I am not sure what Keith is referring to but Mesh Messaging is interactive.
Now this is possible because sender and receiver are ALWAYS authenticated to each other and every message is encrypted. But Mesh Messaging is designed for applications of the form 'Doctor sends consent form to patient to sign'.
Each messaging flow (currently) requires a separate message type and specification. But the existing mechanisms support many use cases.
For example, the confirmation interaction (which can be used as an improved 2FA replacement).
Alice and Bob have agreed to some work to be done on Bob's house and Alice wants an approval.
Alice sends Bob a signed confirmation request: "Do you approve work order for $23,233 as per <document>"
Bob sends signed response: Agreed.
The pattern here is to do all the work to set up the contract in a normal HTML forms based environment and then get an approval on the output.
The same scheme could be applied as a 2FA scheme or to provide authorization of admin actions on a system.
For applications like code signing, we would probably want to go further so that there is a cryptographic binding to the process.
This is also potentially a vehicle for applying tasty cryptography like Micali fair contracts.
It is possible we would want to go further with flows such as 'Invoice' and tie those into payment mechanisms. But a large part of the value prop here is that the Mesh communicator is NOT a full featured browser with a grand canyon sized attack surface. So invoices are going to be XML markup, not unrestricted HTML.
[Only in the prototype it is because the prototype is written on top of Maui/Blazor which has Chromium bundled in because we are leveraging WebRTC. But that can be changed if people start using this on a large scale.]
It is also possible in principle to send forms as a Mesh Messaging attachment. This gets to the notion of workflow. It will take a lot of resources to mine those opportunities. But resources follow realization of an opportunity. People can continue to hope crypto-currencies will have a turnaround or they can shift to a technology that is actually relevant to the problems they purported to solve.
There is actually a huge difference between the following workflows
Email <might be spam, might be spurious>
Web
Email <might be spam, might be spurious>
Web
Email <might be spam, might be spurious>
etc/
and
Mesh message Authenticated
Mesh message Authenticated
Mesh message Authenticated
