Going through the 2.0 info I was disapointed in one area. Midi 1.0 was based on piano type instruments. So it seemed logical to use the keys in the interface definition. These day playing surfaces and instruments have evolved. Note on value should in my opinion no longer be defined by key but should specify the frequency. If the Key is kept as basis you still have to fiddle arround with pitch bend to get tuning variations you seek. Integration into eurorack, modular, microtunal, different scales, application to other instruments and varity in tuning would profit from this. So the question is why?
I would wait for the full MIDI 2.0 spec to be released before making assumptions about what is in it. Backwards compatibility is of paramount importance, but MIDI 2.0 adds so much more that it might actually address your concerns. We'll have to wait a little longer and see. 🙂
Hi --
The 2.0 spec does address this problem exactly, potentially in several different ways. Note On/Off messages add a new 16-bit attribute value that can be used for several purposes, including the specification of the pitch to a precision of 1/512th of a semitone (9 bits of fractional semitone precision). There's a more complicated multi-message approach that permits specifying pitch with 25 bits of semitone data (1/8191 of a semitone)