Bernhard Eversberg wrote: > James Agenbroad schrieb: >> If chapter 13 said "to align the bibliographic naming of works with >> established ways in which users refer to works" as Mr. Eversberg >> suggests, would this include at least allowing the giving of titles >> of works of nonroman origin in their original script?
[HC] If such titles are not established in an English form for an anglophone environment (by force of overwhelming accepted usage -- as, for instance, the Odyssey or, maybe, the books of the Bible), I assume the choice between original script and romanization should depend on the capabilities of the particular system; and possibly upon a preference expressed by the searcher. I am wary, however, of seeming to suggest layers of complexity in bibliographical search interfaces, some of which are already, in terms of online interfaces designed for users with minimal or no training, quite complex enough. > > That will depend on the way this "naming" of works will actually be > accomplished. The most elegant method is having a work ID number in > the bib records, referring to a "work authority" record. With respect, surely an ID number (or other alphanumeric label) isn't a "name"? I think of it (an ID) in a bibliographic system as a token: the value of the token can be converted into a name written in terms (linguistic and semantic) which match the system (or system and user) parameters laid down in the preferences in force at any one time. Those preferences may include choice of language, choice of script, even perhaps choice of name formulation in the form of name-title or title+name-qualifier, according to local or regional cataloguing rules. > Currently, something like this exists only for series and in LCSH records > for works. I suspect you mean "LC/NAF" rather than "LCSH"? > These authority records might well be enriched by vernacular forms. Having sat through a couple of sessions of MARBI meetings in New Orleans lately, I have to say that I find it very difficult to envisage any consensus leading to the implementation of such radical accomodation of equivalent forms. And in the present context of attack on all sorts of "traditional" bibliographic control, I think "steady as we go" and "backward compatibility" are likely to rule in the next decade. > The catalog interface would have to be capable to display > (not the number but) the appropriate form for the local catalog. > Barbara Tillett has emphasized that many times in her presentations. > The searching for vernaculars is of course no easy feature to support. But some libraries are still without systems that accomodate global change of bibliographic data through change of authority data. The apparently simple concept of recording a superseded form of an established heading (perhaps ineligible, in the [AACR2/LCRI/MARC21/NACO] cataloguing/cooperative rules, for being expressed as an "ordinary" cross-reference) so that local systems can process a change automatically, has fallen on the MARBI course and been sent back to the stable for further training. To layer over this presumption of immutability the complexity of equivalent or alternative forms seems to be more than our present structures can deal with. That is indeed a pity. Maybe the constraints are more in our thinking than in our systems? > However! Catalog records, as of now, do not contain ID numbers referring > to authority records, they contain "headings", sometimes called > "textual links". Some local systems, I understand, are capable of > handling ID number links. This would have to be made a requirement, > and ways to be found to enable local systems to access authority > files for the forms they need. Catalogs of the future should no longer > be ridden with the maintenance nightmare that consists of having > to periodically update the authority controlled local data fields. Maybe. As we stand now, the frame of reference of communication of bibliographic information (in the AACR world, certainly) is MARC and Z39.50. It's not easy to step back from that frame and think of bibliographic information (including authority data) in a different frame of communication and presentation, apart from the systems and processes we're familiar with, and know and love (and hate). We're wedded to the unitary record, almost always transmitted in the MARC format; to imagine drawing on separate data resources for elements that combine to make a virtual record for a distinct resource, or set of resources, is genuinely difficult to grapple with! And, since very little is to be had for free, what does a construct which involves drawing some data elements (description) from here, others (name, subject and work/expression data) from there, relationship links from elsewhere, and availability from yet another place, part drawn from an external agency, commit us to by way of transaction costs? In the long term I'm an optimist, in the short term something of a pessimist. I just don't see us having the time to work out really new constructs for recording and presenting bibliographic data; and there's always that "will be compatible with AACR2 data" constraint -- let alone the deadline for releasing RDA in 2008. I fear the Google/keywords brigade are poised not just to pull the rug from under us, but to unravel it from about us -- to drain off matters of data organization and consistency and correlation and interrelation in favour of mere description; and that the more mechanical the better. In other words, we need to focus chiefly on what's practical to achieve in framing RDA within the next twelve months or so. New tools will have to be designed, built and demonstrated before we can apply them to having a better house. Perfection is not at present within our grasp. Hal Cain, happy to be proved wrong Joint Theological Library Parkville, Victoria, Australia [EMAIL PROTECTED]
