- From: Mike Bergman <mike@mkbergman.com>
- Date: Thu, 10 Jan 2013 21:21:08 -0600
- To: Holger Knublauch <holger@topquadrant.com>
- CC: Ivan Herman <ivan@w3.org>, Martin Hepp <martin.hepp@ebusiness-unibw.org>, Dan Brickley <danbri@danbri.org>, Richard Wallis <richard.wallis@oclc.org>, "public-vocabs@w3.org Org" <public-vocabs@w3.org>, Jamie Taylor <jamietaylor@google.com>
Hi Holger, It is great to see you contributing on this list. See below. . . . On 1/10/2013 5:42 PM, Holger Knublauch wrote: > On 1/11/2013 0:09, Ivan Herman wrote: >> Not exactly. Personally, I like the spin 'processing' model. No >> problem there. But I think that syntax, ie, turning SPARQL statements >> into triples, is not really usable for the masses. It may be as simple >> as adding SPIN-like SPARQL statements into Turtle, ie, extending the >> syntax in that direction, but there might be some complications on the >> way that I do not see now. > > I agree that the SPIN RDF syntax is putting some people off, because it > indeed requires tooling to edit. But this is similar to OWL class > expressions (intersections etc) or SWRL that very few people would want > to edit by hand. Furthermore, the RDF syntax does have advantages when > you need to properly identify references to properties, for renaming > etc. Another obvious advantage is that the RDF syntax is independent of > namespace prefix declarations. I did not encourage a pure text syntax > because I wanted to see if we can get by without having to support too > many different modes at the same time and thus further increase the cost > of adoption and implementations. > > Anyway, not sure how off-topic this is here, but I think it would be > easy to allow textual syntax as an alternative to the SPIN RDF syntax. > Some options that I could think of quickly are: > > 1) Allow spin:rule etc to point to literals, e.g. > > ex:Adult > spin:constraint "ASK WHERE { ?this ex:age ?age . FILTER (?age < > 18) }"^^xsd:string > > 1b) This could be further clarified by introducing a new datatype > > ex:Adult > spin:constraint "ASK WHERE { ?this ex:age ?age . FILTER (?age < > 18) }"^^sp:askQueryString > > 2) Allow the blank node to contain the sp:text. sp:text is already > supported by the SPIN spec in cases where people want to store the > verbatim text alongside the RDF syntax, for example to preserve special > cases of indentation, # comments etc: > > ex:Adult > spin:constraint [ a sp:Ask ; sp:text "ASK WHERE { ?this ex:age > ?age . FILTER (?age < 18) }" ] > > 3) Introduce another property > > ex:Adult > spin:constraintText "ASK WHERE { ?this ex:age ?age . FILTER > (?age < 18) }" > > I am sure there are other options. If anyone here has specific > suggestions and if we agree that these would extend the reach of SPIN > without dividing the community too much, then I am sure these are low > hanging fruits that we can start to support quickly. Is there any reason why you can not make these changes to SPIN directly? I'd be happy to see you extend the reach of SPIN unilaterally, since it was your efforts in the first place that gave us this option. My guess is that the market would like it. If I understand correctly, these ideas require no changes to any existing standards. If the market likes it, they will use via SPIN. Thanks, Mike > > Thanks > Holger > > PS: thanks, Martin, for the cc. > > > -- __________________________________________ Michael K. Bergman CEO Structured Dynamics LLC 319.621.5225 skype:michaelkbergman http://structureddynamics.com http://mkbergman.com http://www.linkedin.com/in/mkbergman __________________________________________
Received on Friday, 11 January 2013 03:21:42 UTC