Re: Element names in prov-xml

On Dec 13, 2012, at 4:49 AM, Paul Groth <p.t.groth@vu.nl> wrote:

> Yes, I mean just state the convention that we choose and why we choose it. 
> 
> In the Primer, we use Turtle and PROV-XML.  The problem is that you see something like
> 
> :bob a prov:Entity.
> 
> <prov:entity prov:id=":bob"/>
> 
> and I can see how it's weird that we define in the same namespace both a lower and uppercase

Especially when pro-o uses prov:entity as a property that is different than the class prov:Entity…

-Tim

> 
> Paul
> 
> 
> On Thu, Dec 13, 2012 at 10:28 AM, Stephan Zednik <zednis@rpi.edu> wrote:
> 
> 
> On Dec 13, 2012, at 1:57 AM, Paul Groth <p.t.groth@vu.nl> wrote:
> 
>> Hi Stephan, 
>> 
>> I think this is a fine rationale but I also think it needs maybe a couple sentences of justification. 
> 
> ?  Do you mean add this rationale to the note?
> 
>> 
>> Also, for the primer, we should think about whether we want to show each serialisation in parallel as this may cause unwanted comparison.
> 
> I agree that can be an issue with both a RDF/XML and non-RDF XML serialization.  Especially since they have the same namespace (!).  Perhaps we should just use a TriG serialization for PROV-O in the primer?  That ducks the issue about there being 2 XML serializations, which is something we should probably directly address.
> 
> --Stephan
> 
>> 
>> regards
>> Paul
>> 
>> 
>> On Thu, Dec 13, 2012 at 8:05 AM, Stephan Zednik <zednis@rpi.edu> wrote:
>> PROV-XML is not an RDF/XML serialization and I believe it would be a mistake to create the expectation that it conforms to RDF/XML conventions.  Doing so could introduce incorrect assumptions on how PROV-XML maps against PROV-O.  PROV-XML was intended as a non-RDF encoding of PROV.  For a RDF/XML serialization of PROV use PROV-O.
>> 
>> I believe we went with camelCase in element names because it conformed with PROV-N conventions.  We used PascalCase in complexType names to differentiate element and complex type names.  In the schema the complexType for entity has name "prov:Entity" and the element you use to reference a prov:Entity from the document root has name "prov:entity".
>> 
>> --Stephan
>> 
>> On Dec 12, 2012, at 3:23 PM, Paul Groth <p.t.groth@vu.nl> wrote:
>> 
>>> 
>>> 
>>> 
>>> Begin forwarded message:
>>> 
>>>> Resent-From: <p.t.groth@vu.nl>
>>>> From: Egon Willighagen <egon.willighagen@gmail.com>
>>>> Date: December 12, 2012, 22:56:42 GMT+01:00
>>>> To: "Groth, P.T." <p.t.groth@vu.nl>
>>>> Cc: Provenance Working Group <public-prov-wg@w3.org>
>>>> Subject: Re: Element names in prov-xml
>>>> 
>>> 
>>>> On Wed, Dec 12, 2012 at 10:48 PM, Paul Groth <p.t.groth@vu.nl> wrote:
>>>>> I've been having a chat with Egon Willighagen in twitter about the element name case in prov-xml. You can see excerpts below. The key question is why element names are lower case e.g <prov:entity ...> and not upper case. This does not correspond to the convention in rdf/xml plus it looks a bit weird when sitting next to the turtle.
>>>> 
>>>> The relevant section in the 2004 RDF/XML spec is 2.13 which describes
>>>> the behavior:
>>>> 
>>>> "It is common for RDF graphs to have rdf:type predicates from subject
>>>> nodes. These are conventionally called typed nodes in the graph, or
>>>> typed node elements in the RDF/XML. RDF/XML allows this triple to be
>>>> expressed more concisely. by replacing the rdf:Description node
>>>> element name with the namespaced-element corresponding to the RDF URI
>>>> reference of the value of the type relationship."
>>>> 
>>>> You can test it with this XML snippet
>>>> 
>>>> <rdf:RDF xmlns:rdf="http://www.w3.org/1999/02/22-rdf-syntax-ns#"
>>>>  xmlns:dc="http://purl.org/dc/elements/1.1/"
>>>>  xmlns:prov="http://www.w3.org/ns/prov#">
>>>>  <prov:entity rdf:about="ex:article">
>>>> <dc:title>Crime rises in cities</dc:title>
>>>>  </prov:entity>
>>>> </rdf:RDF>
>>>> 
>>>> here -> http://www.rdfabout.com/demo/validator/
>>>> 
>>>> If you 'validate' it, it will also create other formats, showing that
>>>> the above RDF/XML has a rdf:type prov:entity ... that confirms that
>>>> convention.
>>>> 
>>>> Section 2.13 is not
>>>> 
>>>>> Do we have a good explanation for this?
>>>> 
>>>> Also note that my RDF/XML snippet uses rdf:about rather than prov:id
>>>> ... I have to check whether rdf:ID or rdf:about is more appropriate,
>>>> but that would be closer to RDF/XML too than prov:id ... but that's a
>>>> separate thing you may want to look at.
>>>> 
>>>> Egon
>>>> 
>>>> 
>>>> --
>>>> Dr E.L. Willighagen
>>>> Postdoctoral Researcher
>>>> Department of Bioinformatics - BiGCaT
>>>> Maastricht University (http://www.bigcat.unimaas.nl/)
>>>> Homepage: http://egonw.github.com/
>>>> LinkedIn: http://se.linkedin.com/in/egonw
>>>> Blog: http://chem-bla-ics.blogspot.com/
>>>> PubList: http://www.citeulike.org/user/egonw/tag/papers
>> 
>> 
>> 
>> -- 
>> --
>> Dr. Paul Groth (p.t.groth@vu.nl)
>> http://www.few.vu.nl/~pgroth/
>> Assistant Professor
>> - Knowledge Representation & Reasoning Group | 
>>   Artificial Intelligence Section | Department of Computer Science
>> - The Network Institute
>> VU University Amsterdam
> 
> 
> 
> -- 
> --
> Dr. Paul Groth (p.t.groth@vu.nl)
> http://www.few.vu.nl/~pgroth/
> Assistant Professor
> - Knowledge Representation & Reasoning Group | 
>   Artificial Intelligence Section | Department of Computer Science
> - The Network Institute
> VU University Amsterdam