[openbiblio-dev] openbiblio catchup call today 1600 GMT

[Jim Pitman]

Mark MacGillivray <mark.macgillivray at okfn.org> wrote:

> > Good.  Do you have an example dataset already or should I provide? I have been thinking about this.
> I don't have one.

OK. I suggest we try journal and subject entities first instead of authors, as I we can pull data for those from MathSciNet entries.
Just look at any MathSciNet BibTeX entry, e.g.

@article {MR2295611,
    AUTHOR = {Pitman, J. and Yor, M.},
     TITLE = {It\^o's excursion theory and its applications},
   JOURNAL = {Jpn. J. Math.},
  FJOURNAL = {Japanese Journal of Mathematics},
    VOLUME = {2},
      YEAR = {2007},
    NUMBER = {1},
     PAGES = {83--96},
      ISSN = {0289-2316},
   MRCLASS = {60J25 (60J60 60J65)},
  MRNUMBER = {2295611 (2008a:60183)},
MRREVIEWER = {Romain Abraham},
       DOI = {10.1007/s11537-007-0661-z},
       URL = {http://dx.doi.org/10.1007/s11537-007-0661-z},
}

The lines 

JOURNAL = {Jpn. J. Math.},
FJOURNAL = {Japanese Journal of Mathematics},
ISSN = {0289-2316},

are all attributes of the journal, not the article.

For display of the article, you just need what is required for display, which is typically the
JOURNAL = {Jpn. J. Math.},
and the 
ISSN = {0289-2316},
may be used for linking to a journal entity record, which will have the full journal name

FJOURNAL = {Japanese Journal of Mathematics},

and typically also e.g. the journal url and publisher

http://www.springerlink.com/content/0289-2316/
publisher = Springer

and nice to know e.g. if the journal is open access or not.
We should be doing something to play up the visibility of OA journals like this.

I think all entities we encounter, especially journals, series, events (conferences, talks, ... ), subjects, people, 
and how we identify them, should be treated basically the same way in BibJSON.
This should really help keep BibJSON simple.

--Jim