Re: xmlsch issues discussion

Dave

I think you've misunderstood some of these - I get the impression that the XML 
Schema WG have tried hard not to be antagonistic with their comments.
e.g. they say
[[
Other
    members of the WG feel that it verges on disrespect to assume that you
    need instruction on this point.
]]
at a point where, IMO, they have just made a helpful comment. (Not on syntax 
though).

ISSUE http://www.w3.org/2001/sw/RDFCore/20030123-issues/#xmlsch-09

> "We were unable, on a first reading, to determine whether the default
>    namespace declaration, and thus unprefixed names, were or were not
>    allowed in order to encode 'RDF URI References'."

I understood this as asking, is:

<RDF xmlns="&rdfns;">
</RDF>

a legal RDF/XML document (assumning entity correctly defined).

My understanding is that it is; but, without looking, I wouldn't be surprised 
if the text could do with editorial polish on this point.

[[
ISSUE http://www.w3.org/2001/sw/RDFCore/20030123-issues/#xmlsch-10
"cannonical syntax"

Discussion of out of what I believe is out of RDF Core's charter,
creating a new XML syntax or modifying RDF/XML (created 1998)
to work with XML technology designed later.

  "The RDF Core WG is neither chartered to develop a new RDF syntax,"
  -- http://www.w3.org/2001/sw/RDFCoreWGCharter

Summary: closed.  we don't have to do this, out of charter ??
]]

I am not convinced by the "out of charter" argument. My reading of the 
suggestion is that if we specified a subset of the current syntax that could 
serilalize any graph then expressing some RDF within that subset would 
potentially allow XQuery to work better with it.

This would loosely correspond to the basic syntax in M&S.

e.g.
We could permit:
RDF
nodeElement
nodeElementList
ws
propertyEltList
propertyElt (but not with rdf:li)
literalPropertyElt
parseTypeLiteralPropertyElt
emptyPropertyElt
nodeIdAttr
aboutAttr
resourceAttr
datatypeAttr  (but not with rdf:XMLLiteral)
parseLiteral
URI-reference
literal

I believe that with that, possibly subject to the further constraints:
- that each nodeElement must be uniquely identified by an rdf:about or 
rdf:nodeID

would result in a much more regular subset of the syntax which could be used 
more effectively with XML technologies.
(I have excluded:
  propertyAttrs
  reification syntax
  collection syntax
  container syntax
  all the parseTypes except for Literal - even that possibly should be 
excluded
)


[[
ISSUE http://www.w3.org/2001/sw/RDFCore/20030123-issues/#xmlsch-11
"layering on xml"

There is some explanation to do since they seem to have missed
reading the Syntax Data model section in defining RDF/XML on top of
the pile of XML technologies.

>    Not only does this confuse levels and thus readers, it also runs the
>    risk of inadvertently defining an XML subset. It also appears, on a
>    strict reading, to rule out XML documents not derived from the parsing
>    of character streams as possible RDF/XML (so that it would be
>    illegitimate to regard a data structure created using a DOM interface,
>    for example, as RDF/XML).

It neither creates a subset nor rules out using other XML APIs as
long as an XML Infoset is available.
]]

I tend to agree with your reading, but would see no harm in a more 
conciliatory addition e.g.
in section 6
http://www.w3.org/TR/2003/WD-rdf-syntax-grammar-20030123/#section-Data-Model

The first paragraph talks about "sequence of events", in a way which, to me, 
clearly does not presuppose the existence of a document to hand.

We could consider an addiitonal sentence, at the end of that first paragraph, 
along the lines of:
[[
The sequence of events may arise from a non-document representation
of an XML infoset, for example,. a DOM tree.
]]

On a personal level I would prefer the spec not to say that, because I am 
getting complaints that ARP does not support such functionality! (Not that I 
would admit that it didn't conform if we made the change!)

[[
ISSUE http://www.w3.org/2001/sw/RDFCore/20030123-issues/#xmlsch-12
"capricious syntax"
]]

While your out-of-charter stance is technically correct, and formally our only 
option, I wonder whether the long term good of the semantic web would be 
aided by us indicating that we agree with much of this comment. We could 
point to the new work being started with the HTML WG on embedding RDF inside 
validated xhtml.

It may also help for us to identify those usecases for which RDF/XML *is* a 
good match. There are a few, and it would be good to enumerate them.

Another intersting thing to consider on this one is that from an XML point of 
view one of the more capricious aspects of RDF/XML is its use of qnames.
However, this use appeals to the semantic web community as a whole, even those 
who try not to touch RDF/XML with a bargepole.

Jeremy

Received on Wednesday, 12 March 2003 08:59:41 UTC