[Om] Content Dictionary for Linked Data with RDF

[Christoph LANGE]

Dear Ken,

congratulations on publishing this CD!

I welcome all such initiatives for integrating mathematical knowledge 
representation with the semantic web.  Recall our discussion at ISWC.

2012-02-29 17:34 Wenzel, Ken:
> we've recently developed a simple OpenMath CD for accessing and representing RDF data
> within mathematical expressions. The CD tries to provide a minimal set of symbols
> required to integrate Linked Data and OpenMath.

So that is, in your case, representing linked datasets as OpenMath 
objects, right?

Then my question is the same as Paul's – what kind of OM-aware software 
you are going to apply to these OM representations of LD, and what 
benefits these give you over, e.g., SPARQL queries or OWL reasoning over 
RDF.

And another question, would you also like to make these "LD in OM" 
representations linked data compliant?  I.e. serve them under the same 
URIs as the RDF, and then accept HTTP requests for the 
application/openmath+xml MIME type?  (See my paper linked below for 
details.)

Last question: Do you see certain advantages/disadvantages compared to 
the opposite approach, which I had proposed earlier?  I.e. to turning 
OpenMath CDs (but rather not the OM objects, as they are too complex for 
a reasonable representation in RDF) into linked datasets?

See

@InProceedings{Lange:OpenMathCDLinkedData10,
   author =       {Christoph Lange},
   title =        {Towards {OpenMath} Content Dictionaries as Linked Data},
   crossref =     {OM10},
   eprint = {1006.4057v1},
   eprinttype = {arxiv},
   eprintclass = {cs.DL},
   pubs = {clange,projects/ikyda2012}}
@PROCEEDINGS{OM10,
   editor = {Michael Kohlhase and Christoph Lange},
   title = {23\textsuperscript{rd} OpenMath Workshop},
   booktitle = {23\textsuperscript{rd} OpenMath Workshop},
   url = {http://cicm2010.cnam.fr/om/},
   year = {2010},
   month = jul}

http://arxiv.org/abs/1006.4057v1

Now some comments on the CD, just based on your mail, as I'm offline 
right now.

> The basic idea is to provide an OM vocabulary for representing RDF resources
> and literals (rdf.resource, rdf.literal_type, rdf.literal_lang)

My personal taste would be to keep these names as close as possible to 
their original names in RDF, i.e. using "Resource" and "datatype".  For 
the language there is no such name, unfortunately.

> along with prefix mappings for short names (rdf.prefixes, rdf.prefix).

My understanding of prefixes in RDF is that they are pure syntactic 
sugar, so I'd say any representation of them in OM should be optional at 
most.

> For the retrieval of existing resources and their property values the symbols
> rdf.resourceset, rdf.value and rdf.valueset
> are provided.
>
> The most complex symbol is "rdf.resourceset" which
> accepts an Manchester OWL expression as OpenMath string:
> http://www.openmath.org/cd/contrib/cd/rdf.xhtml#resourceset

Why Manchester OWL?  Linked data does not require OWL, but is first of 
all RDF-based, so Turtle might be a more appropriate notatin.

> This symbol requires an implementor to create a Manchester OWL parser
> that is able to interpret the given class description.
> An alternative would be to express the class description by
> using OpenMath set expressions.
>
> I hope that the RDF vocabulary is already useful and provides a starting point
> for further investigations on this topic.

Indeed I hope so!  I am eager to learn more about it.

Cheers,

Christoph

-- 
Christoph Lange, Jacobs University Bremen
http://kwarc.info/clange, Skype duke4701

→ SePublica Workshop @ ESWC 2012.  Crete, Greece, 27/28 May 2012.
   Deadline 29 Feb.  http://sepublica.mywikipaper.org
→ I-SEMANTICS 2012.  Graz, Austria, 5-7 September 2012
   Abstract Deadline 2 April.  http://www.i-semantics.at