W3C home > Mailing lists > Public > www-xml-blueberry-comments@w3.org > July 2003

Re: XML 1.1 CR comment response for Lewis-01

From: Paul Grosso <pgrosso@arbortext.com>
Date: Mon, 14 Jul 2003 11:31:05 -0500
Message-Id: <>
To: "Amelia A. Lewis" <alewis@tibco.com>
Cc: www-xml-blueberry-comments@w3.org

At 11:59 2003 06 24 -0400, Amelia A. Lewis wrote:
>On Tue, 24 Jun 2003 11:17:38 -0400
>John Cowan <jcowan@reutershealth.com> wrote:
>> Amelia A. Lewis scripsit:
>> > Discussion of the issue revealed that x#D is included in S as part
>> > of compatibility with SGML; the discussion included a rather
>> > grotesque example of hackery that could get this code point to show
>> > up in a document, bypassing normalization.
>> Can you provide the details?
>See the thread beginning at
>(which raised this issue, I believe).
> From examination of that thread in the archive, it appears that the
>example of what one John Cowan referred to as "entity abuse" included
>solely for "backward compatibility" was communicated outside the
>archive.  I regret that I do not appear to have retained the email
>illustrating entity abuse intended to get x#D into the stream.
>> I wouldn't object to adding a motherhood note to the Third Edition
>> (and a fortiori to XML 1.1).
>Something of the sort would make me far more comfortable with rejection
>of this issue. 


The XML Core WG is considering adding a note as an erratum 
to XML 1.0 (thereby also incorporating it into XML 1.1) on
this subject.  The suggested wording currently reads as follows:

  To be inserted just after production 3:

        Note:  The presence of #xD in the above definition is
        maintained purely for backward compatibility with the
        First Edition.  As explained in 2.11 End-of-Line Handling,
        all #xD characters physically present in an XML document
        are either removed or replaced by #xA characters before
        any other processing is done.  The only way to get #xD
        characters into an XML document (as opposed to character
        references) is a complex trick involving character
        references within parameter entity definitions.

In light of this, would you like to amend your response to
our resolution of Lewis-01 which is currently recorded as:
  I'm not happy with this.

in the Disposition of Comments document at



Paul Grosso for the XML Core WG
Received on Monday, 14 July 2003 18:43:21 UTC

This archive was generated by hypermail 2.3.1 : Tuesday, 6 January 2015 20:13:20 UTC