On 08/03/2013 02:02, Robert Maxwell wrote:
The one core relationship in RDA is to record the relationship between the resource being cataloged and the work manifested in it (see RDA 17.3). There are several ways to do this. One of the ways to do it is by using an authorized access point for the work (see 220.127.116.11). In current practice if there is only one work or expression manifested in the resource being cataloged, the authorized access point for the work is recorded in bibliographic 1XX + 240 (or 130 if there is no principal creator). So in this case, the purpose of 1XX/240 (or 7XX author-title) is to record the relationship of the resource being cataloged with the work contained in it, not to unite manifestations/works/expressions with different titles. In this case the title proper of the manifestation is evidently not the preferred title for the work, so the 1XX/240 is necessary to record the relationship between the resource and the work that is in it.
This shows the difference between RDA/FRBR and cataloging rules that came before. RDA/FRBR are philosophical, academic statements while AACR2 and previous rules are pragmatic and based on practical issues. RDA/FRBR posits that every manifestation contains a work, and a specific version of that work, the expression. Therefore, every manifestation must contain the requisite work and expression information, even if there is only one manifestation.
Previous rules did not make such a philosophical statement. They began by creating a record for the item, then if and only if it turned out that your item were related to records of other items, you would make those relations in various ways. In the physical catalogs (card and book), this was achieved through filing those cards together in different ways, by typing the heading at the top of the card, which would tell the card filers where the card should be placed in the catalog. This system was continued into the OPACs. Therefore, before RDA/FRBR, works and expressions were arrangements of records created only when necessary. If not necessary, the cataloger could forget about works and expressions. All very tangible and exceedingly practical.
As Robert points out: “the purpose of 1XX/240 (or 7XX author-title) is to record the relationship of the resource being cataloged with the work contained in it, not to unite manifestations/works/expressions with different titles” and everything becomes much more complex for the cataloger. Every manifestation automatically “contains” a work and expression and therefore, this information must be in the record somewhere. This is definitely more complicated for the cataloger to create, and any real advantages for searchers has never been shown. The traditional FRBR user tasks can now be done using facets by anyone in the world but nobody seems to want to celebrate that success or even want to do it, while the push toward making our catalog records into “data” is also very doubtful.
The example of the typographical error in the title is a great example of all of this: what do you do with a typo in the title? In the past it was simple: you just make an added title with the corrected form, but now this becomes a difference from the “ideal/preferred title” (title of the work), when that “ideal/preferred title” doesn’t even exist! A metaphysical solution! Not very tangible nor very practical. The final product for the searchers will allow them to find the item by using correct spelling, which is exactly what happens today. They won’t notice a thing.
Practical concerns have never been RDA/FRBR’s strong point however. I can only hope that in the BIBFRAME, they will come up with some method to make the creation of the work, expression and manifestation entities as efficiently as possible with a minimum of duplication. Otherwise, the resulting format will be so complex, no web developer will be able to make heads or tails out of it.