[oslc-core] OSLC Compact representation, titles with markup

[Arthur Ryman]

Sam,

By "generic RDF/XML" I mean arbitrary, unconstrained RDF/XML. We started 
out trying to define a constrained RDF/XML for general use because we 
thought it would be easier to parse with standard XML technologies. 
However, we did not want to pay the price of defining this precisely, 
having a validator, etc. We soon found it was very hard to entice toolkits 
to generate RDF/XML that precisely fit our design. The closest we came was 
the so-called Abbreviated RDF/XML in Jena. There was no spec for that. I 
assume it's just supposed to be easier to read by humans. In the end we 
decided to abandon that approach and simply adopt the W3C RDF/XML spec.

However, as a legacy, that compact UI spec still used the constrained 
RDF/XML. However, we don't call it RDF. It's content type is 
application/xml. It just happens to also be valid RDF/XML but service 
providers are not allowed to use unconstrained RDF/XML. They have to 
generate precisely that format. We could even have written an XSD to 
describe the format. The goal was to make it easy to parse in browsers, 
i.e. not require full browser-based RDF parsers.

Yes, it would have made sense to use JSON for this purpose since browsers 
can more easily parse that. However, as a general rule, we want to 
minimize the number of formats a service provider has to support. This XML 
format is a MUST.

___________________________________________________________________________ 

Arthur Ryman 

DE, PPM & Reporting Chief Architect
IBM Software, Rational 
Toronto Lab | +1-905-413-3077 (office) | +1-416-939-5063 (mobile) 





From:
Samuel Padgett/Durham/IBM at IBMUS
To:
Arthur Ryman <ryman at ca.ibm.com>
Cc:
Adam Archer <agarcher at ca.ibm.com>, "oslc-core at open-services.net" 
<oslc-core at open-services.net>, oslc-core-bounces at open-services.net, Randy 
Hudson/Raleigh/IBM at IBMUS
Date:
08/23/2011 11:12 AM
Subject:
Re: [oslc-core] OSLC Compact representation, titles with markup


Hi, Arthur. Thanks for the answers. It clears some things up for me.

Arthur Ryman <ryman at ca.ibm.com> wrote on 08/22/2011 04:40:25 PM:

> I don't think problems using XPath are a valid reason to encode markup 
> since RDF/XML itselt is very difficult to process using XPath. At one 
> point we tried to define an OSLC-variant of RDF/XML that looked like 
> "normal" XML. However, we abandonned that and now require support for 
> generic RDF/XML.

To make sure I understand: By generic RDF/XML, you are talking about what 
the spec calls "constrained RDF/XML" in Appendix B, right? [1]

> The are many equivalent ways to represent a given set of triples in 
> RDF/XML. It would therefore be very problematic to use XPath, XSLT, or 
> XQuery to process RDF/XML. The safe way to process RDF/XML is to use an 
> RDF toolkit like Jena.

This makes sense, and I agree. But for me it also raises a few questions:

- Do we need a JSON Compact representation for consumers who don't use an 
RDF library? This is one of the few resources in OSLC that doesn't have a 
JSON representation, and it seems natural since so often the consumer here 
is a web client.

- Should we define a "constrained RDF/XML" representation if our 
recommendation is to use an RDF toolkit anyway? JSON might be a reasonable 
alternative for those who don't want RDF/XML.

Best Regards,
Sam

[1] 
http://open-services.net/bin/view/Main/OSLCCoreSpecAppendixRepresentations?sortcol=table;up=#Guidelines_for_application_xml