- From: Marc Hadley <marc.hadley@sun.com>
- Date: Fri, 22 Nov 2002 08:43:22 -0500
- To: "Henrik Frystyk Nielsen" <henrikn@microsoft.com>
- Cc: <noah_mendelsohn@us.ibm.com>, "Martin Gudgin" <mgudgin@microsoft.com>, <xml-dist-app@w3.org>
+1 on SHOULD rather than MUST, it shouldn't affect interop IMO as the definitive part of the fault is the code/subcode value. Marc. On Friday, Nov 22, 2002, at 00:26 US/Eastern, Henrik Frystyk Nielsen wrote: > > Yeah, I am not sure either. If a node finds two Russian reason texts > should it then generate a env:Sender fault? That would seem like > overkill to me, especially as I am not convinced that this is really an > interoperability issue. In other words, I don't think this can be any > stronger than a SHOULD. > > Henrik > >> I'm on the fence whether this should be MUST or SHOULD. >> >> Pros for MUST: it's what we mean, having two with the same lang is >> incoherent, it's consistent with enforcing in the schema via Unique, I >> think. >> >> Cons for MUST: turns it into a conformance requirement to >> check. If I >> find an English reason, must I really check that there weren't >> also two >> strings in Russian? That's a lot of implementation effort for a >> non-validating implementation. >> >> As I say, I'm unsure what's best. > > -- Marc Hadley <marc.hadley@sun.com> Web Technologies and Standards, Sun Microsystems.
Received on Friday, 22 November 2002 08:44:04 UTC