[Om] Adding DLMF links to CDs [Re: How to translate csymbol/@definitionURL]

[Professor James Davenport]

On Sun, July 18, 2010 12:10 am, Christoph LANGE wrote:
> Hi James, hi David,
>
>   OK, I got an idea.  It seems to me that, contrarily to what I had
> thought,
> the OpenMath CD language is not quite suitable for expressing RDF-like
> links to external datasets.
>
> My motivation for trying to maintain these links inside the CDs was to
> facilitate maintenance, and to make these links accessible to the
> CD→XHTML
> rendering process.  But there _is_ a way of achieving the same with links
> maintained externally.  For each CD, e.g. transc1.ocd, we could have an
> RDF/XML file transc1.links.rdf, with content e.g. as follows:
This fits in with the more general OpenMath philosophy, of not trying to
fit all information into one basket. So we have .ocd, .sts, .ecc (a pity
that this isn't really current, but that's not OpenMath's fault, it's the
evolution of the theorem-proving world), and I hope we'll have .ntn (on my
list for the next OM standard, and, David, I would hope this IMPROVED
interoperability with MathML). Personally, I would just call it .rdf, but
I'm not an expert.
> <rdf:RDF
>   xmlns:rdf="http://www.w3.org/1999/02/22-rdf-syntax-ns#"
>   xmlns:rdfs="http://www.w3.org/2000/01/rdf-schema#"
>   xmlns:owl="http://www.w3.org/2002/07/owl#">
>   <rdf:Description about="http://www.openmath.org/cd/transc1#sin">
>     <owl:sameAs rdf:resource="http://dlmf.nist.gov/4.14.E1"/>
>     <!-- or <rdfs:seeAlso>, or <relation1:eq>, etc. -->
>   </rdf:Description>
> </rdf:RDF>
>
> With a little discipline we could keep these RDF/XML files sufficiently
> well-structured, so that our XSLTs could easily parse them in order to
> find the outgoing links and add them to the XHTML.
If we continue to believe (as I do) that the XHTML is a more
human-oriented compendium of the others, then that's a good direction.
> If we assign URIs to FMPs etc. (which my CD→RDF extraction already
> does), we
> can obviously also annotate FMPs, e.g. if we don't want to say that
> transc1#sin is the same as something in DLMF, but only that the 1st FMP of
> transc1#sin is the same as some equation in DLMF.
Which indeed it is, mathematically.
> If, in the near future, we shall also serve RDF to linked data clients
> accessing the OpenMath symbol URIs, they would receive an RDF document
> obtained by merging the cdname.links.rdf and any other RDF extracted from
> the
> OCD file (e.g. what name the CD has, when it was created, what CMPs/FMPs
> there
> are for symbols).  It is feasible to automate the generation of such a
> merged RDF.  The whole process would be:
>
>        -cd.ocd#--RDF extraction of CD metadata--> cd.metadata.rdf
>       /                                                  |
>      /   cd.links.rdf# ----------------------------------| merge
>      \       |                                           |
>       \      |                                           v
>        \--------> cd.xhtml (with links)                cd.rdf (for LD
> clients)
>             XSLT
>
> … where files marked with # would be maintained manually, and the rest
> generated.  What do you think about that?
I'm not enough of an expert to comment definitively, but I would add (in
the # category) the .sts (probably only for arity) and .ntn.

James Davenport
Lecturer on XX10190, CM30070, CM30078/50123, CM50209
Hebron & Medlock Professor of Information Technology, University of Bath
OpenMath Content Dictionary Editor and Programme Chair, OpenMath 2009
IMU Committee on Electronic Information and Communication
Council of the British Computer Society
Federal Council, International Foundation for Computational Logic