware and software were controlled by the vendors, and it was a matter of trying to convince the vendors to incorporate what you needed.
Later, when you started getting into handheld devices, it was the members of MCN that really took the lead.
MM: Speaking of vocabularies, I noticed that you were involved in Dublin Core.
Note: Dublin Core is a set of standardized metadata fields to describe objects. It’s used widely today in libraries, museums, and even by Google’s search engine.
HB: I was at the first Dublin Core meeting in 1995. And I was chair, or convener, for the third Dublin Core meeting.
MM: Wow! What was that first meeting like?
HB: [Laughs.] The meeting was really strange, because I think most of the people at the meeting had no idea—there were about 40 people at the meeting, and I’d say 30 of us thought it was a one-off thing. We weren’t very optimistic about much coming out of it, unless we dug in and did things individually.
A few people, like Stuart Weibel from OCLC, had bigger visions for it. But all of that became clearer when there was a second meeting the next year, and a third meeting…
It was a weird bunch of people—back then we said it was the librarians and their sensible shoes, it was the pocket protector crowd; it was sets of people that didn’t frequently come together. The people in the room usually only interacted with their own kind of folks.
Most of us were there to try to solve problems that hadn’t come up yet. Our goal then was to anticipate what happens if you get a million hits, when you put a word into a Web search engine.
The first web browser was released in May of 1993, so the first Dublin Core meeting in 1995 was less than two years after the first Web browser was released. It was before there really were Web search engines.
A million hits seemed pretty optimistic! On the other hand, for a lot of us—particularly people who had knowledge about the library community—there was a lot of content there, it was just a matter of making Web protocols talk to Gopher and FTP protocols, which happened fairly quickly.
Eventually you would have the problem of having too much content, or too many results. How can we narrow down search results? Can you narrow down the search between different entities that each have a different vocabulary?
We started talking about this shared core vocabulary. Communities could establish their own local vocabularies, but they could map a core set of fields into Dublin Core, too. And that shared “Core” could show up in a search within any specialized field, or in a very general search as well.
We looked closely at each core field. We wanted to make our schema much more generic, so we dumped “Author,” in favor of “Creator.” We spent hours finding the word “creator.” It ended up being a good word that can be relevant for artworks, it can be relevant for architectural works, it can be relevant for photographs that are not artistic. It works in a lot of different realms.
We also ended up exploding the idea of date.
For a normal document, there is just a single date. It’s the copyright date, usually. When you start getting to these other types of material, film, for example, you have the first release date in its native language, the release date in the U.S., you have the date it was actually shot, and finished, or edited; there’s all these other dates. So we did a lot of work with date.
/ / / / /
My conversation with Howard continued. Since it was too much for just one blog post, I will post more in a follow up post. Stay tuned for the origins of Artstor, the New Art Trust, and why MCNers are a special kind of people.
It was so fascinating to learn about the origins of the schemas we use across the field today, and talk to one of the people who was there.
Were you also there at a musetech birth? Tell us about it!