Re: ISSUE-101 (plainliterals): equating plain literals with no language tag to xsd:string

On 14 Mar 2008, at 11:29, Michael Schneider wrote:
[snip]
>   Plain literals are considered to denote themselves,
>   so have a fixed meaning.
>
>   The denotation of a typed literal is
>   the value mapped from its enclosed character string
>   by the datatype associated with its enclosed type.

They have a fixed meaning too, at least, relative to OWL/RDFS axioms.

> So for plain literals the semantic interpretation is simply:
>
>   I_{plain}("foo") = "foo"

Sure.

> The second statement about typed literals reads a bit weird,

Is it? It's basically the same thing as above but the particulars of  
the mapping are parameterized by the datatype.

Consider if we coined a datatype "plainliteral". Its lexical-to-value  
space mapping could just ben the "self denotation" one.

> but I
> understand it for xsd:string in the following way: The part /"foo"/ of
> /"foo"^^xsd:string/ is mapped by the datatype (regarded as a  
> mapping from
> syntactical literals to values)

 From the lexical space to the value space. It's exactly the same as  
in plain literals except that the mapping function is indexed to the  
datatype.

> to the string "foo", since the value space
> of xsd:string consists of all strings.

Well, not all:
http://www.w3.org/TR/xmlschema-2/#string

"""The ·value space· of string is the set of finite-length sequences  
of characters (as defined in [XML 1.0 (Second Edition)]) that ·match·  
the Char production from [XML 1.0 (Second Edition)]"""

> Thus:
>
>   I_{xsd:string}("foo"^^xsd:string) = "foo"
>
> So, although /intentionally/ the semantics of xsd:string and plain  
> literals
> differ

No, they don't, afaict. I don't know what you mean by "intentionally".

> (the interpretation functions are defined in different ways,

Different definitions don't necessarily indicate different intensions  
or intentions.

> and
> their domains are obviously disjoint),

? You mean the lexical spaces are disjoints? That's true. I'm not  
sure what it matters...

Consider the datatype defined as the union of nonNegativeInteger and  
negativeInteger. It has the same value space as integer but a  
different lexical space (at least in OWL because in OWL the literals  
include, in a weird way, the datatype; in XML they lexical spaces  
would be the same). I don't see an "intentional" or "intensional"  
difference in these definitions.

> /extentionally/ they are the same
> (they produce the same output for "isomorphic" input).

It certainly would have been more transparent to just declare that  
plain literals were a lexical variant of xsd:string literals, which I  
guess is the real point here :)

> Well, works for me! I hereby withdraw my proposal to reject this  
> issue. :)

Ok.

Cheers,
Bijan.

Received on Friday, 14 March 2008 12:10:47 UTC