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

[Christoph LANGE]

Hi David, hi James (= CD editor), hi Bruce (= DLMF developer), hi all,

  now my questions on how to add links to external datasets to the CDs have
been answered and I know how to proceed.  The next question is:  _Should_ I
proceed?

As a first step, I'd suggest to add to the OpenMath 2 CDs (in
http://svn.openmath.org/OpenMath/www/cdfiles2/cd) links to the DLMF, e.g. to
transc1#sin as given below.  First a sample listing, then some further detail
questions.  After that, the notation census and DBpedia (=
machine-understandable variant of Wikipedia) could follow, and anything else
that is reasonable and relevant.

<FMP>
  <OMOBJ xmlns="...">
    <OMA>
      <OMS cdbase="http://www.w3.org/2002/07" cd="owl" name="sameAs"/>
      <OMS cd="transc1" name="sin"/>
      <OMATTR>
        <OMATP>
          <OMS cd="mathmlattr" name="definitionURL"/>
          <OMSTR>http://dlmf.nist.gov/4.14.E1</OMSTR>
          <!-- this is the link, ... -->
        </OMATP>
        <OMV name="sin"/> <!-- ... the variable is just a dummy -->
      </OMATTR>
    </OMA>
  </OMOBJ>
</FMP>

You see that attributions instead of OMS are needed, as the DLMF URIs don't
fit into the cdbase/cd#name schema.

Questions:

The "links" to DLMF are currently redundantly given in natural language: "the
... function as described in Abramowitz and Stegun, section ...".  Do we want
to keep that text?  (BTW, the DLMF URIs differ from the A&S section numbers;
e.g. sin is in DLMF section 4.14 but in A&S section 4.3)

What kind of equivalence relation do we want to have?  The OpenMath functions
are intended to be the same as the DLMF ones.  We could express that in the
OpenMath way using relation1#eq (Is that applicable to functions?), or using
owl:sameAs, which is more widely understood by RDF-based linked data clients.
The latter would become relevant once we start serving the OpenMath CDs as
RDF in addition to OCD and XHTML.

We do not currently have a reasonable XHTML rendering for such FMPs.  @David,
would you like to implement one in the XSLTs, or would you like me to do it?

@David, a comment about that:

2010-07-16 14:01 David Carlisle <davidc at nag.co.uk>:
> you could do that with an FMP and a suitable OMS to say sameas, or what 
> we did in the old monet project was to have a very thin rdf version of 
> the cds (that threw away almost all information other than the cd and 
> symbol names) then you could use owl or something to say sameas.
> Depends whether you want to express everything in OM or whether you want 
> OM to play in a larger world (or both:-)

I actually want both (see my OpenMath workshop paper).  I would like to have
the links to external datasets inside the OCDs, as I believe that doing
everything in our domain-specific languages increases the acceptance within
the community (@all, please object if I'm wrong) and helps to facilitate
future maintenance, as I suppose that there are more people in the OM
community who wouldn't mind using some strange attributions, instead of people
who would like to learn RDF.  Maybe there will even be some future
OpenMath-aware clients that can make use of such "link" FMPs.
_Alternatively_, I want to serve RDF extracted from the CDs from the URIs of
the CDs, as most existing linked data clients are RDF-based.  That requires
some enhancements to the Apache configuration, which I have to figure out with
Paul.  The OCD→RDF translation is mostly implemented (currently inside the
SWiM wiki, but it can also be used standalone).  Some of the RDF information
that I can provide comes directly from the CDs, e.g.

<http://www.openmath.org/cd/arith1> dc:date "2004-03-30"^^xsd:date

and simply translating the CDs in order to obtain the RDF that I want to serve
is just more convenient than merging the RDF extracted from every OCD with a
separate RDF graph that contains additional information such as

<http://www.openmath.org/cd/transc1#sin> owl:sameAs <http://dlmf.nist.gov/4.14.E1>

Cheers,

Christoph

-- 
Christoph Lange, Jacobs Univ. Bremen, http://kwarc.info/clange, Skype duke4701
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 198 bytes
Desc: This is a digitally signed message part.
Url : http://openmath.org/pipermail/om/attachments/20100716/777c9f3e/attachment.pgp