SPARQL Profile for PATCH [was Re: LDP Patch Format FPWD published]

On 09/18/2014 06:30 PM, Arnaud Le Hors wrote:
> Hi all,
>
> I'm pleased to announce that the LDP WG just published the Linked Data
> Patch Format First Public Working Draft:
> _http://www.w3.org/TR/2014/WD-ldpatch-20140918/_
>
> I want to stress that the WG is seeking feedback from the community at
> large on the direction being proposed.

Thank you for the work that has gone into this!  I am very glad to see 
progress toward supporting an RDF PATCH operation, and I am glad to see 
the thinking that has gone into ensuring simplicity.  However, I also 
have concerns about inventing a new syntax.

Overall, I think progress would be better served if, instead of 
inventing a new syntax, a simple restricted set of operations were 
defined as a *profile* of SPARQL 1.1 Update operations.  I think this 
would provide important benefits over inventing a new syntax:

  - Users would not have to learn yet another syntax that is confusingly 
similar to SPARQL.   Using a single language decreases development and 
maintenance costs.

  - Implementers could simply plug in an existing general-purpose SPARQL 
engine to get a new system up and running quickly.  Later if they decide 
that it is worth the development cost to optimize performance, they 
could replace the general-purpose SPARQL engine with special-purpose 
engine that is stripped down and optimized for this profile.

  - Implementers would have the option of supporting additional SPARQL 
1.1 Update operations, beyond what the profile requires, in a consistent 
100% compatible way.

I suggest that the LDP working group define an RDF PATCH operation as a 
*profile* of SPARQL 1.1 Update, restricted to a set of operations 
similar to those defined in the current Linked Data Patch Format draft:
http://www.w3.org/TR/2014/WD-ldpatch-20140918/

Thanks,
David

Received on Friday, 19 September 2014 21:03:13 UTC