Posting to RDA-L
It seems to me that the problem is not so much the concept of bibliographic identity, although I personally agree that very few members of the public have understood it. The real problem is that the authority records do not function in the modern, computerized environment. Without those authority records, people are blind. Every case of different bibliographic identities has several authority records, but in today’s catalogs, it is almost impossible to see any of those authority records. It all worked much better in a card/physical environment, where it was designed to function.
In a card catalog, the cross-references were an intrinsic part of the entire system. For someone interested in a topic, e.g. “Russia”, you would open the “R” drawer, browse through the cards through “ra” “re” and so on until you would come to the beginning of the card set for “Russia” and there, you could see all kinds of information about how the cards are arranged. You had to see these guide cards–there was no choice for the user. Here are the guide cards for how to find the various concepts of “Russia” in Princeton’s scanned card catalog. http://bit.ly/1127EvP
In the same way today, if someone browses the heading alphabetically for “Twain, Mark” (not “mark twain” or using keyword) that person will see the record that makes sense of everything. http://1.usa.gov/Z7bWfZ (although in Voyager, someone still has to click on the cryptic “Authorized & Notes”, then the incomprehensible link “Scope Note” to finally see the invaluable:
INFORMATION FOR: Twain, Mark, 1835-1910.
Search also For works of this author written under other names, search also under
Clemens, Samuel Langhorne, 1835-1910,
Snodgrass, Quintus Curtius, 1835-1910
Conte, Louis de, 1835-1910,
Alden, Jean François, 1835-1910
The searcher doesn’t have to understand or care about bibliographic identities in this case. They just do what the catalog tells them to do.
In the card catalog (that is, before the idea of bibliographic identities), searching for Alden, Jean François, you would have no choice but to see the card at http://bit.ly/TCPkoT, which would tell you what to do.
So, it seems to me as if all the relevant information exists now but it is all but inaccessible and incomprehensible to the public because it presupposes that the public still browses headings alphabetically and that they will click on things that make absolutely no sense to them. It is obvious to see that this system of cross-references was made for another world. These are some of the reasons why I maintain that the dictionary catalog–not the catalog itself, but the dictionary catalog–is dead, and has been for a long time.
So, I don’t see this problem separately or even something that URIs would fix at all. It is only one symptom of the fact that our system of authorities just don’t work, and haven’t for the last 25 years or so. This is one of the major reasons why the public does not like our catalogs.
Maintaining consistency of access is a completely different matter: that is, whether the cards made before the implementation of AACR2 were changed to the right headings. For example, a book cataloged in 1906 written by “Alden, Jean François” would have gotten the heading “Clemens, Samuel”. Later, was the heading changed to “Alden” or “Twain” or did it stay as “Clemens”? If the heading is not updated correctly–here, “Alden, Jean François”–a search for “Alden” will not retrieve the book received in 1906. This is a variation on the problem I discussed about adding the relator codes to the pre-RDA records. http://blog.jweinheimer.net/2012/09/cataloging-matters-no-16-catalogs-consistency-and-the-future.html
In my experience, I remember updating the older records for pseudonyms only when I would catalog a book new to the collection and when I discovered that I was dealing with a pseudonym. My first reaction was, my heart would sink, and then I was supposed to do the least amount of dismantling and updating of the records to be able to catalog my book, and provide consistent access for the public. I think there was a rule that if it involved updating more than 10 or 20 headings, it was considered a project and should be done separately.