Yesterday and today is the 10th anniversary of the "London Meeting"; the Data Models Meeting between members of the RDA and linked data communities represented by the Dublin Core Metadata Initiative (DCMI), held at the British Library, London.
As noted in a previous blog A small celebration, this is where we (Diane and Gordon) met for the first time, initially with some mutual suspicion that rapidly turned into a working partnership that continues to thrive.
We had already been brought together a few months earlier in the January/February 2007 issue of D-Lib Magazine, with somewhat competing contributions on RDA:
- Diane (with Karen Coyle): Resource Description and Access (RDA): Cataloging Rules for the 20th Century
- Gordon: Distinguishing Content from Carrier: The RDA/ONIX Framework for Resource Categorization
Diane: In fact, there was a fair amount of suspicion and tension all around during the beginning of the meeting that day, but the tension dissolved once it became apparent that there was a cooperative (and exciting) path forward.
So why do we keep talking about that decade-old meeting in London? Sometimes it’s not hard to figure out turning points looking back; it’s much less common to recognize those turning points in real time. I was there at the DCMI conference the previous fall (in Mexico) when the discussions began about the importance of such a meeting. I give a lot of the credit for making it happen to Don Chatham, recently retired from ALA Publishing, who accepted a (somewhat offhand) invitation from me to come to Mexico for the conference. I gave a presentation on RDA (as did Mikael) at the conference, and the conversation began.
Gordon: In 2007 I was a member of the CILIP/BL Committee on RDA, co-opted for expertise on digital libraries, and to augment representation of the interests of cataloguers in Scotland. The invitation to attend a meeting about RDA and (linked) data models began a chain of co-incidents … Highlights for me include:
I arrive late at the hotel but just in time to meet up with the RDA "side" and allay fears that we would have to defend the FRBR model. A few weeks earlier, the Scholarly Works Application Profile (SWAP) working group had based its model for e-prints on FRBR, and one of the group coordinators (Andy) was attending the meeting. So we go into the meeting the following morning with a glimmer of hope that it’s not going to be a waste of (at least) two trans-Atlantic trips and a train ride from Edinburgh.
Someone asks: And how many properties have you identified for RDA? At least 80, we say as we duck behind the table, expecting accusations of bloat. That’ll do nicely, we need a rich vocabulary, they say - and we relax some more.
The DCMI folk propose a collaborative project to develop RDF representations of RDA elements and vocabularies. That’s do-able, we say, despite the shock of the new acronyms.
We get down to real business over dinner; we are joined by Julie, the other SWAP coordinator. The long table splits naturally into two groups, with each, as it turns out the following morning, coming to similar conclusions.
Alistair wakes us all up and wows us in front of Panizzi with "something I put together back at the hotel last night" - a spreadsheet with a draft of what RDA might look like in RDF. (And that’s pretty well how we maintain RDA in RDF today, via csv files, although it looks a wee bit different.)
Diane and I are asked to be co-leaders of a DCMI task group, and I am invited to talk about RDA at a special session at the forthcoming DCMI conference in Singapore. It is the week after IFLA’s World Library and Information Congress in Durban, South Africa, in a few months’ time.
I arrive at the Singapore hotel late, but just in time to meet up with Diane. She shows me the NSDL Registry …
Ten years on, and we are still working on it!
We haven’t quite finished the RDA Application Profile, although most of the components are in place; they include:
- A new version of the RDA Element Analysis table specifying the RDA vocabulary encoding scheme (VES) associated with a property, and indicating properties that are components of aggregated statements using a syntax/string encoding scheme (SES).
- Development of the RDA Datatype and object element sets to accommodate string and thing values in the 4-fold path.
- Ongoing development of the OMR and RDA Registry to maintain application profile data.
We expect the Implementation of the LRM in RDA to develop the entities and elements that will form one or more RDA/LRM description set profiles, so we will aim to complete the first set of deliverables from the London Meeting on its eleventieth birthday, this time next year.
Gordon adds: I also met Tom Baker for the first time at the London Meeting, and it was a great pleasure to see him again at the LD4P / LD4L-Labs Community Input Meeting last week.
Diane Hillman and Gordon Dunsire