It's a step in the right direction but it's the same problem as iMessage supporting SMS in addition to its own protocol. If the proprietary protocol supports something the open protocol doesn't, or that Apple doesn't implement for the open protocol, a competing implementation can't do it. And if it doesn't do that then why does the proprietary protocol exist?
If you're going to make your own protocol, publish a spec.
It's a service and a protocol and a client application.
But it should be possible for a third party to make an interoperating client application using the same protocol, and then it could connect to any service implementing that protocol.
Obviously if you want to be selective in who can use your service then you should create a protocol that supports some kind of federation so that each user's service can forward messages to another service if the two endpoints don't use the same one. Email works like this, for example.
> Should I be able to operate a business that sends bulk advertisements to iMessage users?
Can you do this now by using a Mac with iMessage?
The point of the blue bubble is to ensure the encryption is there.
Kicking them out of the system makes your messages less secure, not more secure.
The hard part is associating some identity with the user's keys, but when the ID is your phone number or email address, the entity doing that is inherently your phone company or email provider. You can standardize a way to do that, i.e. to sign up you get an SMS or email with a code and have to enter the code. The client can automate that if it has access to read your SMS or email, or otherwise you enter it manually.
But also when it comes to managing the keys and syncing across devices it’s also the servers.
Someone willing to do the work should be capable of producing an interoperable implementation.