[topicmapmail] PSIs?

[Murray Altheim]

Lars Marius Garshol wrote:
> * Lars Marius Garshol
> |
> | I agree it is bad for you to create URIs in their domain, but you
> | wouldn't be doing that in this case. OWL already uses the same URIs,
> | so you would just be following OWLs lead. I think it's much better for
> | us to use the same URIs as the W3C for what is certainly the same
> | subjects.   <URL: http://www.w3.org/TR/owl-ref/#rdf-datatype >
> 
> * Murray Altheim
> | 
> | OWL was published by the W3C. I don't see how this is relevant.
> 
> That means that the W3C has already defined these URIs as identifying
> what you are describing them as identifying in your PSI set. That
> means that you don't have to do anything except tell your users which
> PSIs to use and how to use them with topic maps.
> 
> In other words: you don't have to define any new URIs. RDF is already
> using the XSD ones. All you really need to do is annotate them with
> information about how to use them in topic maps.

RDF has created a set of URIs that correspond roughly to a subset of
XSD's datatypes, then added a few of their own and deprecated use of
some of the XSD ones. And those URIs are OWL's creation -- they don't
point into the XSD spec, just use its namespace (base URI). For example,

   http://www.w3.org/2001/XMLSchema#date

doesn't point into the XSD spec at "date", it points to a placeholder
document from February 2001. IOW, RDF did exactly what I'm proposing,
i.e., creating a new set of URIs, except they don't point anywhere.
Mine at least point *actually* to the place in the XSD where a human
can get a definition. I chose deliberately *not* to use the OWL ones
because they're broken in this sense. I could use the OWL ones, but
I'd gain nothing. I'd rather point at XSD directly than point at
XSD via OWL via RDF. That's silly. If someone thinks we need subject
identity with the OWL set, we can add subject identifiers within the
PSI topic map, either within or via a separate topic map.

Also, I'm not an OWL/RDF poodle. I don't care what RDF is doing. XSD
has provided a set of datatypes and I'm using them directly for Topic
Maps. Topic Maps don't currently rely on OWL and I have no reason to
to create that dependency. It's nice as a link, but I don't want to
rely on it, nor do I need to.

> * Lars Marius Garshol
> |
> | The base URL given by xml:base doesn't affect the resolution of the
> | IDs of <topic> elements. Officially this XTM issue isn't resolved yet,
> | but I don't see how it can be resolved any other way.
>  
> * Murray Altheim
> |
> | Yes it does. It's not an XTM issue, it's an XML issue that has already
> | been resolved. Read the spec (particularly the examples in section 3):
> | 
> |     http://www.w3.org/TR/xmlbase/
> | 
> | That's the whole point of xml:base. What other purpose could it
> | serve?
> 
> Please read the references I sent you. This *only* applies to URIs of
> the form "" and "#foo" and the resolution of <topic id>s into absolute
> URIs. All other kinds of URI references are resolved relative to the
> xml:base if one is specified. The thing is that "same-document
> references", as RFC 2396 calls them, are a special case. The XML Base
> specification doesn't discuss those at all.

I think you're misinterpreting fragment identifiers. If you read the
xml:base spec, fragment identifiers are resolved relative to the base
established by the URI reference in the xml:base attribute. And for
the purposes of publishing a PSI set I really could care less about
edge cases empty URI references. Really irrelevant.

Point is, if I put an xml:base attribute in the PSI set, the document
is portable in the sense that resolution of fragment IDs (used to
establish the PSIs) is always the same no matter where the document
sits.

> The relevant bit of RFC 2396 is section 4.2:
> 
>    "A URI reference that does not contain a URI is a reference to the
>    current document.  In other words, an empty URI reference within a
>    document is interpreted as a reference to the start of that
>    document, and a reference containing only a fragment identifier is
>    a reference to the identified fragment of that document.  Traversal
>    of such a reference should not result in an additional retrieval
>    action.  However, if the URI reference occurs in a context that is
>    always intended to result in a new request, as in the case of
>    HTML's FORM element, then an empty URI reference represents the
>    base URI of the current document and should be replaced by that URI
>    when transformed into a request."
> 
> See also 
>   <URL: http://lists.w3.org/Archives/Public/www-xml-linking-comments/2001AprJun/0158.html >

Your misunderstanding of Paul should not lead you to think that
xml:base doesn't work. Your example merely was in error, and he
answered accordingly. Paul is a stickler for details. E.g., if
your xml:base had included a document reference there'd be a
resource to use. He says this, if you read it again.

>   <URL: http://lists.w3.org/Archives/Public/w3c-rdfcore-wg/2002May/0060.html >
> 
> In addition to all this there is the logic given in the rationale that
> Graham and I put into our XTM resolutions proposal document that I
> linked to in my previous email.

Like it or not, the can of worms is already open, as xml:base is
already used in XTM. xml:base conforms to RFC 2396 by using the
concept of allowing a base URI to be embedded in a document:

   http://www.w3.org/TR/xmlbase/#rfc2396

If this is still unclear, check the examples at

   http://www.w3.org/TR/xmlbase/#syntax

These are very clear as to how to use xml:base. When it comes
to fragment identifiers, Paul's answer about the "bare name" form
of XPointer is simply his way of stating what I've been saying,
that you append a "#" and the ID reference to the document
reference, the same way as almost every link on the web is done.

In his paragraph beginning "The document infoitem's [base URI]
property..." you'll see where xml:base operates, by redefining
the base URI.

Murray

......................................................................
Murray Altheim                  <http://kmi.open.ac.uk/people/murray/>
Knowledge Media Institute
The Open University, Milton Keynes, Bucks, MK7 6AA, UK

    Boundless wind and moon - the eye within eyes,
    Inexhaustible heaven and earth - the light beyond light,
    The willow dark, the flower bright - ten thousand houses,
    Knock at any door - there's one who will respond.
                                     -- The Blue Cliff Record