Internationalization last call comments on Infoset

Dear Infoset WG,

Here are the last call comments from the i18n WG/IG,
after discussions in the IG (see
http://lists.w3.org/Archives/Member/w3c-i18n-ig/2000Jan/0000.html
and followups) and in the WG (see
http://www.w3.org/International/Group/2000/01/telecon1/minutes#OtherDocuments).

- in 2.6.2.3, add the list of predefined XML entities
  to avoid misunderstandings.

- 2.9.2.7: Currently, this says:
  [charset] The name of the character encoding in which the entity is
  expressed. This property is derived from the XML or text declaration
  optionally present at the beginning of the document entity or an external
  entity respectively.

  This must be changed to take into account that the information
  can come from outside the document (e.g. HTTP header). I.e. the
  text must make excruciatingly clear whether this is the the
  encoding in which the entity was expressed (which may have
  been derived as UTF-8 or UTF-16 by the XML default rules,
  or may have been derived from external information, e.g.
  an HTTP header, in cases where the 'encoding' pseudoattribute
  is missing or even conflicting with the external information),
  or whether this is just the literal value of what is on an
  eventual 'encoding' pseudoattribute, which may be different
  from the character encoding in which the entity is expressed.

  Also, it looks like this only makes sense for external entities;
  it is not clear whether it's also possible to have this
  information for internal entities. Please clarify.

  In addition, we discussed whether we would prefer a different name
  for this, e.g. [encoding] instead of [charset]. The WG prefers
  the current name; arguments for both solutions can be found in
  the mail archive cited above.

Please feel free to contact me if there are any further questions.


Regards,   Martin.


#-#-#  Martin J. Du"rst, World Wide Web Consortium
#-#-#  mailto:duerst@w3.org   http://www.w3.org

Received on Sunday, 23 January 2000 20:23:20 UTC