Karen Coyle is an inspiring speaker.  I really enjoyed her presentation on Linked Data during the MidWinter pre-conference (http://alamw12.scheduler.ala.org/files/alamw12/Are%20We%20There%20Yet%20by%20Karen%20Coyle.pdf).

One of the things she really drove home is the fact that we need to be investing in fields associated with data, not fields associated with text.  For something like a title to become “data” and not just be text in a 245, there must be a unique identifier associated with it.  There was some discussion at the end of the pre-conference about how to do that in a simple fashion with MARC records.  It was suggested that we could begin using the $0(zero) subfield, which in RDA will be defined to contain a unique identifier, and is allowable to add to any field (at least as I heard it).  The RDA $0 apparently was not created with the intention of adding URIs to every field, but it’s the best solution to transforming MARC records to linked data that I’ve yet heard.

The above is a bit of an aside, however…

Karen mentioned her “5 star Linked Data” mug.  That ends with the question “Is your data 5 [star]?”  Here’s what’s on the mug:

Linked Data
* On the web, open-license
** Machine-readable data
*** Non-proprietary format
**** RDF standards
***** Linked RDF

Karen went on to reword each of the 5 stars, and then added a “5 star +”

* Data, not text
** Identifiers for things
*** Statements, not records
**** Machine-readable schema
***** Machine-readable lists
*****+ Open Access on the Web

What she means by “open access on the Web” is that once libraries have edited the MARC records they have in their local catalogues, they should then re-share them – make it possible for anyone to easily take the information they have added while tailoring a MARC record to fit the item they have in their library.

I guess some would argue that we already do this with OCLC, but then, there are a lot of limitations that OCLC places on the records they accept.  I can guarantee that content of the records my library has attached to in OCLC are not an accurate reflection of content of the records we have in our local catalogue.

So why don’t we figure out a way to just put our local records out there for the taking?

I’m really not sure.

Advertisements