Thanks Paul for sharing your thoughts. Yes, I agree that Confidence is a Key here and Testing is all about making things Certain :-) .
Regards Arun On Tue, Jan 8, 2013 at 11:10 PM, Paul Kyzivat <[email protected]> wrote: > On 1/8/13 10:51 AM, Arun kalia wrote: > >> There is a test suite which has 100% coverage of the spec, if it is run >> against the product and it shows 100% PASS. Then yes, of-course the >> product is 100% compliant. >> > > In that case it is up to the publishers of the test suite to say if they > have sufficient confidence to guarantee that two implementations adhering > to it will interoperate. > > At best I would think this would be phrased as confidence levels. There is > no certainty. Even a direct interop test is no guarantee that every use in > practice will work. > > Thanks, > Paul > > Thanks & regards >> Arun >> >> On Tue, Jan 8, 2013 at 8:01 PM, Paul Kyzivat <[email protected] >> <mailto:[email protected]>**> wrote: >> >> On 1/8/13 1:54 AM, Arun kalia wrote: >> > Hi All, >> > >> > If two products are 100% compliant to any specification, for e.g; >> rfc 3261, >> > then is the Interoperability testing required for these two >> products? >> >> How would you know that a product is "100% compliant" to a spec? >> >> Thanks, >> Paul >> >> ______________________________**_________________ >> Sip-implementors mailing list >> >> [email protected].**columbia.edu<[email protected]> >> >> <mailto:Sip-implementors@**lists.cs.columbia.edu<[email protected]> >> > >> >> https://lists.cs.columbia.edu/**cucslists/listinfo/sip-**implementors<https://lists.cs.columbia.edu/cucslists/listinfo/sip-implementors> >> >> >> > _______________________________________________ Sip-implementors mailing list [email protected] https://lists.cs.columbia.edu/cucslists/listinfo/sip-implementors
