Home > Blog > #PKMChat 2016-06-15 Bots, Chatbots and Integrations

Bots, Chatbots and Integrations

A few definitions taken directly from Wikipedia A bot (derived from the word “robot”) is an automated or semi-automated tool that carries out repetitive and mundane tasks … source: https://en.wikipedia.org/wiki/Wikipedia:Bots So bots relates to automation, the A of the four direction of producity from Tim Ferris. A PKM Bot would be an agent specifically made to us accomplish to tasks related to dealing with incoming information, classifying it, searching it etc.. A chatbot (also known as a talkbot, chatterbot, chatbot, Bot, chatterbox, Artificial Conversational Entity) is a computer program which conducts a conversation via auditory or textual methods. Such programs are often designed to convincingly simulate how a human would behave as a conversational partner … . Chatterbots are typically used in dialog systems for various practical purposes including customer service or information acquisition. Some chatterbots use sophisticated natural language processing systems, but many simpler systems scan for keywords within the input, then pull a reply with the most matching keywords, or the most similar wording pattern, from a database.( source: https://en.wikipedia.org/wiki/Chatterbot) A chatbot defines a precise interface with a bot: Natural language, not more, not less. It could vocal or typed, still it’s language. In computing integrations Architecture refers to a pattern of bringing additional services and capabilities to a central service by adjoining interfaces to other services. For WordPress it would be plugins or Themes, for Slack it is integrations, for Facebook it’s apps etc.. The concept is always the same: A host service defines an API , the rules of the game. If offers attractive capabilities, a welcoming infrastructure and exposure. Most chatbots are pretty complex to setup and attach to a dominant platform: Siri, Cortona. However once the decoding of the request is done, the actual execution of it depends on the existence of a multitude of microservices offered by integrations. The more integrations, the more the chatbot can actually help. So using the combo of bot, chatbot and integrations we end up having automated tasks running on hubs, commanded by language doing repeated tasks as requested and leveraging integrations for this purpose. It’s the pattern we can expect in most cases. Example in Slack: http://www.inc.com/kevin-j-ryan/why-slack-is-about-to-become-your-boss.html A chatbot establishes a conversation, not just simple queries without no follow ups. So “search a youtube video on collective learning” don’t require a chatbot but just a natural language interface, while “search a video on collective learning from the same guy you found a video last week on social media” “this one? ____” “yes, what else” “can’t find from same author but a similar one, shall I show it?” “yes, download it and remind me later to take a look, if its good continue with this author as well, otherwise search others.” This requires a real chatbot. There is a context maintained from line to line but also a history of requests and interactions being kept, a foreseeable actions plan. The context of the user (“later”) is also taken into account. Once given those elements, I invite you to devise during this #PKMChat to imagine which services we would love to have to help us being more productive in PKM, to maintaining our knowledge and for our Lifelong learning. Join and share your wildest desire of being obeyed for your routines knowledge work. I felt compelled to do this topic since so much rumors are shared and the software engineer in me felt #PKMChat could be the place we debunk some of them. After all theses technology (plus AI and 3D/VR) have been my everyday work for years, I can claim I’m qualified. PKMChat being about Personal Knowledge Management encompass Knowledge lifecycle in general. Our first chat was about learning, acquiring Knowledge. Our second is about sharing it. Weeks after weeks we will switch from one end of the lifecycle to another while exploring all the channels that could be used: social, formal, writing, videos. Feel free to suggest topics by tweeting to @pkmchat.
TBC

Links:

Links on Diigo #PKMChat group with TBC tag https://groups.diigo.com/group/pkmchat/content/tag/TBC TBC

Questions:

  • Q1: TBC
  • Q2: TBC
  • Q3: TBC
  • Q4: TBC
  • Q5: TBC
  • NB: Questions are subject to change without notice.