For one of my own projects, I chose standardized codes and names from references to represent IDs and namess for:
1) institutional affiliations using Carnegie Classification for universities; 2) company names with ticker symbols in financial registry, etc.; 3) postal codes for names of cities and states, etc.; 4) Geo-codes for geo-mapping of location; and other codes and metrics. Public library access to network was hard to get access to. Therefore, we have to pre-pack those codes for offline access with update notification as prerequisite for all applications built for RDA/FRBR. As for the form of choices for names in NARs, a preferred label will be used for fuller form of the name if that is a predominantly known name, e.g. 80/20 rule, and alternate label will be used for the abbreviation of the name. In the past, the place name was used as a qualifier in order to differentiate person A from person B if both have the same name. With geo-tagging and mapping capability, I have seen such pratice being extended to personal and institutional names for local, mobile, and social apps. That's why roles of a person, his/her professions, field of expertise, interests, homepages, emails, phone numbers, native languages, etc. are getting critical for disambiguation. We don't display such data without his/her consent even if such data is in published literature. In addition, we must keep them confidential in our DBs in every possible way. That's the exquisite services we deliver and differ from others. Hope it helps!!! Amanda Xu Sent from my iPhone On Sep 24, 2012, at 12:09 PM, John Hostage <host...@law.harvard.edu> wrote: > This is one of the stranger provisions in RDA. In some areas some of us are > intent on using authorized access points in as many elements as possible, > even where it is not necessary, such as affiliation, while here we are > instructed to use abbreviations with the apparent intent to record places in > the form they would have in an addition to an access point, though this is > not stated explicitly. When would a place name be used in an access point > for a person? Does this use of abbreviations help us approach a future world > of linked data? > > If abbreviations are desired in an access point, shouldn't that instruction > come at the point the access point is constructed, e.g. 11.13.1.3, not when > the place is recorded? > > A person completing a MARC record while looking only at the MARC format > documentation will not be aware of which RDA rules apply to which subfield > and what peculiarities they contain, spread across chapters 9, 11, and 16, at > least. I'm not saying the MARC format should contain cataloging rules; in > fact, there's far too much of that already. But pointers to the appropriate > content rules would not be out of place. > > ------------------------------------------ > John Hostage > Authorities and Database Integrity Librarian > Harvard Library--Information and Technical Services > Langdell Hall 194 > Cambridge, MA 02138 > host...@law.harvard.edu > +(1)(617) 495-3974 (voice) > +(1)(617) 496-4409 (fax) > >> -----Original Message----- >> Speaking of consistency, I am noticing lots of records input by many >> different libraries where the place name abbreviations from Appendix >> B.11 have not been followed when recording a location in the 370 field. >> I have seen things like: >> >> United States >> Connecticut >> Rhode Island >> New York >> New Zealand >> Disbury, Alberta >> Wellington, New Zealand >> Western Australia >> Saskatchewan, Canada [doubly incorrect, since Canada shouldn't be >> included] Alberta, Canada [just like the one above, both input by Uk] >> >> Even more strangely, at least one library is inputting the established >> forms of states, provinces, countries and adding a $2 naf even though >> RDA instructions in all of the instructions on recording these elements >> (e.g. >> 9.8.1.3, 9.9.1.3, 9.10.1.3, 9.11.1.3, 11.3.1.3, etc.) say to >> "Abbreviate the names of countries, states, provinces, territories, >> etc., as instructed in Appendix B (B.11), as applicable." >> >> It's pretty easy to search the authority file in OCLC for incorrectly >> entered places, since you can search for example for "New Zealand" in >> the Entity Attributes index and then find the ones in 370 that are >> wrong. >> (And I fixed that country last night.) It would also sure be nice if >> OCLC searching allowed us to do more specific targeted searches in >> specific fields or elements, so you don't have to wade through records >> that have the words "New Zealand" in field 371 or 373 for example. >> >> Anyway, it would be really good to remind catalogers to use the >> abbreviations in B.11, even when recording just a state, province, or >> country. >> >
