ISSUE-92: Possible compromise from Holger Knublauch on 2015-11-12 (public-data-shapes-wg@w3.org from November 2015)

From: Holger Knublauch <holger@topquadrant.com>
Date: Fri, 13 Nov 2015 07:04:18 +1000
To: public-data-shapes-wg <public-data-shapes-wg@w3.org>
Message-ID: <5644FED2.1010701@topquadrant.com>

One technique of moving forward with conflicting proposals is to include 
them all, but into different dialects. This was done by many other W3C 
groups before, and while unpopular, may be the only achievable solution 
once again. We did this before by introducing SHACL Core and SHACL with 
SPARQL extensions.

I'd be OK with some variation of Arthur's proposal to ISSUE-92 if we 
make it non-mandatory, just like the SPARQL extensions are not 
mandatory. The cost of implementing the partitioning is high (both for 
engines and UI tools), and performance will likely be slow too. It also 
isn't a use case that I have encountered often, and a work-around to use 
SPARQL already exists.

What do others think?

Holger

Received on Thursday, 12 November 2015 21:04:51 UTC