W3C home > Mailing lists > Public > public-xml-processing-model-wg@w3.org > July 2010

base URI phrase in... New editors' draft of XML Processor Profiles

From: Grosso, Paul <pgrosso@ptc.com>
Date: Wed, 14 Jul 2010 11:39:35 -0400
Message-ID: <9B2DE9094C827E44988F5ADAA6A2C5DA5F33B0@HQ-MAIL9.ptcnet.ptc.com>
To: <public-xml-processing-model-wg@w3.org>

> -----Original Message-----
> From: public-xml-processing-model-wg-request@w3.org
> processing-model-wg-request@w3.org] On Behalf Of Henry S. Thompson
> Sent: Wednesday, 2010 July 14 6:33
> To: public-xml-processing-model-wg@w3.org
> Subject: New editors' draft of XML Processor Profiles
> Now available, with a new (not what I had expected) approach to
> invariants:
>   http://www.w3.org/XML/XProc/docs/xml-proc-profiles.html
>   http://www.w3.org/XML/XProc/docs/diff.html

A perhaps minor issue, but I'm disturbed by the various
uses of "base URI" in this document.  I was first alerted
to this when I noted that, in the terminology section, the
definition of base URI references RFC 3986.  I would have
expected it to reference either XML Base or Infoset.

So I looked at uses of the phrase throughout, and became
more concerned/confused.

The phrase occurs:

1.  In the terminology section with a reference to RFC 3986.

2.  In point 2 of each of the four profiles where the phrase
    is "base URI property...in conformance with XML Base".

3.  In section 3.1 of the invariants where it is shown as
    [base URI] and is pretty clearly representing an Infoset

It is not really clear what the phrase "base URI property" 
in section 2 means.  If "base URI" in this spec is really 
defined by RFC 3986, then it makes no sense to say "base
URI property" in section 2.  And [base URI] in section 3.1
is certainly a reference to the Infoset.

I see two choices:

a.  Leave the reference to RFC 3986 in the terminology section
    and delete the word "property" from the phrase "base URI
    property...in conformance with XML Base" in section 2;

b.  Change the terminology section to say:
        The term *[base URI] property* is used in this specification
        as it is defined in [XML Information Set (Second Edition)].
    and then in section 2, replace:
        base URI property of each element in conformance with [XML
        [base URI] property of each element in conformance with 
        [XML Information Set (Second Edition)].
    (Note that Infoset says:
        Several information items have a [base URI] or [declaration
        base URI] property. These are computed according to [XML Base]. 
    so a normative reference to Infoset is a normative reference
    to XML Base.)

Regardless of any of the above, I think we need to review our
normative and non-normative references.  In particular, given
section 3, I would think that Infoset should be a normative

If we go with my choice b, RFC 3986 is no longer a reference.

The latest NS 1.0 version is the third edition, not second.

Received on Wednesday, 14 July 2010 15:40:18 UTC

This archive was generated by hypermail 2.3.1 : Tuesday, 6 January 2015 21:32:48 UTC