Comparison of cataloging schemes

With the end of the semester upon us, I’m reposting some material from my class discussion boards before they vanish into the ether.

Dublin Core

DC is built not just to fit current technology but also the current technological climate of openness and collaboration. DC reminds me of Drupal, an open source CMS. DC’s and Drupal’s deceptive simplicity framework belies their limitless extensibility. The key is in that word, “framework:” DC, Drupal, and similar open systems are designed to be starting points, not fully-realized end-products (Drupal, of course, additionally serves as a universal standard for cross walking/sharing metadata). In their base form, they are far too general to fit all the needs of a given set of users–instead, DC and Drupal are foundations upon which to build customized solutions that perfectly fit each set of users, while retaining interoperability with other users of the general system. Moreover, in the same way that open-source Drupal modules are shared amongst all users, so that anyone can borrow from (and extend) anyone else’s previous work, by using Qualified DC and RDF users can make public and share controlled vocabularies and relationships.


PBCore’s granularity is impressive. The XML wrappers take up a lot of space when viewing a raw record but the point of such a language isn’t how it looks in its raw form–ideally, users (or even most catalogers) will never see the XML version. Still, the use of descriptive tags instead of MARC codes means that–even absent an application to render the XML into a pretty webpage–untrained users can still piece together an item’s metadata.

I like that PBCore provides two ways of bringing related items together: by embedding multiple instantiations in a record or by identifying relationships between separate records. That allows for a simple yet flexible hierarchy.


It’s painfully obvious that MARC was designed for an earlier era, when space was primary consideration. I have seen it argued that MARC endurance over the years shows that the schema has been flexible enough to accommodate our changing world. That may be true, but the perspective I default to is this: if you were designing a schema from scratch, with today’s technology, to do what MARC does, would it look like MARC? Of course not. In fact, this scenario played out this decade and the result was MODS, which bridges the gap between MARC’s level of detail and DC’s simplicity. (There’s also MARCXML, but that is simply a direct translation of MARC’s byzantine terminology into XML. That helps with interoperability and the like but it’s still handicapped by the legacy of MARC.) MARC is similar in many respects to LCSH–to outsiders/beginners the bulk and complexity of the rules seem arbitrary and overly complex. But those rules are complex because they cover such a breadth of subject matter with unparalleled detail. That said, I’ll be happy if I never have to create metadata using either ever again :)

Leave a Reply




You can use these HTML tags

<a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <s> <strike> <strong>