RE: Import/Include - a proposal

I'm not sure one actually has to define an infoset model, as it can be
implied 
from any xml doc as long as certain restrictions are 
followed (as defined in infoset spec) - but we should verify this.

Martin.



>-----Original Message-----
>From: Yves Lafon [mailto:ylafon@w3.org] 
>Sent: 17 September 2004 14:27
>To: Martin Chapman
>Cc: 'Nickolas Kavantzas'; 'Steve Ross-Talbot'; 'Monica J. 
>Martin'; public-ws-chor@w3.org
>Subject: RE: Import/Include - a proposal
>
>
>On Thu, 16 Sep 2004, Martin Chapman wrote:
>
>> Surely to close this resolution properly we need to define the 
>> mechanism similar to Xinclude but for cdl documents which are not 
>> infoset based.
>
>We have two choices there:
>1/ have CDL defined in terms of infoset
>2/ define our own inclusion mechanism (with all the rulse 
>saying that it
>    is only a syntactic inclusion, etc...)
>
>SOAP 1.2 is defined in terms of infoset.
>WSDL 2.0 has its own data model, and they defined an inclusion 
>mechanism 
>different from XInclude for that reason.
>
>In 1/ we have to evaluate the impact on CDL, in 2/ we have to 
>argue why we 
>need that new inclusion mechanism (which is related to 1/), so 
>I have no 
>real preference, as long as we keep the original proposal 
>which is to have 
>a syntactic only inclusion mechanism (ie: including a variable 
>definition 
>form another choreo doesn't mean importing its semantic)
>
>[1] http://www.w3.org/TR/2004/WD-wsdl20-20040803/#imports
>
>-- 
>Yves Lafon - W3C
>"Baroula que barouleras, au tiéu toujou t'entourneras."
>

Received on Friday, 17 September 2004 18:42:48 UTC