WORK IN PROGRESS
IOTICS Twins' inputs can be used to model any async request/response.
-
A requestor twin specifies a "response input"
-
A responder twin specifies a "request input"; the payload for the request command is specified semantically as a set of key/value/unit triples. The payload must allow for specifying the link to the "response input" and a response reference.
-
The requestor sends the request to the "request input"
-
The responder sends the response to the "response input"
Multiple responses can be sent to the "response input" each with their own sequence number
In case the response has a stream semantic (hence the request command is a subscription), the responder must allow a request command for unsubscribe.