Dear Rich,
I'm not quit sure I understand "flattened". If you are asking whether the entire citation should be written as one string, e.g.
Baldwin, C. C. and W. L. Smith (1998). Belonoperca pylei , a new species of seabass (Teleostei: Serranidae: Epinephelinae: Diploprionini) from the cook islands with comments on relationships among diploprionins. Ichthyological Research 45(4):325-339.
such as would be included in a Dublin Core tag dcterms:bibliographicCitation then I think this would be a BAD THING. By all means have this as an additional element, but keep the individual bibliographic elements (volume, issue, pagination, etc) separate. This greatly simplifies the task of anybody trying to find the reference, or assign a GUID.
If by flat you mean that the journal information is a link to another LSID for the journal, rather than containing a string for the journal name, to me it seems needlessly complex not to include the journal name in the RDF that has the rest of the bibliographic information (that is, as well as any link to the journal). The ISSN is already included, which applies to the journal not the article, so if the idea was to cleanly separate articles from containing journals, then the model has already failed to do that.
Lastly, I'm sure I missed the discussion, but I don't see why the RDF at http://rs.tdwg.org/ontology/voc/PublicationCitation has it's own vocabulary, rather than using existing vocabularies such as Dublin Core (see http://dublincore.org/documents/dc-citation-guidelines/) and PRISM. Another aspect of integration is sharing vocabulary as well as GUIDs.
Compare
http://nsdb.bishopmuseum.org/authority/metadata/? lsid=urn:lsid:bishopmuseum.org:tnu:20889795-7EC7-42F3-A4C3-D1D97704A609
with
http://www.connotea.org/rss/uri/876ae25827e2e94426845b277686efca
Wonder which one is more likely to be integrated into other databases...?
Regards
Rod
On 29 Nov 2007, at 20:45, Richard Pyle wrote:
Hi Rod,
If, however, there's no GUID then I would want all the relevant metadata to hand to be able to try and find one, I don't want to have to go up citation tree to get the journal name, I just want to take the bibliographic metadata in the RDF for the name and search for the GUID (e.g., using an OpenURL resolver).
Right -- and I agree. But my question was whether the citation tree should merely be "flattened" ("collapsed"?) into a single text string that is instered into "parentCitationString", or should there be some sort of nested structure to accommodate each level in the tree. My gut feeling -- at least for now (and for simplicity), is to collapse it into a single text value.
You get to sleep -- luxury! ;-)
:-)
Actually, I have to confess: I had 11 hours of sleep last night. After a marathon week, I often just collapse on the sofa when I get home, and don't open my eyes again until the next morning.
Personally I would strip out any formatting, especially from a field that will be interpreted by other software. Furthermore, what would you do if the metadata gets served in another format, such as n3, or JSON, etc.?
I tend to agree -- but I wanted to get a sense for what others were doing.
Aloha, Rich
---------------------------------------- Professor Roderic D. M. Page Editor, Systematic Biology DEEB, IBLS Graham Kerr Building University of Glasgow Glasgow G12 8QP United Kingdom
Phone: +44 141 330 4778 Fax: +44 141 330 2792 email: r.page@bio.gla.ac.uk web: http://taxonomy.zoology.gla.ac.uk/rod/rod.html iChat: aim://rodpage1962 reprints: http://taxonomy.zoology.gla.ac.uk/rod/pubs.html
Subscribe to Systematic Biology through the Society of Systematic Biologists Website: http://systematicbiology.org Search for taxon names: http://darwin.zoology.gla.ac.uk/~rpage/portal/ Find out what we know about a species: http://ispecies.org Rod's rants on phyloinformatics: http://iphylo.blogspot.com Rod's rants on ants: http://semant.blogspot.com