To:
"'ietf-provreg@cafax.se'" <ietf-provreg@cafax.se>
From:
Rick Wesson <wessorh@ar.com>
Date:
Tue, 30 Mar 2004 11:26:23 -0800 (PST)
Sender:
owner-ietf-provreg@cafax.se
Subject:
[ietf-provreg] schemaLocation required?
in RFC3730 it states... 2.2. Protocol Identification All EPP XML instances MUST begin with an <epp> element. This element identifies the start of an EPP protocol element, the namespace used within the protocol, and the location of the protocol schema. The <epp> start element and the associated </epp> ending element MUST be applied to all structures sent by both clients and servers. Example "start" and "end" EPP elements: <epp xmlns="urn:ietf:params:xml:ns:epp-1.0" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" xsi:schemaLocation="urn:ietf:params:xml:ns:epp-1.0 epp-1.0.xsd"> </epp> My question is: is the xsi:schemaLocation attribute a MUST ot a MAY? the above language isn't expressly clear about it. Since the xmlns clearly defines what we are talking about and the xmlns:xsi and xsi:schemaLocation redundancy remove all the chars recovered in the shortning of element names. These elements cost 119 chars which is more than we recovered in the element name shorning scheme. IMHO we should either reduce the requirement (if there is an implied MUST) for xmlns:xsi and xsi:schemaLocation or use real terms for the element names insted of shorting them to acronyms. could scott clarifiy his intention with xmlns:xsi and xsi:schemaLocation? thanks, -rick