Call for a Moratorium on Metadata
I just finished reading Dick Bulterman’s Is it Time for a Moratorium on Metadata?. He argues that metadata has become needlessly complex and not very useful for locating data, which he feels is metadata’s primary job:
Locating information is a useful activity. It’so useful that it is a problem that has been studied for centuries. For most of that time, there was an implicit assumption that the description of the object being located would, by necessity, be separate from the object itself. Consequently, searches could at best be only indirect with respect to the source information encoding. This is the card catalogue model, in which data is separated from content because a card file is easier to search than a stack of books. During the past decade, it’s become clear that for electronic assets, locating text is best done using the text itself rather than relying on metadata, because the context of the search is defined at query time rather than catalogue time. Also during the past decade, it has become clear that locating nontext assets remains an open problem.
Bulterman spends a bit of time demonstrating how contorted metadata for nontext has become, focusing particularly on MPEG-7. In this area he has a solid argument, since metadata for nontext items is a rather backwards way of going about it. Certainly not all metadata is bad, however, as noted by Henk Ellermann over at DigiLib, but Bulterman’s central point is still extremely important and often neglected: that everyone needs to be more conscious of the purpose of metadata and realize that it’s not necessarily the best solution for every application. In fact, it has the potential to be very counterproductive.