This is preparation for a module at Roehampton which will probably be in spring 2018. As referenced previously Conversational UI project and the case study. These both show the steps in how a conversation evolves with a chatbot by looking at the communication rules, building the conversation frame, use of syntax and messaging formats. They came up with conversational UI definition
An interface based on a holistic system of functional, adaptive and meaningful messages exchange, in which both sides of the conversation use and interpret the language codes, maintaining and complying with the constitutive and normative rules in a friendly, informal way 1
An interesting overview of the programming elements in a conversational UI explains the developments of chatbots from performing tasks to more advanced forms of conversation using AI processing and scripts built using Chatskills (based around Amazon Alexa which could retrieve photos as well as respond to queries). This is designed for using applications on computer or phone rather than e.g. a chatbot service for ordering taxis etc
There is a breakdown of a query into namespaces which mean that words for the chat or content are separated from the same word for processing or script e.g. XML namespaces example.
In the article, it describes namespaces
Namespaces allow us to separate logic and tasks, which might re-use similar conversational text keywords. By namespacing services, we can allow other services to use the same keywords, while still identifying which service we’re currently working in. A bot might have many different namespaces. In addition to the epix service, perhaps it may also have a photos service, a users service, and an orders service. 2
This could be applied to different contexts where a service could be users (students) but another alternative service to – show a photo of…, for our students, it could be asking it for an explanation of a concept based on pre-built libraries that the conversational bot can search and retrieve an answer. If there is interaction with students, another Kory Becker primary objects article looks at emotion and AI using sentiment analysis. This has got a lot more sophisticated since I explored it a little in 2008 when looking at media and the financial crisis. The code and any computers & applications involved in the processing isn’t actually showing empathy, it’s retrieving words that have been described as a suitable human response. Shrinkwrapped has a psychoanalytic look at empathy and AI and what different fields are doing to try and ensure that future computers and their processing are friendly in their interactions with humans and this universe.
I have always been and remain opposed to covert & non-consented experimentation of computing and human interaction but students will be creating future worlds not me, so who knows.
1. 2016, Zawadski L, Conversational UI Principles — Complete Process of Designing a Website Chatbot, available at Medium: https://medium.com/swlh/conversational-ui-principles-complete-process-of-designing-a-website-chatbot-d0c2a5fee376
2. 2016, Becker K, An Overview of Conversational UI, available at Primary Objects: http://www.primaryobjects.com/2016/04/12/an-overview-of-conversational-ui/