Definitely the idea of a record representing a person needs to be kept seperate
from the various names established according to various rules to label the
person. Kept seperate conceptually, and kept seperately addressable in our
data.
That doesn't mean that there necessarily need to be two records though. One
record is perfectly capable of including information about 'preferred label in
RDA', 'preferred label used at spanish national library', 'alternate label
according to X', etc, etc. indefintely. You don't need a seperate record for
that. But you do need a record capable of expressing what the relationship of a
name label to the person is, what rules it's established under, etc. One record
can do that though. I'm not sure if there are benefits to keeping this in
seperate 'record' packages? Really, as long as the data is clearly identified
and seperately addressable, some people can keep it in one record, and others
in multiple records, and it can still be exchanged and shared and translated.
Personally, I think the idea of a name as 'access point' is no longer useful to
us. The 'access point' terminology kind of has bundled up in it that a name is
used BOTH as a display label AND for collocation -- that is for identifying the
person entity the record belongs to (that's really what 'collocation' means).
Personally, I've argued before that we should be trying to move away from using
one string for both these purposes. But certainly the legacy practice needs to
be accomodated while we move away from it, sure. And it can be, with one
record or multiple records.
Jonathan
________________________________________
From: Resource Description and Access / Resource Description and Access
[rd...@listserv.lac-bac.gc.ca] On Behalf Of John Attig [jx...@psu.edu]
Sent: Tuesday, July 14, 2009 8:44 PM
To: RDA-L@LISTSERV.LAC-BAC.GC.CA
Subject: Re: [RDA-L] Urls in access fields
The situation may be a bit less clear than Karen describes. It is not clear
that the registry record represented by the URL will in fact identify a person.
Currently it identifies the person known by a particular name; in other words,
we tend to merge the concept of an entity and the name of the entity.
In most of the discussions of FRBR implementation that I have heard, it is said
that the record that represents the person (which most people are comfortable
calling an authority record, but which some are calling something more generic;
Tom Delsey suggested using "registry record" for records representing ANY of
the FRBR entities) will include a preferred access point for the person. That
makes the record less a representation of the entity (e.g. person) and more the
representation of one name for that person.
In some of my presentations, I have suggested that we should consider using
both entity ("registry") records -- to represent the person -- and name
authority records to represent one of the many possible access points
representing the name of the person. This allows us to record -- and share --
the factual information that is not dependent on particular rules or
interpretations of the facts, while also providing a record for a particular
access point that could give the context in which that access point is
appropriate -- e.g., the form established according to RDA rules by an agency
in the United States and appropriate for U.S.-English-language users. Needless
to say, the "name" records would all be linked through identifiers to the
appropriate "person" record.
I have not heard anyone else pick up on this idea, but I think it is definitely
worth considering because (in my opinion) it is the only way that we can
support the kind of identification of persons that Karen describes in her post.
John Attig
ALA Rep to the JSC
