update RDF dataset discussion #151
Conversation
Add explicit comparison of SPARQL 1.2 and RDF 1.2, vs SPARQL 1.1, RDF 1.1, and RDF 1.0. Might also want specific difference with SPARQL 1.0. This specificity may not be needed for users who have never before encountered RDF or SPARQL, but I think it helpful for users who *have* worked with any previous version(s) — like me.
|
Note that we do discuss changes to RDF in the Abstract:
RDF 1.0 did not have named graphs, and therefore a discussion of the use of IRIs for graph names is not appropriate. (SPARQL 1.0 had it's own concept of named graphs). |
| @@ -1021,11 +1021,13 @@ <h2>RDF Datasets</h2> | |||
| can avoid interoperability issues by not ascribing importance to | |||
| the presence or absence of empty named graphs.</p> | |||
|
|
|||
| <p>SPARQL 1.2 [[SPARQL12-QUERY]] uses | |||
| <p>Version 1.2 of the SPARQL Query Language [[SPARQL12-QUERY]] uses | |||
There was a problem hiding this comment.
It's all SPARQL specs which is why it says SPARQL 1.2.
The link is provided because that is the starting point. Arguably, the link could be removed.
There was a problem hiding this comment.
The link is provided because that is the starting point. Arguably, the link could be removed.
It seems odd for [SPARQL12-QUERY] to be the starting point.
[SPARQL12-CONCEPTS] or possibly [SPARQL12-NEW] makes more sense to me, and not only because these are listed before [SPARQL12-QUERY] in the List of Documents (to which I cannot find a way to link directly, nor to auto-expand after load)
It's all SPARQL specs which is why it says SPARQL 1.2.
It seems to me, this lends weight to linking to [SPARQL12-CONCEPTS] or possibly [SPARQL12-NEW] instead of [SPARQL12-QUERY].
Removing the link is indeed also an option.
| and RDF version 1.0 | ||
| only allowed graph names to be IRIs.</p> |
There was a problem hiding this comment.
Better to remove this.
RDF 1.0 didn't mention datasets at all nor were they concurrent.
This is a note. RDF 1.1 is 10 years old.
I don't know about you, but when I am looking things up in a spec, and even when I am trying to read the entire document, I often overlook things in the Abstract (because I don't understand enough to absorb them properly) — and may even skip it entirely, because I expect it to echo things said later but at a higher level of, sorry, abstraction. For these reasons, I suggest that anything we mean for the reader to actively digest should be placed elsewhere in the document, even if it is also said (preferably not word-for-word) in the Abstract.
My bad, probably. I should have re-re-re-re-re-re-read the RDF 1.0 & 1.1 and SPARQL 1.0 & 1.1 docs more closely. (In fact, I did go back to those docs, and based my own writing on what they said. I should have linked to my source material. I'll see if I can re-find it.) |
Add explicit comparison of SPARQL 1.2 and RDF 1.2, vs SPARQL 1.1, RDF 1.1, and RDF 1.0. Might also want specific difference with SPARQL 1.0.
This specificity may not be needed for users who have never before encountered RDF or SPARQL, but I think it helpful for users who have worked with any previous version(s) — like me.
Preview | Diff