Re: Proposal for Resolution of LC99.: Message Reference Component is Underspecified

done. thx.

On 4/21/05, Arthur Ryman <arthur.ryman@gmail.com> wrote:
> Here is my revised proposal.
> 
> LC99 [1] : The Message Reference Component description doesn't define the
> semantics of the message when the optional {message content model}
> property is absent.
> 
> I propose to make the property REQUIRED and to set it to #other if the
> element attribute is missing. This means that the message content is
> defined by a non-XML extension type system.
> 
> We should add a token, #other, to the allowed values (in addition to
> #any and #none). The meaning of #other is that the message content
> model is defined by a non-XML extension type system.
> 
> The spec should be changed as follows:
> 
> 1. Change 2.5.1 The Message Reference Component
> 
> Current text: {message content model} OPTIONAL
> 
> Proposed text: {message content model} REQUIRED
> 
> 2. Change Table 2-6. Mapping between Message Reference Component
> Properties and XML Representation , {message content model} Row
> 
> Current text: If the element attribute information item is present and its
> value is a QName, then #element. Otherwise the actual value of the element
> attribute information item, if any, otherwise empty.
> 
> Proposed text: If the element attribute information item is present and
> its value is a QName, then #element. Otherwise the actual value of the
> element attribute information item, if any, otherwise #other.
> 
> [1] http://www.w3.org/2002/ws/desc/4/lc-issues/issues.html#LC99
>

Received on Thursday, 21 April 2005 18:05:57 UTC