Re: Test assertion markups for SOAP/JMS spec completed

Greetings, an interesting subect. eric is correct in pointing out that the 
work on a JMS Scheme in IETF has become a URI. And it is also the case 
that conversion from an IRI to a URI is possible. What I think we will 
need to do is decide where and when such an transformation takes place.

Regards, Roland
FBCS, CITP




Eric Johnson <eric@tibco.com> 
Sent by: public-soap-jms-request@w3.org
18/06/2008 18:00

To
Yves Lafon <ylafon@w3.org>
cc
"SOAP/JMS (list)" <public-soap-jms@w3.org>
Subject
Re: Test assertion markups for SOAP/JMS spec completed







Hi Yves,

Yves Lafon wrote:
> On Tue, 17 Jun 2008, Eric Johnson wrote:
>
>> Given Peter's complete run through, it appears that he noted several
>> editoral/content issues not related to testing.  Seems to me that we
>> should turn that into a list for us to resolve the issues, and or have
>> the editors apply changes.  So I thought it useful to put together the
>> following:
>>
>> Editorial changes:
>>  *  Document refers to IRI scheme instead of URI throughout.  Should be
>>     changed to URI. Sections 1.1, 1.5, 2.2, 2.2.1, 2.2.2, 2.2.3, 2.2.4,
>>     2.2.5, 2.3, 3.2, 3.3, 3.4.1, 3.4.5, 3.5,
>>  *  Section 1.1: Third bullet currently reads "The IRI specification 
for
>>     ..." -- suggestion to change to "How the SOAP over JMS uses the 
[URI
>>     Scheme for JMS]."
>
> Are you sure you want to restrict to URIs only? In SOAP 1.2 [1], IRIs 
> are allowed (provided they are encoded in form of URIs). What forbid 
> using IRIs at a higher level, then encode in URI when needed ?
I think we're in complete agreement here.  The folks over at IETF 
corrected our original use of IRI instead of URI for the "jms" scheme. 
They pointed out that the IETF only supports registration of URIs, not 
IRIs.  So we renamed that specification to be about the "jms" URI, not 
the "jms" IRI.  You can, of course, use an IRI representation of the 
equivalent URI wherever it might be supported.  The above two points are 
specifically about the editorial change to properly refer to the JMS URI 
scheme.

One question for me is that we certainly *can* allow the use of IRIs, 
but we perhaps need to decide whether we generally want to use IRIs 
everywhere we can.  The JMS message API of course uses Java Strings, and 
thus is not subject to byte encoding issues, so an IRI can be carried 
without problems by JMS.

Pouring over section 3.2.17 of XML Schema datatypes 
(http://www.w3.org/TR/xmlschema-2/#anyURI), I find it very difficult to 
suss out whether, in WSDL, I can put the IRI form of a URI wherever WSDL 
expects xsd:anyURI.  It doesn't seem like it.

-Eric.









Unless stated otherwise above:
IBM United Kingdom Limited - Registered in England and Wales with number 
741598. 
Registered office: PO Box 41, North Harbour, Portsmouth, Hampshire PO6 3AU