ACTION 2003-03-14#7 Editor's working draft (Semantics), review.

Review as per above action: in particular, review with an eye on changes
arising out of pfps -04 -05 -06 -07 -08 -10

In general, this is ok, however there are still two things I think need
addressing.


The first is the wording in section 3.1 to do with the denotation of
XMLLiterals. The text in question is preceeded by

@@The wording of this needs to be aligned with the terminology used in
Concepts@@

and I'd agree completely: in fact, I think it'd be useful to have JJC
cast an eye over this to ensure once and for all that what semantics and
concepts say on this are totally in sync.


The second point is that there seems to be an ambiguity in the text
dealing with datatyped literals in general. This ambiguity is at the
core of PFPS' continued objections as discussed at the telecon today (I
think).

The issue was, what if eg:foo owl:sameIndividualAs rdf:XMLLiteral?

Firstly I would note that section 3.4 (the new text) seems to rule out
such aliasing where eg:foo and rdf:XMLLiteral denote datatypes.

However, both in section 3.4 and section 4.3 it is not clear what the
ddd in the following form is:

   "aaa"^ddd

From context, it would APPEAR that the ddd is a uriref. However it is
used somewhat interchangeably with the datatype denoted by ddd -
particularly in section 4.3.

If ddd is a lexical URIref only, then the

    "xxx"^eg:foo =? "xxx"^rdf:XMLLiteral

issue looks to be trivially solved; there's no longer a blur between
lexical form in the abstract syntax and denotation. I think the text
would benefit from an explicit definition of the "ddd" - that could
simply be changing the text

	For any typed literal "sss"@ddd in G,

to something like

	For any typed literal "sss"@ddd (with ddd a URIref) in G,

In section 4.3, the clarifying change would be something like:
Before:

[[
This corresponds to the following rule, for each string sss that is a
legal lexical form for the datatype ddd.
]]

After:
[[
This corresponds to the following rule, for each string sss that is a
legal lexical form for the datatype I(ddd).
]]


-- 
jan grant, ILRT, University of Bristol. http://www.ilrt.bris.ac.uk/
Tel +44(0)117 9287088 Fax +44 (0)117 9287112 http://ioctl.org/jan/
Ceci n'est pas une pipe |

Received on Friday, 21 March 2003 11:32:42 UTC