It seems to me we need to be flexible about expressions in our data models.
Some cases warrant a "record" (or data) that identifies a specific expression; but others (like, for example the expression that results from a translation of another expression into a different language) it may be sufficient just to identify a category of expression. --Ben Benjamin Abrahamse Cataloging Coordinator Acquisitions, Metadata and Enterprise Systems MIT Libraries 617-253-7137 -----Original Message----- From: Resource Description and Access / Resource Description and Access [mailto:RDA-L@listserv.lac-bac.gc.ca] On Behalf Of Brenndorfer, Thomas Sent: Thursday, October 18, 2012 10:17 AM To: RDA-L@listserv.lac-bac.gc.ca Subject: Re: [RDA-L] RDA promoting the "work" entity? > -----Original Message----- > From: Resource Description and Access / Resource Description and > Access [mailto:RDA-L@LISTSERV.LAC-BAC.GC.CA] On Behalf Of Jonathan > Rochkind > Sent: October 18, 2012 10:01 AM > To: RDA-L@LISTSERV.LAC-BAC.GC.CA > Subject: Re: [RDA-L] RDA promoting the "work" entity? > > I'm not actually sure that the need to distinguish between expressions > is that important -- outside of particular minority cases involving > voluminous works with many editions that are the subject of study by > literary scholars. I'm not saying it's useless, but I'm dubious that > it's as important as some people are suggesting. > > (I'm not saying it's wrong to include it on our entity model! I think > it's exactly right to; I just am not convinced it's a high priority > for improving amongst all the things that need improvement). > > Certainly compared to just being able to collocate works, and present > a search results where manifestations of the same work are grouped > together under one entry -- that's something most of our systems still > can't do, and I think a LOT higher priority. > > Once you've collocated by work, simply displaying a list of items > under that work grouped by _format_ is probably more useful than by > expression (the majority of works will only have one expression, but > many of these will still have multiple formats; print, audiobook, > online, etc.) Print and audiobook would constitute different expressions, each of which could be found in different manifestations. With print, revisions and different language versions would constitute different expressions, and with audiobook, different narrators would constitute different expressions. Solving the problem with grouping works shouldn't mean ignoring the problem with grouping expressions-- one would think solutions for consistent data should apply across the board. Thomas Brenndorfer Guelph Public Library
