Re: Review Comments

Souri - thanks for the comments.  I have made changes as noted below as they 
are editorial.

Changes in v1.417

[Eric - I didn't address the comments on section 11]

	Andy

Souripriya Das wrote:
> The following  minor comments are based on 
> 
> http://www.w3.org/2001/sw/DataAccess/rq23/ 1.413
> 
> Thanks,
> - Souri.
> 
> 1) In 10.1 [Solution Sequences and Result Forms] mainly:
> - In Some potential confusion may arise in readers' minds due to use of 
> following terms: Solution Sequence, Set of Solutions, Solution Modifier, 
> Solution Sequence Modifier.
> 
> Could some minor editing based on the following basics be useful?
> - A query pattern generates an unordered collection of solutions.
> - DISTINCT makes it a set.
> - ORDER BY makes it an ordered collection. In case of duplicate ordering 
> key values, the ordering could be partial.
> - Projection, LIMIT, and OFFSET would work on any collection (ordered, 
> partially ordered, unordered, or set). For LIMIT and OFFSET, working on 
> ordered collection makes more sense, so typically would be used with 
> ORDER BY.

The first paragraph of 10.1 has been changed to:

"""
Query patterns generate an unordered collection of solutions, each solution 
being a function from variables to RDF terms. These solutions are then treated 
as a sequence, initially in no specific order; any sequence modifiers are then 
applied to create another sequence.  Finally, this latter sequence is used to 
generate one of the SPARQL result forms.
"""
which tries to be more specific about "collections" and "sequences".

The definitions of solution sequence and solution sequence modifier have also 
been changed slightly, s/set/collection/ and explicitly stating the 
unorderedness of the sequence initially.

> 
> 2) In 10.1 [Solution Sequences and Result Forms]:
> - "Projection" section does not explicitly show an example. We could 
> refer to one.

Example added.

> 
> 3) In 11.2.3.2 [sop:bound]:
> - Typo in example1: change OPTIONANL to OPTIONAL

[Eric]

> 
> 4) In 11.2.3.10 [sop:logical-and]:
> - "... Mapping table [17] specifies a which ..." --> Remove the "a"

[Eric]

> 
> 5) In 5 [Including Optional Values] (before 5.1):
> - "... complete structure can not be ..." --> use "cannot"

This editor prefers "can not"

> 
> 6) In 2.7 [Blank Nodes]:
> - "...although it can not be ..." --> use "cannot"

This editor prefers "can not"

> 
> 7) In 8.1 [Accessing Graph Names]:
> - The query in the example does not seem complete until we use FROM 
> NAMED clauses to specify the set of named graphs that the variable, 
> ?src, can be bound to. So, proposing inclusion of the following and a 
> reference to Section 9 definition of FROM NAMED clause.
> 
>   FROM NAMED <http://example.org/foaf/aliceFoaf>
>   FROM NAMED <http://example.org/foaf/bobFoaf>
> 

The RDF dataset for a query need not be described in the query itself - it is 
executed against a dataset given externally to the syntax of the query.  To 
keep this clear, the description of datasets does not occur until sec 9 and 
sec 8 examples use the dataset noted in sec 8.

	Andy

Received on Saturday, 9 July 2005 16:37:39 UTC