The IoT Conversation
0 likes577 views
1) The document discusses three technologies - financial trading, massively multiplayer online gaming, and conversational social media - that can provide lessons for building Internet of Things (IoT) systems. 2) For financial trading, the value of time-sensitive data declines rapidly so IoT systems need low-latency processing of real-time data, like Complex Event Processing (CEP). 3) Online games involve asynchronous, multi-player interactions and conversations between users and systems, which are important concepts for designing human-machine interfaces for IoT. 4) Early social media like Twitter emerged from simple status updates but gained rich conversational features through user contributions, showing how IoT systems could also evolve through collaboration.
The IoT Conversation
- 1. Matt Biddulph Thington Inc @mattb Continuing the IoT conversation
- 2. âIt is a magniïŹcent feeling to recognize the unity of complex phenomena which appear to be things quite apart from the direct visible truth.â Albert Einstein Today Iâm going to talk about three areas of technology that Internet of Things systems can learn from. On the surface, none of them are IoT technologies.
- 3. One of my favourite concepts is the idea of âconsilenceâ, described by Edward O. Wilson. He believes that knowledge can be uniïŹed across scientiïŹc disciplines. Itâs the idea of âthe fundamental interconnectedness of all thingsâ as Douglas Adams put it.
- 4. Tech number 1: time-oriented data in ïŹnancial trading
- 5. In ïŹnancial trading, the most valuable data is the freshest data. The value of much ïŹnancial data declines over time. The same is true of IoT data when we use it to monitor the real world and trigger actions for users.
- 6. As Kevin Slavin described in his brilliant talk âHow algorithms shape our worldâ there are buildings in Manhattan full of servers positioned as close as possible to the London transatlantic data cables. There is almost nothing that humans could do in that real estate that would be more valuable than the lowest-possible latency connection that those servers use to trade on.
- 7. The need for low-latency correlation, query and response on realtime data is generally known as CEP: Complex event processing. One such open-source system that we used to build Thington is called Esper.
- 8. The Devops community has been using CEP tools for its own real-time time-series analysis and alerting for a long time.
- 9. âThe problems we look at have temporal constraints ranging from: 5 seconds (counters and statistics) toâš 1 second (fraud detection) toâš 10 milliseconds (user-action reaction) âš andâš everywhere in between.â Theo Schlossnagle, OmniTI Theo Schnossnagle has described using Esper for its in-memory time-oriented constructs in order to deal with a wide variety of data on behalf of clients at OmniTI.
- 10. Change over Time The essence of real-world data is that it describes change over time. This is a fundamental concept for IoT architectures.
- 11. Tech number 2: massively-multiplayer online gaming
- 12. Some of the most complex realtime systems in the world are group events such as raiding in World Of Warcraft.
- 13. The Internet of Things can turn single-player things such as a car with a single owner and set of keysâŠ
- 14. ` ⊠into multiplayer objects by adding a small amount of hardware and a service layer such as Zipcar
- 15. Games are essentially asynchronous, essentially multi-character / multi-player, and full of conversations.
- 16. Early games such as the Monkey Island are brilliant solutions to giving the player a sense of open conversations in the absence of strong AI and processing power
- 17. These techniques are still used in modern games such as 80 Days by Inkle Studios
- 18. They open-sourced the narrative engine that their designers (not just their developers!) use to create in-game conversation. We have adapted this open-sourced engine to create a conversational interface in Thington.
- 19. Human2Machine The essence of games is the human-machine interface. This is a fundamental concept for IoT architectures.
- 20. Tech number 3: conversational social media
- 21. This is a sketch by Jack Dorsey of an early idea for status updates that became Twitter
- 22. âThe emergence of @ to mean a reply was a crucial development in Twitterâs history.â âThe ïŹrst-ever hashtag, @-reply and retweetâ â qz.com âEarly on, its founders struggled over whether the service was primarily for status updates or conversation.â From a simple system of status updates, a number of emergent features were co-created by Twitterâs userbase that formed the axioms of a rich human ecosystem of conversation.
- 23. One example of this is when Twitter technical architect Blaine instituted the rules of @reply scoping, meaning that when one user made an @reply to another, only users that followed both users would see the reply. This helped users managed the ïŹrehose of content, and is a lovely example of a technical semantic decision changing the nature of an entire product.
- 24. Retweets Hashtags Links Cards Polls The same is true of several other emergent Twitter features.
- 25. this slide from RafïŹ Krikorian At Thington we were inspired by a proposal published by Twitter to their developer community that was never implemented: Annotations.
- 26. this slide from RafïŹ Krikorian The idea was to have tweets with âView Sourceâ - that there could be a packet of data underlying any tweet that was deïŹned by the userâs own app (any app that used the Twitter API). We expanded on this idea to create Thingtonâs timeline view, where smart home devices talk about themselves in human language, but the data underlying each update is always available for use in automation.
- 27. Conversation is Collaboration The essence of social media is that humans are really good at collaborating through conversation. This is a fundamental concept for IoT architectures.
- 28. Social media: conversation is collaboration Games: human2machine Data: change over time
- 29. âThe love of complexity without reductionism makes art; the love of complexity with reductionism makes science.â Edward O. Wilson: Consilience
- 30. the Internet of Things should be a conversation
- 33. This material is in part based upon work supported by the National Science Foundation under Grant Number (1621491). Any opinions, ïŹndings, and conclusions or recommendations expressed in this material are those of the author(s) and do not necessarily reïŹect the views of the National Science Foundation.