Conversational Agents (aka chatbots and intelligent assistants) are very popular for customer service, whether text based agents embedded in web pages, or speech-based. There are a range of conversational agents markup languages, and AIML is one of the best known. Others include BOTML, SIML, and CML. Is it timely for W3C to organize a workshop on conversational agents to see if there is interest in standardizing a chatbot markup language and scripting API given the commercial popularity of chatbots? The workshop could include proposals for extending SSML for conversational agents (see Amazon's extensions), the possibility of standard logging formats as a basis for machine learning tools, and the role of intents as a means to support ecosystems of services, where third parties register an intent for a particular service, or just for ease of maintenance through modularising conversational agents into conversations covering particular intents.
Intelligent assistants will benefit from access to personal information that enables a positive customer service interaction. This points to new opportunities for managing privacy in innovative ways.
Conversational Agents (aka chatbots and intelligent assistants) are very popular for customer service, whether text based agents embedded in web pages, or speech-based. There are a range of conversational agents markup languages, and AIML is one of the best known. Others include BOTML, SIML, and CML. Is it timely for W3C to organize a workshop on conversational agents to see if there is interest in standardizing a chatbot markup language and scripting API given the commercial popularity of chatbots? The workshop could include proposals for extending SSML for conversational agents (see Amazon's extensions), the possibility of standard logging formats as a basis for machine learning tools, and the role of intents as a means to support ecosystems of services, where third parties register an intent for a particular service, or just for ease of maintenance through modularising conversational agents into conversations covering particular intents.
Intelligent assistants will benefit from access to personal information that enables a positive customer service interaction. This points to new opportunities for managing privacy in innovative ways.