OASIS Mailing List ArchivesView the OASIS mailing list archive below
or browse/search using MarkMail.

 


Help: OASIS Mailing Lists Help | MarkMail Help

ws-rx message

[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index] | [List Home]


Subject: RE: [ws-rx] i021 proposal


 

> -----Original Message-----
> From: Paul Fremantle [mailto:paul@wso2.com] 
> Sent: Friday, Feb 10, 2006 6:01 AM
> To: Doug Davis
> Cc: ws-rx@lists.oasis-open.org
> Subject: Re: [ws-rx] i021 proposal
> 
> Doug
> 
> I need someone to explain to me how the other specs work, as I'm not 
> familiar with them.
> 
> For example when there is a security policy for web service, is the 
> default to assume that all response messages are unsecured?

Hang on though. If the attachment is on an endpoint, all the messages
that are captured by the endpoint will be subject to the policy. If the
attachment is on the message level, only a specific message in a
particular direction (input/output/fault) will be affected. 

The current proposal allows either of these policy subjects, therefore
it is possible to express RM engagement at the endpoint level, or
message level, etc. 

The semantics depend on the specific assertion scope and since we are
defining it, we can define it suitable for our needs. 

> 
> Paul

--umit

> 
> Doug Davis wrote:
> >
> > Paul,
> >   I'm just looking at the other ws-* specs w.r.t. policy and they 
> > don't seem to make the
> > kind of distinctions we're thinking about making.  Either 
> 'foo' is on 
> > or off - that's about
> > all they say - and its very simple.  If we deviate from 
> that pattern 
> > we run the risk of
> > RM being a special case.  If we believe those other specs are wrong 
> > then we should
> > push the policy experts to agree to that and then we can be 
> the first 
> > to "do it right", but
> > until then I'm not keen on RM being the oddball.
> > -Doug
> >
> >
> >
> > *Paul Fremantle <paul@wso2.com>*
> >
> > 02/10/2006 08:45 AM
> >
> > 	
> > To
> > 	Doug Davis/Raleigh/IBM@IBMUS
> > cc
> > 	ws-rx@lists.oasis-open.org
> > Subject
> > 	Re: [ws-rx] i021 proposal
> >
> >
> >
> > 	
> >
> >
> >
> >
> >
> > Doug, Umit,
> >
> > I think there is a simplicity to this. What you are sating 
> is that there
> > is no way of expressing in a req-resp WSDL the fact that 
> both directions
> > are expected to be reliable.
> > That may be a logical conclusion but its not clear that it fits the
> > expected usage of RM in a req-resp scenario.
> >
> > Paul
> >
> > Doug Davis wrote:
> > >
> > > "Yalcinalp, Umit" <umit.yalcinalp@sap.com> wrote on 02/09/2006
> > > 09:43:16 PM:
> > > ...
> > > > It seems to me if it would be cleaner to leave the 
> one-way policy
> > > > assertion at the input message only, so that for the "outbound 
> > messages"
> > > > the receiving end's policy assertion would apply. I am 
> thinking in 
> > terms
> > > > reconciling the policies of RMS and RMD at both ends (including
> > > > extensibility). I think binding/operation/input message 
> should be
> > > > sufficient and is simpler.
> > >
> > > +1 - keep it simple (and consistent with the other ws-* specs)
> > >
> > > -Doug
> >
> > -- 
> >
> > Paul Fremantle
> > VP/Technology, WSO2 and OASIS WS-RX TC Co-chair
> >
> > http://bloglines.com/blog/paulfremantle
> > paul@wso2.com
> >
> > "Oxygenating the Web Service Platform", www.wso2.com
> >
> >
> 
> -- 
> 
> Paul Fremantle
> VP/Technology, WSO2 and OASIS WS-RX TC Co-chair
> 
> http://bloglines.com/blog/paulfremantle
> paul@wso2.com
> 
> "Oxygenating the Web Service Platform", www.wso2.com
> 
> 


[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index] | [List Home]