- From: Don Box <dbox@microsoft.com>
- Date: Wed, 6 Feb 2002 17:11:20 -0800
- To: "Jacek Kopecky" <jacek@systinet.com>, "Tim Ewald" <tjewald@develop.com>
- Cc: "XMLDISTAPP" <xml-dist-app@w3.org>
> -----Original Message----- > From: Jacek Kopecky [mailto:jacek@systinet.com] > Sent: Wednesday, February 06, 2002 8:13 AM > To: Tim Ewald > Cc: XMLDISTAPP > Subject: Re: Issue with soap-rpc:result > > Tim, > the intention is that the return value accessor is named > soap-rpc:result _and_ strongly typed. If we remove the namespace > qualification, we have to solve possible conflicts in this name and > the names of the other accessors. In fact, your problem comes from > using XML Schema to describe messages created with SOAP Encoding. This > is one of the cases where XML Schema is not the perfect language for > the job and where a different schema language would be better, > something like SOAP Data Model Schema language. I don't think adding yet another way to define data representations is the answer. No one thinks XML Schema is perfect, yet it is the core type system that SOAP relies on. XSD got increasingly baked into the wire as SOAP progressed, and I think we have gone too far to retract that (not that I would advocate that anyway). If someone wants to pitch yet another schema-like language, it should be taken up by the WSDL WG, not SOAP WG. To that end, I think that core aspects of the protocol (e.g., Envelope/Header/Body, and fault) need to rely on XML Schema and nothing else. The reality is, SOAP mandates XML Schema-awareness. WSDL just makes explicit what we made implicit in SOAP/1.0 and SOAP/1.1. DB
Received on Wednesday, 6 February 2002 20:12:12 UTC