I had the opportunity to take part in Jane Greenberg’s Metadata Marathon this afternoon, or as she preferred to call it, a Metadata Sprint. It was a really great chance to meet professional metadata enthusiasts, since many local (and not so local) library and information professionals gave interesting lightening talks. Here are highlights from several that particularly resonated with me:
Senior Principal, Information Management, Knowledge Lead for Metadata and Taxonomy Project Performance Corporation/AE Group
This lightening talk was about metadata in social media. One of the most interesting points was the idea of analyzing trends in your own social media metadata, specifically your LinkedIn profile. Trends can be the result of relocating (e.g., you will have a trend towards making connections in your new city) changing jobs, developing new interests, or reconnecting with old colleagues. Identifying these connections as a trend can help you strengthen your ties to this emerging area. It made me realize that one of my trends is the connections I’ve made working through various institutions of the Episcopal Church, and I’m thinking about joining the Church and Synagogue Library Association or the Theological Library Association (especially since they have good deals for student members…)
He also brought up a fun site called Portwiture, which takes words from your most recent tweets and gathers photos from Flickr with similar words in the metadata. I don’t really tweet enough to create my own image, but I did one here for FakeAPStylebook. You can click on the images to see the metadata they have in common.
UNC Chapel Hill School of Information Science
As the self-professed “metadata heretic” of the group, Ryan Shaw asked us to consider whether metadata can become irrelevant if tools generating metadata can grab so much of a work. As an extreme example, he asked to consider if metadata exists that is so complete that it includes every word of a book, is it really separate metadata, or is it another iteration of that book? This is especially interesting (to about .01% of the population…) when you consider this idea with FRBR Group 1 entities; we did an exercise in Metadata class earlier this semester where we tried to think of something that you can’t apply metadata to in some way. Is there anything? Are things like tables and chairs actual things, or are they just manifestations of the work “table” that some clever IKEA employee came up with? Interesting stuff, riiiiiight? (I think so!)
Also, I pull up Barbara Tillett’s “What is FRBR?” so often that I probably need to just bookmark it to my toolbar…
Head, Carolina Digital Library and Archives (CDLA), UNC-Chapel Hill
Jenn Riley makes the point that library metadata was designed for finding physical things, but that’s not what we really need it to do anymore. Most libraries’ collections are composed of fewer and fewer things, and we can start thinking instead about how we can be using library metadata to “contribute to a global encyclopedia.” (That quote is an approximation; she said it better in real life but you can only take so many notes…) She also called for an end to copy cataloging and suggested that a better idea might be something like an all encompassing omnibus catalog that libraries share and select appropriate records from. It sounds a little bit like WorldCat and OCLC, but each institution needn’t literally copy all of the relevant information into their own catalogs/servers/so forth.
Chief Information Officer, Dublin Core Metadata Initiative (DCMI)
Thomas Baker gave the longest presentation at the very end of the Marathon, and he had a good explanation of Resource Description Framework (RDF). Library Science is full full full of acronyms, and this was one I was only kind of familiar with. Like, it’s definitely been mentioned and showed up on a few slides, but I haven’t done any in depth reading or practice with it. But, now I think I understand it better! RDF can be thought of as a grammar for languages of description, and the triples are like sentence structure (with subject, predicate, object.) The subject (x) is the unique ID assigned to the item, the object (y) is an attribute of the item (such as title or author) and the predicate (–) describes the relationship of the Object. For example, X –has the author– Y. X –has the title– Y. This is useful terminology to have at your disposal when considering the relationship of items to metadata, and Thomas Baker makes a good point that it’s a technique that could liberate siloed information and bring it into a bright, interoperable future.
Thanks for a great Marathon, everyone!