W3C home > Mailing lists > Public > www-tag@w3.org > February 2008

Partial review of Extending and Versioning XML Languages

From: Norman Walsh <ndw@nwalsh.com>
Date: Wed, 20 Feb 2008 15:01:22 -0500
To: www-tag@w3.org
Message-ID: <m2d4qrxur1.fsf@nwalsh.com>
I've been reviewing the 4 July 2007 draft of [Editorial Draft] Extending
and Versioning Languages: XML Languages:
http://www.w3.org/2001/tag/doc/versioning-xml-20070704.html

Section 1.1, para 1, "An XML Language is an where all the Texts MUST..."

  There seems to be something missing.

Section 1.1, diagram

  Despite the fact that we've spent a fair bit of time talking about the
  diagram, I don't find it very helpful. I find it particularly difficult
  to get my head around "Information Set" as distinct from "XML Information
  Set" and "Act of Consumption" as distinct from "Act of Interpretation".
  What is the relationship between "string_set" and "semantics"?

  I don't find a description of "string_set" in either this document or
  part 1.

Section 2, bullet 1, "Often this results in Schemas that are incomplete in..."

  Incomplete in what sense?

Section 2, bullet 2, "...like XHTML's P element"

  But there is only one namespace for the XHTML P element, no?

Section 3, initial list

  I think "different schemas" should be added to that list. I might
  have a vocabulary with six versions and assert the rule that if an
  instance is valid according to one (or more) of those schemas, then
  that is the version (or versions) of that document.

Section 3.2, para 1

  I really don't understand this paragraph. I think I have a sense for
  what "top typing" is, but I don't get "bottom typing" at all.

Section 3.2, para 3

  What's the distinction between component definition and component type?
  The introduction talks about "component" but doesn't mention either
  "component definition" or "component type".

Section 3.3

  Why does it follow that "a schema cannot refer to the correct middle type"?

Section 5.2.1, para 3, "This allows producers to extend instances without
changing the extension element's parent namespace"

  How does a must-understand flag allow this?

I'll try to post more comments on the rest of the document in the next
day or two.


                                        Be seeing you,
                                          norm

-- 
Norman Walsh <ndw@nwalsh.com> | To rule one's anger is well; to prevent
http://nwalsh.com/            | it is better.--Tryon Edwards

Received on Wednesday, 20 February 2008 20:01:45 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Thursday, 26 April 2012 12:47:52 GMT