Email is a bootstrap that no "next-generation" communications solution can do without. Every startup, whether focused on communications or not, wants your email address. That is not a coincidence.
From a marketing perspective, email also has the greatest reach, greater than the web. The majority of the still relatively small percentage of the world's pop. that is connected to the internet are more likely to use email than the web, and they spend more time using email than they do the web.
If you do the research I believe you will find that these are the facts.
Email needs to be improved (e.g. cut out the middlemen -- "email providers" -- somehow, perhaps run email servers within trusted peer-to-peer groups: groups that are established using only the internet and without the need for a web browser, javascript, etc.). Email does not need to be "replaced".
Email is a simple and effective concept and as such, it is not going to "go away". So let's work on trying to improve it, architecturally.
The "walled garden" approach to communications, maybe using different gardens for different types of contacts to whom you provide your mail address, makes sense. Maybe it wouldn't stop spam, but it might contain it, e.g. to certain gardens. But any solution dubbed an "email killer" that based on the public web is not truly walled. It relies on a complex web browser. There will be ways to get in. And it's the same old story: people want to get in to spam you or scan your email for hints they can give to advertisers.
Imagine if the post office scanned every letter for hints they could give to advertisers, like the ones who fill your physical mailbox with junk postal mail. Why do we accept this with email?
Improvements on email need to start at a lower level: the IP layer. Peer-to-peer. Move away from "calf-cow". Take back control of our mail.