Re: Update to proposal report.

On 03/23/2017 08:58 AM, Andy Seaborne wrote:
> 
> 
> On 23/03/17 15:24, Peter F. Patel-Schneider wrote:
>> I am confused as to how this community group is supposed to work.
>>
>> As far as I know there are two proposals that have been put forward and no
>> consensus has been achieved.  This report contains only one of the two
>> proposals.  Why is the other proposal not included?
> 
> There has been no write-up. The other proposal is in at least two emails,
> where one problem was identified with BIND.
> 
> There has been no response to the problems it has for FILTER, GRAPH, UNION and
> MINUS except to describe them as "weird"; it is further from the current
> specification.  If there is no scope to change, there isn't much to discuss.
> There is no systemic characterisation of what changes, nor has it been shown
> that there are different queries that can written at all.
> 
>     Andy
> 


The initial message for Proposal A is
https://lists.w3.org/Archives/Public/public-sparql-exists/2016Sep/0012.html
which mirrors a paper presented at ISWC 2016.  For reasons of space the
paper omits a related change to variable scoping.  This was accidentally
omitted in the message as well.  The scoping change is:

\item Modify the scoping rules so that variables in-scope at a {\sf FILTER}
   are in-scope at the beginning of the pattern argument to any {\sf EXISTS}
   in the {\sf FILTER} expression.  This is independent of the change to a
   mapping-based definition but fixes an error that affects {\sf EXISTS}.

The message identifying Proposal A and Proposal B is
https://lists.w3.org/Archives/Public/public-sparql-exists/2016Oct/0006.html
which has only a short description of the proposals.

I can see messages describing how Proposal A works, but no messages about
technical problems with the proposal.

Proposal A has the main goals of being well-defined and simple.  It has
secondary goals of covering as many SPARQL queries as possible. and
conforming to SPARQL notions such as variable scope and bottom-up sub-query
evaluation.  It sacrifices some compatability to achieve these goals.

I'm quite willing to produce a new writeup of Proposal A for inclusion in
https://w3c.github.io/sparql-exists/docs/sparql-exists.html


Peter F. Patel-Schneider
Nuance Communications

Received on Saturday, 25 March 2017 17:48:29 UTC