[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index] | [List Home]
Subject: RE: [ws-rx] i119: EPR comparisons overly restrictive
An EPR is more than an address. I trust that someone implementing a comparrison for them would understand that. I think that is where our opinions diverge on this. -----Original Message----- From: "Doug Davis" <dug@us.ibm.com> To: "ws-rx@lists.oasis-open.org" <ws-rx@lists.oasis-open.org> Sent: 5/22/06 5:23 PM Subject: RE: [ws-rx] i119: EPR comparisons overly restrictive "Marc Goodner" <mgoodner@microsoft.com> wrote on 05/22/2006 08:02:20 PM: > Being overly prescriptive seems good, in fact it is limiting in that > some implementations could provide additional value and > differentiate themselves in this area, in either being liberal or > strict. I think Jonathan?s 3.a and 3.b perfectly illustrate this. > Do we really need to get so specific just to remind implementers of > the spec that EPRs can have ref params? I'm confused - by reminding them of ref-p's are you suggesting that they are important or not? If they are, then everyone should take them into account. If not then we should state that too so that both side are aware of how the other will determine where acks will be sent. IMO, this isn't a question of whether someone does it better than someone else its a question of whether its done correctly (in an interop sense) or not. > I?m not understanding your thoughts on the debugging thing. Do you > now see a problem where you didn?t before? Your response directly to > Gil seemed to support the idea that arbitrary headers like this > aren?t harmful in and of themselves. That was w.r.t. how to debug misguided headers - I think that's a different topic. > The scenario you describe below > sounds like a product that shouldn?t have been shipped. Again, I'm lost - if you're saying that an impl that only compares wsa:Address should not have been shipped then I think we need to ask why? They're spec compliant - its not a question of being right or wrong, since if we do as you suggest and close w/no action, they're doing what the spec says - nothing/anything. Its an interop issue. thanks, -Doug
[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index] | [List Home]