Lee Feigenbaum wrote: > A *key* part of driving a healthy standards process is extension & > innovation performed by implementations. If implementations blindly > adhered to the standard, there would be no experience to draw on to > motivate and coalesce around designs for future specifications. The sort > of innovation done by Virtuoso and many other SPARQL engines is key to a > healthy future for SPARQL. > > The one caveat is that I do agree that, in general, extensions really > ought to be compatible with the core standard. This means that if a > query is asked or a request made via the SPARQL protocol, it's almost > always best for the implementation to behave as per the spec. This > allows generic SPARQL clients to safely and consistently access a > service, while other clients with knowledge of your extensions can still > take full advantage of them. > > Lee
That is my opinion as well - and it basically echos what I just wrote to Kingsley in an private email. Regards, Michael ------------------------------------------------------------------------------ _______________________________________________ Dbpedia-discussion mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/dbpedia-discussion
