I personally have no problem with the hybridization of sync/async communication. That's where everything is headed. People send emails and sometimes expect them to be delivered and/or read immediately. OTOH, people send messages over these "chat" services that aren't time sensitive, simply because it's easy and available.
My problem is that all these networks are proprietary and isolated. When someone comes up with a really bad idea in their UI, you're SOL until they decide to fix it. When someone comes up with a really good idea, you have to hope their competitors re-implement it. And all they can do is gently try to guide their network into the place they want to occupy on the sync-async spectrum via features.
Looking back at the history of computing, I see that when a protocol is open and has many competing implementations, it lives on far beyond what the original designers ever intended. When a protocol is proprietary and implemented only by a single proprietary UI, it has a fairly short lifespan. What I can't tell is if these companies don't know their history, or if they think they're going to be the first ones ever to beat this trend, or if they don't care and just want to grab money while they can.