- From: Jeremy J Carroll <jjc@syapse.com>
- Date: Wed, 27 Nov 2013 09:40:11 -0800
- To: public-rdf-dawg-comments@w3.org
Received on Wednesday, 27 November 2013 17:40:40 UTC
I find the text confusing, and believe editorial work on the paragraph would help.
http://www.w3.org/TR/2013/REC-sparql11-query-20130321/#aggregateExample2
[[
g avg c
<http://example.com/data/#x> 2.5 2.5
<http://example.com/data/#y>
<http://example.com/data/#z> 2.5 2.5
Note that the bindings for the :y group is not included in the results as the evaluation of Avg({1, _:b2, 3, 4}), and (_:b2 + 4) / 2 is an error, removing the bindings from the solution.
]]
My issue is that bindings for the :y group are included, specifically ?g is bound. Moreover this does answer questions about the more formal text in section 18 about aggregates.
I suggest the following rewrite:
[[
Note that in the bindings for the :y group neither ?avg nor ?c are bound, as the evaluation both of Avg({1, _:b2, 3, 4}) and of (_:b2 + 4) / 2 are errors.
]]
Jeremy J Carroll
Principal Architect
Syapse, Inc.
Received on Wednesday, 27 November 2013 17:40:40 UTC