My own view on this is that when the wire protocol changes, the handling in the new code should recognize and process the incompatibility. The crashing code in question can recognize the old wire protocol and process it. This is a burden on the future, and also incentive to stop changing the wire protocol.
I believe this backward compatibility is part of becoming a responsible, mature standard. Of course, this only applies to official releases, not 0.9. But from here we should take responsibility for our mistakes. John Light Intel OTC OIC development -----Original Message----- From: iotivity-dev-bounces at lists.iotivity.org [mailto:[email protected]] On Behalf Of Thiago Macieira Sent: Wednesday, March 16, 2016 9:17 AM To: Iovene, Salvatore <salvatore.iovene at intel.com> Cc: iotivity-dev <iotivity-dev at lists.iotivity.org> Subject: Re: [dev] Another crash when trying to get the representation of a resource (from Java) On quarta-feira, 16 de mar?o de 2016 11:40:48 PDT Iovene, Salvatore wrote: > This might be the case. Server is 1.0, client is 1.0.1. A colleague > hinted the same thing, tho I'm surprised that something versioned 1.0 > would be incompatible with something version 1.0.1. > > Hope others don't find the same issue! There is indeed a wire incompatibility. But that's no excuse for crashing. If you can still reproduce the crash with 1.1, we could use a detailed bug report (with debug symbols). -- Thiago Macieira - thiago.macieira (AT) intel.com Software Architect - Intel Open Source Technology Center _______________________________________________ iotivity-dev mailing list iotivity-dev at lists.iotivity.org https://lists.iotivity.org/mailman/listinfo/iotivity-dev
