More semantic comments

Global sp:
truthvalue or truth-value (both occur)

After figure 1

[["whatever"^^ex:b]] => [["whatever"^^<ex:b>]]


[[1.6 Unlabeled nodes ...]] => [[1.6 Blank nodes ...]]

suggest-delete
[[See http://www.w3.org/...#rdfms-identity-anon-resources for a
sumamry and pointers to further discussion on this issue.]]

[[unlabeled nodes]]=>[[blank nodes]] * 2

[[Notice however that since two unlabeled nodes cannot
have the same labal, ]]
==>
[[Notice however that since two different blank nodes are not equal, ]]


before Monotonicity Lemma
[[
Finally, the following is a trivial
<<suggest-add: but important>>
consequence ... entailment,
<<suggest-delete:
but it may be worth stating ...
satisfy it.
 >>
]]
(I don't think it's appropriate or clear to talk about many implemented 
systems here)

3.1
possible spell construing or constructing? subsets of the universe

3.2
[[When the RDFS vocabulary is added, ... semantic conditions.]]
unclear
suggest-replace (in brackets)
[[(The RDFS part of the formal semantics does cover domain and
range conditions for the properties used in this vocabulary)]]

Given that you have gone for the MAY vocabulary on extensions, I think 
in 3.3 you are trying to say:

[[.. lesser burden on implementors ]]
suggest-add:
[[
Semantic extensions MAY strengthen the semantics to iff.
]]

3.3.1
[[There is however no way in current RDF to specify exactly ...]
==>
[[It does not specify exactly ..]


3.4

I found the following unintelligible, possibly because I don't have 
access to the XSD example in your (or mayber Peter's) head.

[[Users shoudl take care to distinghuish the value space ...
identical when viewed as class members.]]


4.
<<suggest-delete>>
[[Implementors who wish to check any kind ...
backchaining on the clousre rules, for example.]]

(any normative suggestions about impl techniques may degrade the value 
of others)


That's it - no more comments at this stage.

Jeremy

Received on Monday, 11 November 2002 07:45:14 UTC