[open-bibliography] Basic RDF/XML to bibtex/RIS conversion?

[Jim Pitman]

Ross Mounce <ross.mounce at gmail.com> wrote:

> > the scripts I wrote that parse the BNB files and spit out the bibjson
> yes, I think that sounds like something that would be useful. Although
> I may have to ask the same type of question again:
> How to get from bibjson -> bibtex  (or RIS)

I have working code for bibjson -> bibtex though it may be lossy in some cases.
There are issues of how to embed lists of objects in BibTeX, espcially links. I suggest using e.g.

link_1_url = { ... },
link_1_anchor = {... },
link_1_rel = {... },
link_2_url = {... },
link_2_anchor = {... },

Of all the things that are difficult to squash into bibtex, links are by far the most important.
>From memory, CiteULike uses something similar. But there are no standards for this.
Another embedding I have used extensively is

links = {http://...  anchor 1
	http://...  anchor 2
         .....  },

which contains the same info except for the "rel" and other attributes of the link.
A third option would be 

link_1_html = {... },
link_2_html = {... },

where the ... is a plain html rep of the link e.g. <a href="$url$" rel = "$rel$" title = "$title$" .... >$anchor$</a>
It is a general question for this group to consider:
How best to choose and promote such simple ad hoc extensions of common standards which can be
round-tripped to BibJSON without loss of information? 
Promotion of such extensions could play a big role in adoption/expansion of BibJSON/BibServer.
--Jim
----------------------------------------------
Jim Pitman
Professor of Statistics and Mathematics
University of California
367 Evans Hall # 3860
Berkeley, CA 94720-3860

ph: 510-642-9970  fax: 510-642-7892
e-mail: pitman at stat.berkeley.edu
URL: http://www.stat.berkeley.edu/users/pitman