W3C home > Mailing lists > Public > public-ws-addressing@w3.org > December 2004

RE: i037: Replace QName's with anyURI

From: Jonathan Marsh <jmarsh@microsoft.com>
Date: Mon, 6 Dec 2004 11:33:10 -0800
Message-ID: <7DA77BF2392448449D094BCEF67569A505D83BE5@RED-MSG-30.redmond.corp.microsoft.com>
To: "Srinivas, Davanum M" <Davanum.Srinivas@ca.com>, "David Orchard" <dorchard@bea.com>, "Hugo Haas" <hugo@w3.org>, <public-ws-addressing@w3.org>

It rates fairly poor in the readability department, but is still IMO far
surpassed in opaqueness by the second example.  Since you chose a QName
for your example that is already pretty unreadable, this makes it clear
to me that even the worst QName gets uglier when mapped into a URI.

> -----Original Message-----
> From: public-ws-addressing-request@w3.org [mailto:public-ws-
> addressing-request@w3.org] On Behalf Of Srinivas, Davanum M
> Sent: Monday, December 06, 2004 11:04 AM
> To: Jonathan Marsh; David Orchard; Hugo Haas; public-ws-
> addressing@w3.org
> Subject: RE: i037: Replace QName's with anyURI
> 
> 
> Jonathan,
> 
> :) I was trying to think aloud what "Readability" means. So the
> example
> is NOT really readable. Right?
> 
> Thanks,
> dims
> 
> -----Original Message-----
> From: Jonathan Marsh [mailto:jmarsh@microsoft.com]
> Sent: Monday, December 06, 2004 1:19 PM
> To: Srinivas, Davanum M; David Orchard; Hugo Haas;
> public-ws-addressing@w3.org
> Subject: RE: i037: Replace QName's with anyURI
> 
> I'm amused that you contrived as a "readability" example a case where
> the localName is longer than the URI and contains more escaping
> characters.
> 
> > -----Original Message-----
> > From: public-ws-addressing-request@w3.org [mailto:public-ws-
> > addressing-request@w3.org] On Behalf Of Srinivas, Davanum M
> > Sent: Friday, December 03, 2004 1:32 PM
> > To: David Orchard; Hugo Haas; public-ws-addressing@w3.org
> > Subject: RE: i037: Replace QName's with anyURI
> >
> >
> > Replace PortType in sample with ServiceName since Issue [1] clearly
> > includes ServiceName as something we need to check if we can
> replace.
> >
> > Look at the EPR as is defined today.
> > <wsa:EndpointReference xmlns:wsa="..."
> > xmlns:fabrikam="http://bosnewslife.com/">
> >     <wsa:Address>http://www.fabrikam123.com/acct</wsa:Address>
> >
> >
> <wsa:ServiceName>fabrikam:BosNewsLife_x0020_Information_x0020_Services
> > </
> > wsa:ServiceName>
> > </wsa:EndpointReference>
> >
> > Is this more readable than what's above?
> > <wsa:EndpointReference xmlns:wsa="...">
> >     <wsa:Address>http://www.fabrikam123.com/acct</wsa:Address>
> >
> >
> <wsa:ServiceName>urn:BosNewsLife_x0020_Information_x0020_Services:http
> > :/
> > /bosnewslife.com/</wsa:ServiceName>
> > </wsa:EndpointReference>
> >
> > [1] http://www.w3.org/2002/ws/addr/wd-issues/#i037
> >
> > -----Original Message-----
> > From: David Orchard [mailto:dorchard@bea.com]
> > Sent: Friday, December 03, 2004 4:18 PM
> > To: Srinivas, Davanum M; Hugo Haas; public-ws-addressing@w3.org
> > Subject: RE: i037: Replace QName's with anyURI
> >
> > It is probably out of scope for WS-A to define a WSDL 1.1 or WSDL
> 2.0
> > portType/Interface to URI mapping.
> >
> > Cheers,
> > Dave
> >
> > > -----Original Message-----
> > > From: Srinivas, Davanum M [mailto:Davanum.Srinivas@ca.com]
> > > Sent: Friday, December 03, 2004 1:11 PM
> > > To: David Orchard; Hugo Haas; public-ws-addressing@w3.org
> > > Subject: RE: i037: Replace QName's with anyURI
> > >
> > > Let's me illustrate better readability with Harris' example [1]
> > >
> > > Look at the EPR as is defined today.
> > > <wsa:EndpointReference xmlns:wsa="..."
> > > xmlns:fabrikam="http://bosnewslife.com/">
> > >     <wsa:Address>http://www.fabrikam123.com/acct</wsa:Address>
> > >
> > >
> >
> <wsa:PortType>fabrikam:BosNewsLife_x0020_Information_x0020_ServicesSoa
> > p<
> > > /wsa:PortType>
> > > </wsa:EndpointReference>
> > >
> > > Is this more readable than what's above?
> > > <wsa:EndpointReference xmlns:wsa="...">
> > >     <wsa:Address>http://www.fabrikam123.com/acct</wsa:Address>
> > >
> > >
> >
> <wsa:PortType>urn:BosNewsLife_x0020_Information_x0020_ServicesSoap:htt
> > p:
> > > //bosnewslife.com/</wsa:PortType>
> > > </wsa:EndpointReference>
> > >
> > > When you look at the PortType, you can see both the port type and
> > the
> > > namespace, it's right there. In the current you need to scan the
> xml
> 
> > > document to look for xmlns:fabrikam and what then figure out what
> > > namespace that it is mapped to....
> > >
> > > [1] http://bosnewslife.com/webservices/bnl_services.asmx?wsdl
> > >
> > > Thanks,
> > > dims
> > >
> > > -----Original Message-----
> > > From: public-ws-addressing-request@w3.org
> > > [mailto:public-ws-addressing-request@w3.org] On Behalf Of David
> > Orchard
> > > Sent: Friday, December 03, 2004 2:58 PM
> > > To: Hugo Haas; public-ws-addressing@w3.org
> > > Subject: RE: i037: Replace QName's with anyURI
> > >
> > >
> > > There's been no proof that the problems that the TAG finding talks
> > about
> > > are relevant to WS-A.  I generally refuse to do things "because
> > somebody
> > > smart said so", and *especially* when the smart people didn't
> outlaw
> 
> > > QNames and noted many cases where they were useful.  I remember
> when
> > I
> >
> > > mentioned to TimBL that OASIS WS-Security had moved to URIs from
> > QNames,
> > > and he sighed and said "there goes readability".
> > >
> > > Dave
> > >
> > > > -----Original Message-----
> > > > From: public-ws-addressing-request@w3.org
> > > [mailto:public-ws-addressing-
> > > > request@w3.org] On Behalf Of Hugo Haas
> > > > Sent: Friday, December 03, 2004 11:35 AM
> > > > To: public-ws-addressing@w3.org
> > > > Subject: Re: i037: Replace QName's with anyURI
> > > >
> > > > * David Orchard <dorchard@bea.com> [2004-12-03 06:56-0800]
> > > > > In general, +1.  It seems to me that any rationale for moving
> > part
> > > of
> > > > > WSA QNames to URIs would be to provide some kind of benefit.
> > I'm
> > > not
> > > > > strongly against moving relationshipType to URIs, but I'd like
> a
> 
> > > > > stronger reason than "because".
> > > >
> > > > I think that the motivation is not just "because", but the TAG
> > finding
> > >
> > > > on QNames that I'm sure you're familiar with:
> > > >
> > > > | In so far as the identification mechanism of the Web is the
> URI
> > and
> > > > | QNames are not URIs, it is a mistake to use a QName for
> > > identification
> > > > | when a URI would serve.
> > > >
> > > > As this is internal to Addressing, it seems like a simple and
> > natural
> > > > change to do.
> > > >
> > > > So I quite like Harris's proposal.
> > > >
> > > > Regards,
> > > >
> > > > Hugo
> > > >
> > > > --
> > > > Hugo Haas - W3C
> > > > mailto:hugo@w3.org - http://www.w3.org/People/Hugo/
> > >
> > >
> >
> >
> >
> 
> 
> 
Received on Monday, 6 December 2004 19:33:25 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Tuesday, 2 June 2009 18:35:00 GMT