[openbiblio-dev] Pubmed data: URI forms for journals

[Peter Murray-Rust]

Well done.


On Tue, May 3, 2011 at 12:29 PM, Ben O'Steen <bosteen at gmail.com> wrote:

>
> I've got responses from the leasing arm of NIH! Which is brilliant, as we
> have the go ahead to use the files as specified. Still waiting on FTP
> specifics, but this is not a problem.
>
> We have to conjure URIs for journals....
>
> The journals in pubmed have either an issn (very common, but not blanket
> coverage), and/or an eissn.
>

At this stage I think we should do the easy obvious things even if the
coverage is lower. So if there is no ISSN just ignore them. I'd be surprised
if the main journals did not all have ISSNs


> What seemed obvious at the time, was to use either:
> bibliographica.org/issn/1234-1234 or /eissn/.... (Issn ans eissns have the
> same form unfortunately and conflict so precludes /journal/{issn/eissn
> number})
>
> Is there now a commonly adopted purl or similar stub for issns/eissns?
>
> Anyone have a better idea?
>
> Also, need an agreed way to note journal abbreviations, both ISO and other.
>
>
> Psuedo record (as typing on phone): Caps indicate unknown/unchosen
> predicates.
>
> <http://bibliographica.org/issn/0003-410X> [
>   dcterms:title "Annales de medicine interne" .
>   bibo:issn "0003-410X" .
>   ISOABBR  "Ann Med etc....."
>   MEDABBR "...."
>   dcterms:identifier "nlmid:0171744" .
>
Presumably ISOABBRs could be mapped to ISSNs if these aren't included in the
record (of course that assumes that this mapping is stable and can be
applied historically)

P.
-- 
Peter Murray-Rust
Reader in Molecular Informatics
Unilever Centre, Dep. Of Chemistry
University of Cambridge
CB2 1EW, UK
+44-1223-763069
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.okfn.org/pipermail/openbiblio-dev/attachments/20110503/e803ae12/attachment.html>