- From: Jean-Jacques Moreau <moreau@crf.canon.fr>
- Date: Thu, 31 Jan 2002 09:51:24 +0100
- To: "Williams Stuart" <skw@hplb.hpl.hp.com>
- CC: "'Jacek Kopecky'" <jacek@systinet.com>, xml-dist-app@w3.org
I think this whole discussion is pointing out an inconsistency
between section 4.4.3 "SOAP faultactor Element" and section 2
(with or without the ed's proposed rewrite).
Specifically, section 2 introduces the notion of roles, which
nodes play. IMO, this section places a greater emphasis on roles
than on nodes. In contrast, section 4.4.3 is mute about roles
(see quote below), which I find quite disturbing since a node is
allowed to play multiple roles. I would be tempted to say that
the faultactor attribute really ought to identify not just the
node that faulted (coarse-grained), but the exact role in which
that node operated (fine-grained).
"faultactor [...] is intended to provide information
about which SOAP node on the SOAP message path caused
the fault to happen [...]. It is similar to [...] SOAP
actor [...] but instead of indicating the target of a
SOAP header block, it indicates the source of the
fault. The value of [...] faultactor [...] identifies
the source of the fault.
Jean-Jacques.
PS. You guys never seem to go to bed.
Received on Thursday, 31 January 2002 03:52:59 UTC