Re: "What if" thoughts on a class of shape use cases

Hi Simon,

See this post by Mr Dodds:
http://blog.ldodds.com/2012/06/12/principled-use-of-rdfxml/

John



> On July 29, 2014 at 5:54 PM Simon Spero <sesuncedu@gmail.com> wrote:
> 
> 
>  There seems to be some users shape goals could be (crudely) summarized as
> wanting an equivalent of RELAX-NG that can be used to *syntactically* check an
> *isolated* document's worth of RDF statements.
> 
>  There is a concrete syntax for RDF that uses XML.
> 
>  There are libraries to convert between RDF formats.
>  There is a RELAX-NG style language for XML (I can't remember the name off the
> top of my head :)
> 
>  Can this very specific class of use cases be taken care of by just converting
> the RDF to a canonical form of RDF/XML (e.g such that the number of entities
> is minimized, or by requiring that the submitted RDF be in a canonicalized
> format), and then using RELAX-NG to specify the syntactic constraints on the
> XML?
> 
>  If that would be sufficient to address a subset of use cases, and the case
> users are familiar with XML, and do not need or want any of the features and
> complexities of RDF qua RDF, this approach could solve their problems without
> new tools or recommendations.
> 
>  It could also simplify the task of developing more RDF like validation
> recommendations since the purely syntactic use cases would not have to be in
> scope.
>