Re: suggestions about "RDF* and SPARQL*" draft

Bob, thanks a lot for looking into these details (and thanks Pierre-Antoine 
for addressing almost all of Bob's points).

Find my responses to the remaining points below.

On tisdag 1 december 2020 kl. 11:46:18 CET Pierre-Antoine Champin wrote:
> [...]
> > - "the purpose of this section will be to provide an informal
> > introduction" How long do you want it? I could draft something.
> 
> Feel free to submit a PR

Bob, thanks a lot for offering to draft this section! This would be very much 
appreciated!!

To answer your question: it is difficult to say. It should be as long as it 
needs to be but not longer ;-)  I mean, it should describe the major aspects 
of the approach in a tutorial/primer style with examples; in this sense, it 
should cover both the RDF* side as well as the SPARQL* side.

> [...]
> Again, I expect the motivation and overview section to provide this kind
> of information. So hopefully, this will make section 2 easier to understand.
> > - "Each named graph is a pair consisting of an IRI or a blank node
> > (the graph name), and an RDF* graph" This would be clearer as "Each
> > named graph is a pair consisting of either an IRI or a blank node
> > serving as the graph's name and an RDF* graph". This plays up the
> > importance of role played by this IRI or blank node more.
> 
> it is more than stating a role, it is a definition. I changed it to
> "(called the graph name)" in order to make it clearer.

I have additionally added the word "either" as suggested by Bob.

> [...]
> > - After the reference to "function eval(D(G), algebra expression)"
> > there are two references to "function eval" without showing the
> > parameters or any formatting of "eval" to show that it's a reference
> > to a syntax expression (although I suppose a high-level one). I think
> > saying "the eval function" instead of "function eval" would read
> > better in those two references.
> 
> I leave this one to Olaf. ;)

Good point. I have rephrased these two occurrences as suggested, and I have 
linked the word "eval" in these cases to the corresponding section of the 
SPARQL REC.

> [...]
> > - 4.4 after "following four properties:" the bulleted list looks like
> > the conversion from a sentence to a bulleted list wasn't quite
> > finished. The bulleted items shouldn't have the commas or "and" after
> > them.
> 
> I am not a native spealer, but in French, I would consider this more as
> a matter of taste than a strict rule. As this is Olaf's work, I leave it
> to him to make the fix or not.

I am seeing this style of bulleted lists in many research papers. Anyway, I am 
not a native speaker either. So, I have taken Bob's word on it and removed the 
commas and the "and" (and I have turned it into an enumeration list which 
makes it easier to refer directly to any of the four items in the list).

Thanks,
Olaf


> 
> > See the numbered list in 6.1, although the 4.4 ones don't need a
> > period because they're not complete sentences.
> > 
> > - "semantics, in order" drop comma

Received on Wednesday, 2 December 2020 13:55:55 UTC