Best of the Web
Key Dates

A&MI home
Archives & Museum Informatics
158 Lee Avenue
Toronto, Ontario
M4E 2P3 Canada

Search Search

Join our Mailing List.

Published: March 15, 2001.


Digital Strategies

Jim Blackaby


While the National Research Council study, LC21: A Digital Study for the Library of Congress, was developed particularly to address questions about a single institution, the materials in it are instructive for all who manage digital data and content not only for the issues they raise but because the shortcomings of the Library of Congress in these areas tend to be reflected in the practices of the nation's libraries and the emphases placed by granting agencies on digital projects. In the end, one can see that museums are good partners for libraries in the areas of managing digital materials, especially digital content, because museums have many years of experience in those realms. On that account, at this particular juncture, it is important to work collaboratively.

Digital Strategies

In 1998, The Library of Congress (LC) requested the National Research Council to convene a committee to spend a couple of years reviewing "the status of information technology planning and implementation" at the Library -- a digital strategy.

What did they hope for? Probably, what any institution hopes for from an outside study -- some simple answers to questions, a justification for emphasizing some projects and personnel, and, of course, a justification for de-emphasizing others. They specifically asked that the committee not venture into the murky waters of their mission.

What did the committee do? Probably, what any group with such a charge might do -- some of what it was supposed to do, some of what it was not supposed to do, and a quite a bit that no one seemed to have thought of before. The committee thought about LC's ability to create and manage metadata, to work with digital surrogates of (primarily bibliographic) materials, to collect born digital materials, and to effectively utilize digital content.

The whole report, LC21: A Digital Study for the Library of Congress, can be read on line at or copies are available from the National Academy of Sciences Press. The bottom line was that while the committee thought that the LC was filled with lots of swell stuff and that it had played a key role in the development of libraries, promotion of standardized information management and exchange, and bibliographic preservation, it had moved with great difficulty beyond its bibliocentric universe and hardly into the realm of digital information and content. The committee did take note of the extraordinary quantities of such digitalia that LC had produced -- the mass of MARC records and the five million digital surrogates of the National Digital Library, but came away unconvinced that either of these efforts included strong platforms from which to address the challenges of the digital revolution.

Some high points of the criticisms of LC included the facts that:

  • No highly placed individual was specifically charged with digital strategy for LC. A leadership position was recommended.
  • In spite of various efforts, there was no connection between the copyright process and cataloguing process. Improvements in capturing copyright data and using that data as part of the cataloguing process were recommended.
  • By selecting an off-the-shelf cataloguing system, LC had locked itself into a backwards looking mode that supported its bibliocentric past rather than a forward looking solution that might accommodate more complex kinds of data.
  • LC had failed to develop metadata standards and tools for coping with much beyond the central bibliographic record (areas like Prints and Photos that have been at the library since the turn of the 20th century have received scant supportive development; more recent things like electronic media have received none). In the digital area, in order to receive a certificate of copyright for a website, for instance, one must submit screen printouts of the site.
  • The National Digital Library project (NDL) in various forms reflected opportunism more than solid exploration into digital content, its uses, audiences, and so on.

Many other issues -- large and small were raised (the need for research into digital preservation, the complex relationship between the various branches of LC and their relationship to managing or collecting digital materials, and so on).

It is easy to not bother looking at this report. After all, it is about a library, and we're from museums. And, even if we are from a library, we are not from The Library, which has a different set of circumstances than we might. Or, if one does choose to look at the report, it is easy to think of it as being about something quite different from one's self. Libraries are by nature huge, highly structured, highly specialized and formalized institutions. Museums by comparison have a rustic, amateurish quality. On some level it is comforting to find that the LC is not perfect. But, I want to suggest that in very profound ways, this report, its findings, its implications, and perhaps what grows out of it is important for us to pay attention to. (In the wake of the report, and certainly other efforts as well, Congress has given LC $ 100 million to address digital questions.)

First off, museums have no reason to feel smug. If one looks at any museum, one finds

  • No data czar
  • Only the vaguest connections between acquisitions processes and cataloguing
  • Backwards looking cataloguing systems
  • No real or active efforts at developing standards
  • Opportunism of all kinds

So, why is it okay to throw stones? LC comes to be the library of last resort for those of us who are consumers of library science and its related fields rather than producers of it as a university library might be. In the absence of other places to look, rank and file librarians look to LC for their authority, and it from such rank and file that the librarians and library staff and volunteers are drawn. More important, perhaps, is the effect the LC has on funding for digital projects -- direct (as in the case of their Ameritech grants), indirect (as a model that funding agencies like IMLS might actively look to), or remote (as in the many foundations that support museum and library projects without necessarily having complex knowledge of the state of information science).

Museums have done a poor job of finding any center like the LC for developing standards. For many years, the Smithsonian Institution played such a role informally, until its budget was cut. Then the Getty stepped in to support standards developments, but its priorities changed with new leadership. Off and on the fringes of LC (prints and photos, e.g.) have served the interests of museums, but they have been cut as well. This is a problem, because now LC is moving out into our territory -- the territory of content management.

Museums may not be very good at saying what content management -- digital or not -- should be, but we know when it is being done and we know when it is not being done, and it is fair to say that at this time the NDL is not it in a very big, public way.

Information management tools, like any other tools, reflect the problems and solutions that they enabled. Typically, this is most clearly seen in the way that information is stored. That library systems were developed at an earlier period and on earlier technologies than museum systems helps to explain some of the differences in the ways of thinking between content managers and content holders. Just by virtue of coming later, museum information systems show the potential to do a pretty good job of starting to manage content and complex data -- there is the direct metadata, but then there is associated material as well. A history museum, for instance, has to know about the clock and the clock maker, who he learned from and worked with, and so on, as well as the basic metadata that describes the object. Some of this associated material can be easily brushed off as metadata after a fashion, but some of it -- biographical sketches of makers, for instance -- becomes content in its own right

Note, that just as the Internet brings a new storage paradigm, we also find that metadata and associated materials aren't quite enough to capture the richness of what museums know about objects (digital or otherwise) or to respond to the kinds of demands that our users put on us. There is something beyond those two kinds of information -- metadata and associative materials -- that has to do with "context" -- in sense of what the (often) fragment of material you are looking at is and how to understand it or how it relates to other things. For example -- museums happen to know a lot about their content -- whether it is an object that is full of specificity (like a log book -- we know about the captain, the owner, the vessel, and maybe some of the crew) or one that represents a class (like a sailor's bag -- maybe anonymous or not, but we can use this to generalize about sailors and their life).

But we start to move beyond this, growing from knowing things like "this is the oldest piece of upholstered furniture in America" which arises out of 1) having metadata about the piece; 2) having associated material that allows extrapolating from the piece to all furniture; and 3) being able to go out to some context of understanding.

There are related "contextual" elements that are being considered by many:

  • Reliability of metadata and associative material
  • Age specific content drawn from information repositories
  • Alternative interpretations and their history
  • Relations among things or between things and places, people, and events.

Others will be discovered and explored.

Museums have succeeded in presenting the mix of content, metadata, associative materials, and contextual elements in a variety of ways -- some more effective and clearly realized than others. But, however well we have done this, it is the case that managing this complex mix of content and data is something that museums have done or tried to do for many years. If nothing else, we do know when we are doing something absurd by following guidelines of other disciplines (though this may not stop museums from doing those things) or throwing out good material to conform to someone's notion of "standards." (Consider, for example, that while the basic or even extended data elements of the Dublin Core may work well for purposes of indexing--which was their original intent--they make a particularly tight fitting strait jacket for much of what we know about objects.)

Does LC need to be engaging in the same complex polylog of object, metadata, associative materials, and contextual elements that museums have been? Perhaps not, but they do need to be aware of the need to question what they are doing in these arenas and to constantly make sure that it is something that reflects current Knowledge Management and Information Architecture (which will change) rather than something that is reductive and inward/backward looking. And, again, it is important that even if they do not address these kinds of issues head on, it is important for them to be aware of them so that funding agents look out to new kinds of thinking rather than trying to limit materials.

LC has an obligation to be attentive to those who are experimenting with management of digital assets of all kinds -- and it happens in museums in fumbling kinds of ways and from my experience in Universities away from the mainstream (like in the photo collections or rare books or things that are too odd to be mainstream) -- are too easily squelched by those who are apt to point to the LC and say, "That's not how Library of Congress does it, and without their authority, it will not be the way that we do it either." One has to look no further than the words of Thomas Jefferson on providing his personal library as the core of the present collection to see that this kind of reductive and restrictive thinking is not what made LC great: "I do not know that it [the library] contains any branch of science which Congress would wish to exclude from their collection; there is, in fact, no subject to which a Member of Congress may not have occasion to refer."

So, where might we go from here? LC21 is a start, but it seems to me that those with complex knowledge to manage, those with experience managing such complexities, and those charged with holding content need to think harder about working together. This is a wishy washy kind of conclusion -- we, especially the "we" that is the museum community -- don't have natural centers to turn to to develop these kinds of programs; we don't get $100 million just to think about these kinds of things. But, we do think more and less formally about these issues, and a concerted effort to think about how to include universities, libraries, and, yes, even, The Library, in our thinking and discussions is important. In that respect, there are crucial roles that IMLS and foundations can play if they are willing, but together we need to work towards keeping open minds and lines of communication. (Indeed, another reason to read the LC21 report is to imagine the rich complexity of the committee's ideas coming together as a model for moving forward.)