The Conversation service allows to define intents, entities, dialog. This provides the framework to analyze a conversation and trigger actions when needed - or call an external service to get more information.
When it comes to managing the conversation and these actions, a developer would typically creates a small app to receive the user input, pass it to Conversation, look at the detected intents, potentially call external services, reply to the user. We see such loop in https://github.com/watson-developer-cloud/conversation-simple/blob/master/app.js#L41 (nodejs) or here https://github.com/watson-developer-cloud/conversation-enhanced/blob/b9508152fc914df7f0818af1e2538c5b570767e0/src/main/java/com/ibm/watson/apis/conversation_enhanced/rest/ProxyResource.java#L107 (java).
If I want to write a simple bot that answers to "what is the weather in Paris?" I would need to write such a small app to handle the conversation loop and the external call to the Weather service.
But instead, what if Conversation could handle this and allow me to declare actions that could be implemented with something like OpenWhisk? That's the ask of this requirement. The ability to define actions in Conversation and to link these actions to OpenWhisk actions. Ideally this should allow me to build a fully serverless conversation bot, with Watson Conversation driving the conversation and not a small app I would have to write.
NOTICE TO EU RESIDENTS: per EU Data Protection Policy, if you wish to remove your personal information from the IBM ideas portal, please login to the ideas portal using your previously registered information then change your email to "firstname.lastname@example.org" and first name to "anonymous" and last name to "anonymous". This will ensure that IBM will not send any emails to you about all idea submissions