The theme for this year’s Dublin Core Conference, was œMetadata for Semantic and Social Applications. Like previous DC conferences this was a dual track event with working groups running along side peer reviewed presentations. I attended the conference primarily to participate in the working groups and I have to confess that many of the academic presentations were somewhat outwith my domain, however there were a couple that caught my attention.
Jennifer Trant of Archives & Museums Informatics gave a thought provoking keynote Access to art museums on-line: a role for social tagging and folksonomies which presented findings from Steve: The Museum Social Tagging Project. Steve is:
engaged in systematic research into how social tagging can best serve the museum community and its visitors.
Jennifer began by asking if tagging could help museums answer common queries such as:
Im looking for a picture of a well dressed man standing in front of a window. Can you help me find this painting?
Museum catalogue records, or œtombstone data, are not good at answering questions such as these.
To illustrate this point Jennifer showed us a painting by Winslow Homer called œThe Gulf Stream.
Copyright Â© 2002-2008 www.winslow-homer.com This work is licensed under a Creative Commons Licence.
Now it’s hard to miss the sharks here but nowhere, Jennifer pointed out, does the catalogue record mention the word œshark (Personally I was stunned by the sheer tackiness of this painting but thats another story¦ The Steve project attempted to address the œmis-match between the vocabulary of the visitor and the museum by inviting the public to tag almost 2000 works in 11 museums. 86% of the tags allocated by the public were œnovel to the resource and the intersection with museum terms was primarily œgenre terms such as photograph, sculpture etc. Clearly the public found these terms useful but the project concluded that the tags also had significant value for searching and broader retrieval purposes. More importantly they were able to incorporate new points of view into museum records. The public had effectively created an alternative scholarly vocabulary.
Im not sure how many parallels we can draw between the museum sector and the teaching and learning domain, however I liked the idea that users can generate living dynamic descriptions of resources which can complement more static œtombstone data. I also whole-heartedly agreed with Jennifers concluding assertion that
we dont know enough about how people search and what they are searching for.
This is certainly something we have been aware of in the education domain for some time. Its difficult to create effective educational metadata profiles if we dont fully understand how teachers and learners search and what they are searching for. Semantic density anyone? We do need to recognise that individual teacher and learners will view and experience educational materials in very different ways and we also need to acknowledge that theres more to metadata than œdefinitive static records. In order to improve access to educational resources we need to create services that can accommodate dynamic resource descriptions from a range of sources rather than mausoleums of tombstone metadata. (Okay Ill stop pushing that metaphor now¦)
The other keynote which caught my attention was by Paul Miller of Tallis. What struck me about this presentation was that, excepting a very interesting potted history of the dotcom era, it was awfully reminiscent of a keynote I heard Tim Berners-Lee present at a World Wide Web conference in Toronto almost 10 years ago called The Challenges of the Second Decade (these slides provide only a bare outline of the presentation I remember) and Im not sure how much has changed in the intervening years. I already know that
the Semantic Web is about facilitating connections between data and unlocking value in all the data we collect and maintain
but has this happened yet? And if not, when will it? Dont get me wrong, Im very interested in the potential of semantic technologies and in particular the affordances they may offer to teaching and learning however I did rather feel like I was stuck in a bit of a time warp. I wonder if we need to wait a third decade before the Semantic Web becomes a reality?