Re: Rework on SOAP 1.2 Part 2: Section 2 and 3

----- Original Message -----
From: "Jacek Kopecky" <jacek@systinet.com>
To: "Williams, Stuart" <skw@hplb.hpl.hp.com>
Cc: "'Jean-Jacques Moreau'" <moreau@crf.canon.fr>; "Martin Gudgin"
<marting@develop.com>; "Noah Mendelsohn" <noah_mendelsohn@us.ibm.com>;
<xml-dist-app@w3.org>
Sent: Friday, March 15, 2002 4:34 PM
Subject: RE: Rework on SOAP 1.2 Part 2: Section 2 and 3


> Stuart,
>  as Herve put it, the serialization does not necessarily contain
> the names structA and structB. The misunderstanding might be
> caused by the fact that we don't yet know for sure what we start
> with - either a root node or a root edge (starting nowhere).
>  Root node: the name of the appropriate element information item
> is not ours (Encoding's) to decide.

I think we start here. Algthough that may cause problems in the RPC case as
we don't know the name of the top level element. Off the top of my head I
would suggest that we add some prose to Section 4 stating that the name of
the top level element in the RPC case is the name of the method. In non-RPC
cases, people can use whatever method names they like.

>  Root edge: the name of the appropriate element is set by the
> label of this edge, the edge MUST be labeled.

I don't think this makes much sense, having an *inbound* edge with no origin
just seems weird to me.

>  Either way, we probably should say it explicitly.

Agreed

Gudge

Received on Thursday, 21 March 2002 09:32:40 UTC