I'd like to follow-up on something but it's  too difficult to decide on whose conversation to latch on to!  I was burned by this earlier this summer so it's still uppermost in my mind.  I agree that an explosion of identifiers will be necessary.  As we make our transition to linked data, ALL controlled headings will need an identifier, not just the ones we decide to create according to the PCC standard.  The rub comes in that we don't create identifiers for persons, we create them for controlled strings.  I used to think this was a semantic difference of minimal importance but it turns out to be key in RDF.  Someone showed me what would be necessary to move from an authority for a controlled string, through a Real World Object, to something representing a person and it ain't pretty.

If we agree that linked data and RDF are our future, then we need to come to grips with this soon.  OCLC has recently redefined its authoritative strings for people as person to function in this context.  I think we have a couple of options.  First, we could follow the same path as VIAF and revise the NACO authority file to look to the future.  We are already mucking around with it.  Maybe this could be included.

Option two would be to continue to focus on authorities as controlled strings.  If we do this , however, we will need a parallel file to represent these strings as persons.  This option does seem a dangerous path but not one without attractions.  We could continue our work as we do now, we would not have to once again alter all of the LC NAF, we would not have to focus on a NACO-lite (and it does seem odd to have a light version of something whose purpose is authoritative strings).  These "lite" headings could be created in the parallel "person" file.  Almost anyway could be authorized to create a "person" identifier.  It's those authoritative strings that get us every time.

The downside to this is those of us that move to linked data sooner rather than later would only need the person identifiers.  We would probably stop our NACO work for creation of identifiers in this new file.  Like I said, I never even thought about this until a few months ago but it will be one of the most important issues we resolve moving forward.


On 10/3/2014 8:33 AM, Beth Picknally Camden wrote:
[log in to unmask]" type="cite"> Everyone:

It's time for one last strategic theme for discussion.   I would like us to look at Recommendation 3 for the ACI report:  "Significantly expand the ranks of those who can create
identifiers/contribute authority data".   They propose two ways of expanding:
What are the pros/cons of these options?   Should we do both, or neither?   Other thoughts on this recommendations?

Beth Picknally Camden
Goldstein Director of Information Processing
University of Pennsylvania Libraries
215-746-4149            [log in to unmask]

"You must be the change you wish to see in the world".
--Mahatma Gandhi

Philip E. Schreur
Head, Metadata Department
Stanford University
650-725-1120 (fax)