Re: Syntactic subsetting

On 10/31/12 10:50 AM, Dan Brickley wrote:
> RSS 1.0, XMP are best not understood as RDF vocabs, but as the package 
> of some syntax rules with some vocabs. We don't have a good name for 
> such things.
How about structured data formats constrained by schemas where entity 
relationship semantics are implicit and at best coarse-grained.

Conflating RDF with the following is eternally problematic:

1. data model
2. entity relationship semantics
3. data representation formats.

I've never seen the wisdom in passing disambiguation of the above over 
to end-users and developers. It always leads to problems, as history has 
shown repeatedly.

We have similar patterns with SPARQL now, it leaves end-users and 
developers to disambiguate:

1. query language
2. query dispatch and results handling protocol
3. query results formats.


Spec writers should be responsible for disambiguation. Passing that over 
to end-users and developers simply leads to the kind of confusion and 
stunted adoption that we have right now.

-- 

Regards,

Kingsley Idehen 
Founder & CEO
OpenLink Software
Company Web: http://www.openlinksw.com
Personal Weblog: http://www.openlinksw.com/blog/~kidehen
Twitter/Identi.ca handle: @kidehen
Google+ Profile: https://plus.google.com/112399767740508618350/about
LinkedIn Profile: http://www.linkedin.com/in/kidehen