On 8/10/15 12:54 PM, Kevin M Randall wrote:
> I'm quite baffled by the statement that RDA is not an identifier-based cataloging code.  RDA has "identifier" as the very first option for recording all relationships!

In RDA description identifiers are things that come with the resource. 
"Take information on identifiers for the person from any source." "Take 
identifiers for the manifestation from any source." These are elements 
of description that happen to be called "identifiers" but they aren't 
what I am talking about. They're defined as alphanumeric strings (e.g. 
any string) that are taken from the piece, recorded as part of the 
description, and generally are assigned outside of the library workflow 
(e.g. ISSNs ISBNs). This is an example of an identifier used to record a 
relationship:

For Example:
Library and Archives Canada control
number: 0062A7592E

There is no guidance on what is an actionable identifier, no 
relationship between the identifiers and the things they identify, no 
relationship between the identifier and information about the thing 
identified. Just saying "Use an identifier if there is one" is really 
useless without some sense of what you are identifying and how that fits 
into your model. I would call RDA's approach a nod to identifiers while 
still maintaining a text-based approach. It's like someone thought "Oh, 
we should be including identifiers as an option." And aside from those 
few sentences, the full remainder of RDA is about how to create text 
strings.

Maybe I made a mistake in not qualifying that for identifiers to be 
usable to establish relationships in a computer environment, they have 
to also follow the rules for machine-actionable identifiers, which the 
above is not. I'll say here now that when I speak of identifiers in the 
context of RDF, those identifiers follow the RDF standard and are 
machine-actionable in a web environment and using software that is RDF- 
and http://-identifier aware. The above identifier is a tiny bit better 
than the textual name of the organization, but there's really nothing 
you can do with it. This shows me that the developers of RDA did not 
understand what "identifier" means in the IT sense. In any case, what we 
have here is two different ducks: they both quack like a duck, but only 
one is a fully usable duck. The other one could be anything from a duck 
to a dog toy to a distraction in a bath tub.

OK, as you can tell I'm at the end of my tether. I just despair that 
we'll never bring library data in to the 20th century much less the 
21st. oooooof.

kc

-- 
Karen Coyle
[log in to unmask]  http://kcoyle.net
m: +1-510-435-8234
skype: kcoylenet/+1-510-984-3600