- From: <paul.downey@bt.com>
- Date: Mon, 26 Jul 2004 21:23:44 +0100
- To: <sanjiva@watson.ibm.com>, <www-ws-desc@w3.org>
FWIW i cast myself as being "a user" and i find it OK, no less subtle or obscure than XPath or W3C schema patterns .. -----Original Message----- From: www-ws-desc-request@w3.org on behalf of Sanjiva Weerawarana Sent: Mon 26/07/2004 21:02 To: www-ws-desc@w3.org Cc: Subject: Re: Issue 189 proposals: "Hugo Haas" <hugo@w3.org> writes: > > If I have: > > <data> > <a>1</a> > <b>2</b> > <c>3</c> > <d>4</d> > </data> > > then the draft as it stands says that: > - location="{a}/{b}" serializes as "1/2?c=3&d=4" > - location="{a}/{b}/" serializes as "1/2/?c=3&d=4" > > Now, your proposal is that location="{a}/{b/}" serializes as "1/2", > and that the rest be serialized as application/xml, i.e.: > > <data> > <c>3</c> > <d>4</d> > </data> > > Am I right? I have to say .. this is just too much black art for me. How do you expect users to distinguish between "{a}/{b/}" and "{a}/{b}/" with such serious implications? I know we accepted this, but its not a winner feature IMO. Sanjiva.
Received on Monday, 26 July 2004 16:24:03 UTC