Going forward, I have more hope in OSC support in hardware compared to MMC/MTC. If each OSC server defines its own payload format, for example, it's still an improvement as I see it compared to a midi device that has nothing but midi clock to sync on (all my gear).
I don't see that as much of an improvement. I think you're basically saying "nothing will ever change in hardware that doesn't support X, but new products might support OSC instead".
The only hardware recorder I know of that supports MMC/MTC is the Joeco blackbox. The only new hardware sequencer that supports MMC/MTC is the Cirklon - with a 1 year waiting list. Using my Korg Oasys to send midi note data to my Jomox 999 drum machine, for example, requires me to implement my own code with a note_map_t,
I can't imagine why anyone would make a hardware sequencer that responded to MTC. The timebase is completely non-musical. MMC start/stop commands have some sort of semantic sense, but many others still use video-derived time that has no role in a hardware sequencer.
With control voltage for 1 volt per octave pitch and zero to +5VDC for gate on/off, every analog piece of gear I own supports it. Every analog sequencer will accept and send the same gate signal or can be adjusted to do so.
but like OSC, it has no semantics at all, and unlike OSC, "zero syntax". that means that the semantics are defined by the patching, and nothing else. This is not possible for digital signals., because at the very least they all have syntax, and most of them have semantics too.