 |
 |
Yes, answering my own email, but I've learned something else... the property "publisher" is only used as an access point, something that is relatively rare. I'm also catching on to the fact (if it is a fact) that all of the access points are defined in FRAD, and are also defined in FRBR (groups 2 and 3). I'm trying to sort this out in my mind, but it's starting to seem that one should take Group 1 entities from FRBR and groups 2 and 3 from FRAD. Then one can establish relationships between instances of the groups 2&3 with those of group 1. FRBR in RDF treats the FRBR entities as classes, the relationships as properties, and does not include the FRBR attributes. RDA seems to be mainly about the attributes, and calls them elements. If anyone can untangle this mess, I'd love to hear about it. I'll just keep reading meanwhile. kc On Sat, Dec 27, 2008 at 7:26 PM, Karen Coyle <[log in to unmask]> wrote: > I've just done a re-read of some parts of FRBR, and discovered some > things that I should have known all along. (See: > http://kcoyle.blogspot.com/2008/12/frbr-and-group-2-3-oddities.html ) > > One of these is that the Group 3 entities (concept, event, place, > object) can ONLY be used as subjects. Therefore: > > rda:placeOfProduction ex:E ; > ... > ex:E rdf:type frbr:Place ; > rdfs:label "New York" ; > > can't use "frbr:Place" because it's not a subject. Place of > publication/production has to be a literal. There is no entity > provided in FRBR for this. > > I believe the same is true for "publisher" -- the RDA element is a > string transcribed from the title page, and should probably use the > property: > > http://rdvocab.info/Elements/publishersName > > We could finesse this, however, by having two elements: one would be > "publisher name" and that would be a literal; the other would be > "dcterms:publisher" -- could that be defined as a FRBR corporate > entity? If not, then we simply won't have a publisher entity in our > examples. I suspect that publisher as an entity does not exist in RDA, > so leaving it as a literal string is probably the correct RDA way. > > kc > > -- > -- --- > Karen Coyle / Digital Library Consultant > [log in to unmask] http://www.kcoyle.net > ph.: 510-540-7596 skype: kcoylenet > mo.: 510-435-8234 > ------------------------------------ > -- -- --- Karen Coyle / Digital Library Consultant [log in to unmask] http://www.kcoyle.net ph.: 510-540-7596 skype: kcoylenet mo.: 510-435-8234 ------------------------------------