Annotation syntax for Turtle* #58
Conversation
|
PR #10 was based on an old version of the document (when it was still called |
pchampin
changed the title
|
@gkellogg I requested a new review mostly for the informative note I added on SA-mode + PG-mode |
|
Yes, the updated language looks good, and we should resolve any other outstanding issues pertaining to PG-mode and SA-mode in favor of the SA interpretation of |
Co-authored-by: Ted Thibodeau Jr <tthibodeau@openlinksw.com>
rdf-star-cg-spec.html
Outdated
| @@ -873,7 +873,7 @@ <h2>SA-mode and PG-mode</h2> | |||
|
|
|||
| <ul> | |||
| <li>The <code><< ... >></code> syntax represents an <a>embedded triple</a> without asserting it, satisfying the need formerly filled by SA-mode.</li> | |||
| <li>The <code>{| ... |}</code> syntax creates triples where the subject is interpreted as an <a>embedded</a> version of the triple previously requiring distinct <a>assertion</a>, without the need to repeat it, satisfying the need formerly filled by PG-mode.</li> | |||
| <li>The <code>{| ... |}</code> syntax (annotation) creates triples where the subject is interpreted as an <a>embedded</a> version of the triple <a>asserted</a> just before the annotation, without the need to repeat it, satisfying the need formerly filled by PG-mode.</li> | |||
There was a problem hiding this comment.
| <li>The <code>{| ... |}</code> syntax (annotation) creates triples where the subject is interpreted as an <a>embedded</a> version of the triple <a>asserted</a> just before the annotation, without the need to repeat it, satisfying the need formerly filled by PG-mode.</li> | |
| <li>The <code>{| ... |}</code> annotation syntax creates triples where the subject is interpreted as <a>asserting</a> the triple <a>embedded</a> just before the annotation, without the need to repeat it (i.e., to explicitly assert it), satisfying the need formerly filled by PG-mode.</li> |
There was a problem hiding this comment.
Exactly what's trying to be said is not 100% clear, so I'm not sure this says it ... but I think it's closer than it started.
There was a problem hiding this comment.
No sorry, I think we are getting further from what I'm trying to say (but this is cumbersome to put in prose).
:a :b :c {| :p1 :o1; :p2 :o2 |}
The annotation is the {| ... |} part. The triple :a :b :c is asserted, as it would be if there was no annotation. What the annotation is doing is creating triples X :p1 :o1 and X :p2 :o2, where X is << :a :b :c >>, an embedded version of :a :b :c, the triple asserted just before the annotation.
I think the "is interpreted as" is confusing and should be removed. I also agree that getting "annotation" out of parenthesis is better. Finally, I'll add :p :o inside the annotation syntax example, to hint at the fact that it contains predicate-object pairs, but no explicit subject.
There was a problem hiding this comment.
@TallTed what do you think now?
Co-authored-by: Ted Thibodeau Jr <tthibodeau@openlinksw.com>
|
This was discussed in today's call: https://w3c.github.io/rdf-star/Minutes/2020-12-11.html#item05 |
| <p>The notion of <a href="#grammar-production-annotation">annotations</a> in the <a href="#turtle-star">Turtle*</a> syntax was introduced to remove the need for different modes. Rather than interpret the same syntax differently in each mode, which would have caused interoperability problems and required a switch for those modes, it was decided to provide a different syntax for each use case.</p> | ||
|
|
||
| <ul> | ||
| <li>The <code><< ... >></code> syntax represents an <a>embedded triple</a> without asserting it, satisfying the need formerly filled by SA-mode.</li> |
There was a problem hiding this comment.
| <li>The <code><< ... >></code> syntax represents an <a>embedded triple</a> without asserting it, satisfying the need formerly filled by SA-mode.</li> | |
| <li>The <code><< ... >></code> syntax represents a triple that is <a>embedded</a> without being <a>asserted</a>, satisfying the need formerly filled by SA-mode.</li> |
|
@pchampin -- I think we're getting closer. I've tweaked both paragraphs. See what you think? |
|
Oops ... missed the merge. |
last tweak to #58, which was suggested after commit
Preview | Diff