W3C home > Mailing lists > Public > www-xml-schema-comments@w3.org > July to September 2007

[Bug 3226] Terminology: "character string"

From: <bugzilla@wiggum.w3.org>
Date: Mon, 17 Sep 2007 20:13:47 +0000
CC:
To: www-xml-schema-comments@w3.org
Message-Id: <E1IXMyZ-0002eX-TE@wiggum.w3.org>

http://www.w3.org/Bugs/Public/show_bug.cgi?id=3226


cmsmcq@w3.org changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
           Keywords|                            |needsDrafting




------- Comment #3 from cmsmcq@w3.org  2007-09-17 20:13 -------
The XML Schema Working Group discussed this issue in its telcon of
7 September 2007 and instructed the editors to prepare a wording
proposal with the following properties:

  - The spec should use the words 'string', 'integer', etc.
    in their usual technical meaning.  (Optionally, existing
    circumlocutions like 'character string' may be changed
    to use the simpler form 'string'.)
  - When the spec needs to refer to a string, integer, etc. 
    qua member of the value space of a particular XSDL datatype
    (i.e. when we are speaking of a value identified, in the
    course of validation, as the value corresponding to the
    lexical form specified by a given information item), 
    some explicit phrasing should be used, not the simple words
    'string', 'integer', etc.
  - The spec should at some appropriate point make clear (a) that
    the terms 'string', 'integer', etc. are used in their normal
    technical sense, (b) that other phrasing (to be specified)
    is used when speaking of values qua members of the value
    space of a specific XSDL datatype, and (c) that of course
    all strings (as we use the term) are in fact members of the
    lexical and value spaces of xs:string (and similarly, mutatis
    mutandis, for integer), so that this is purely a question of
    connotation, not denotation.
  - The terms 'string', 'integer', etc. should be given formal
    definitions for each usage, and each occurrence of the terms
    should be hyperlinked to the appropriate definition, so that
    in case of doubt a reader can check what we think we mean.

I'm marking this as needsDrafting.
Received on Monday, 17 September 2007 20:13:53 UTC

This archive was generated by hypermail 2.3.1 : Wednesday, 7 January 2015 14:50:06 UTC