> -----Original Message----- > From: Jonathan Rochkind [mailto:[email protected]] > Sent: November 30, 2010 11:08 AM > To: Resource Description and Access / Resource Description and Access > Cc: Brenndorfer, Thomas > Subject: Re: [RDA-L] Dates in call numbers for RDA > > It makes sense for the _Library of Congress_ to use year of receipt, > since publishers generally deposit with the LC when something is > published -- not always, but often enough that that seems like a fine > decision for LC to make for it's own cataloging, to me, for a fairly > reliable date guess which will on average be better than nothing. If I > was using a record created by LC, I'd be happy to have that date there. > > It doesn't make any sense for a random library that buys something > possibly long after it's published to do that. > > (Although I wonder if one of RDA's several dates would allow the LC to > actually say it was the date LC received it, not pretend it was the > copyright date. But the point is, I care about what date the LC received > it, that's useful information in the absence of any other dates, even to > other libraries. I don't care about what date some random library with > it's own purchasing decisions received it, that's not such useful > information). > > Perhaps an example of the problems of using LC internal guidelines for > other libraries. Got to use them with judgement as to how you are > different than LC. > > Jonathan >
Attaching provenance to elements is a topic I've seen mentioned in presentations about the Semantic Web translation of RDA (such as here http://www.slideshare.net/smartbroad/introduction-to-application-profiles). The JSC paper http://www.rda-jsc.org/docs/5editor3.pdf discusses the elements in the context of their traditional displayed values and their substitutions by schemes such as the fixed fields. There is also the possibility of dividing elements into subelements. And there is always the possibilty of using notes that explain the choice of Date of publication (there is an example of a note for explaining probable dates in RDA 2.20.7.3). The practice of using Copyright date as a probable Date of publication seems to be already ensconced in MARC. Fixed field 008/07-10 is for an actual or probable "Date of publication/release/production/execution" even if taken from a value such as 260 $c c2010. Thomas Brenndorfer Guelph Public Library
