[userbinator]

It's odd that, despite there being an open protocol for realtime chat that is over 25 years old, older than HTTP or HTML, I would guess that the majority of the population hasn't heard of or used it. On the other hand, almost everyone has heard of and uses email, which also dates from around the same time period.

> A unified communication service would not only allow me to communicate easily on any device it makes sense to, it would also unify the 3 main communication formats into one platform: voice, video, and messaging.

The closest to that seems to be SIP and its related protocols, but they don't seem all that popular for some reason.

[pjc50]

Which protocol are you referring to - IRC? talkd?

Voice and video are fundamentally different from messaging because they are connection-orientated; I think it's a mistake to conflate the two.

SIP is actually pretty good when the firewall punching works. I suspect there are enough cases when it doesn't for that to be a problem. And of course, if it's hard to monetise then it's hard to advertise, hence the proprietary push.

[userbinator]

IRC.

I see how messaging doesn't need to be connection-oriented, but in many cases it is (especially if you want near-realtime latency, and not "whenever the next poll is".)

[anon4]

I'd say because they're hard (or at least tricky and not trivial) to implement and rely on more infrastructure than a lean startup wants to maintain and once you do, you again have a separate service that doesn't interoperate with anybody, but now your technology stack is an order of magnitude larger than the competition's and therefore more prone to breakage.