On 5/15/12 2:29 PM, Kevin M Randall wrote:
[log in to unmask]" type="cite">

As I understand RDA, the pre-coordinated headings are not the only way to make the relationships between entities.  Entity identifiers (work identifiers, expression identifiers, person identifiers, etc.) are the way that RDA is trying to get us to go.  But the authorized access points are what we need in the meantime, until we're in an environment that can be based on the identifiers.  I think the idea is that, in the future, systems will be creating the authorized access points (*if* they're still needed) on the fly.  The identifier will be the key, and will gather up all of the elements for whatever the use is at the moment (OPAC sorting and display, generation of a bibliography, collection analysis, etc.).  But until then, our current bibliographic infrastructure absolutely needs to have those pre-coordinated headings.

This is what I want to see mocked up. Much depends on what we decide to identify. RDA doesn't say much about that, it just says: "Use an identifier for the work." I think we're going to have to dig deeper into this. The examples are not sufficient, IMO, to illustrate the principles of identification.

For the expression, the example is:

Library of Congress control number:
Identifier for English translations of Beowulf

This is an identifier that resolves to an authority record that is combination of the Work (Beowolf) and the Expression (English translation).  In other words, the identifiers in RDA appear to be identifiers for authority records for the same *headings* that we have today. To me this means that nothing has changed except what string we record in our data.

If you look at FRBR you see that what connects an expression to a work is a link:

expressionA --> expresses --> workX

RDA does not define that link, but I think that link is going to be between an identifier for an expression and an identifier for a work.

If you look at the authority record for "Beowolf. English" (this is the example of an expression identifier in RDA), there is a "530 (see also) Beowolf", but of course that text string does not create a machine-actionable link to the work identifier.

I'm not at all convinced that this is how identifiers are going to actually work in a machine-actionable record based on FRBR. I don't think that what is illustrated in RDA works. I agree with what you say below, but I don't find that in RDA. What I do find in RDA is different from what you say. We seem to have a gap that needs to be filled.


[log in to unmask]" type="cite">

To use the RDA identifier concept in your "Ask you like it" example, the expression identifier would be in an expression table, which would have a column for work identifier (the work table would contain columns for the name identifier and the title), and columns for date and language of text.

You said:  "What we need is an analysis of our data, not a transformation of it "as is" to a new technology. If we aren't ready to admit that some traditional practices, like headings, are no longer useful or usable in today's technological environment, we cannot have any hope that our data will be relevant in the future."  Looking at the FRBR family of documents, and RDA's emphasis on identifiers (they are listed as the *first* option for relating entities), it looks to me like (some? most? all?) this data analysis has already been done.

Kevin M. Randall
Principal Serials Cataloger
Bibliographic Services Dept.
Northwestern University Library
1970 Campus Drive
Evanston, IL  60208-2300
email: [log in to unmask]
phone: (847) 491-2939
fax:   (847) 491-4345

Karen Coyle
[log in to unmask]
ph: 1-510-540-7596
m: 1-510-435-8234
skype: kcoylenet