Various issues on WG11N173

[David Price]

All,

A quick response to a few of Pascal's issues:

1, 2, 11 and 12 - The audience of Part 28 is an implementor who is fluent in
EXPRESS and the XML 1.0 Recommendation. Therefore, repeating the definitions
found in the XML spec for ?, *, +, %, ENTITY, NOTATION, etc. in the XML
syntax is not appropriate info to be included in Part 28.  The XML Spec is
freely available on the Web and, to be frank, if P28 becomes popular it's
more likely the readers won't know EXPRESS than that they won't know XML.

Thanks,
David

 -----Original Message-----
From: owner-wg11 at steptools.com [mailto:owner-wg11 at steptools.com]On Behalf Of
Pascal Huau
Sent: 2001 October 19 07:49
To: WG11
Subject: Various issues on WG11N173


  Dear all,


  please find hereafter issues against the current draft (WG11N173) of
Part28:

  1 - there is no information about the element DOCTYPE:
      - is it mandatory? For all the bindings?
      - what is its required content?
      - what is the meaning of the value SYSTEM found in the examples?

  2 - there is no information about the meaning of the symbols ?, *, +, %
in the specification of the elements (e.g. see 6.1)

  3 - OSEB: it is not clear how an entity that has several supertypes shall
be processed when they inherit from their supertypes attributes that have
the same name

  4 - OSEB: it is not clear whether a value of a derived attribute or of an
inverse attribute shall or shall not be provided in XML data files. Add a
subclause in §9.9 about that

  5 - OSEB: it is not clear whether a value shall or shall not be provided
in XML data files for Where rules or Unique rules. Add a subclause in §9.9
about that.

  6 - OSEB: it is not clear how an instance of an entity shall be encoded
when this entity has a derived attribute, inherited from a supertype
      - How shall the value of the derived attribute be encoded?
      - shall a value be provided for the attribute of the supertype?

  7 - OSEB: it is not clear how an attribute value that is an empty
aggregate shall be encoded (e.g. if the value of a inverse attribute is to
be exported)
      - shall the attribute be present or not? (NB: there may be a problem
with the #REQUIRED in the !ATTLIST of aggregate types)
      - how the empty collection shall be represented? enpty string ""?

  8 - annexes E and G deal with about the same topic (URN). They should be
merged in one annex.

  9 - annex E: the name of the AIM schema of AP203 is config_control_design
(without _schema)

  10 - annex E: the way an Express schema is identified is not satisfactory
as the concept of version is not so clear (becuase intermediate drafts, TCs
and amendments need also to be considered.
  An Express schema should only be identified by its ASN1 registration code
and Part28 should specify:

  urn:iso10303-28:p28_binding/iso_standard-isonum/schema
identifier/schema_name


  11 - §10.3 - The meaning and the structure of the element !NOTATION should
be explained

  12 - §10.3 - The meaning and the structure of the element !ENTITY should
be explained, especially because of the name of the element.


  Regards,
  Pascal Huau
  Association GOSET
  107,111 avenue Georges Clemenceau
  92000 Nanterre
  FRANCE
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://lists.steptools.com/pipermail/wg11/attachments/20011019/34656e6e/attachment.html