Skip to 0 minutes and 8 secondsSo in the early days of the Internet of Things, there was a need for platforms that simply handled the data from sensors and devices and what have you. And in many cases, these were essentially web-based time series databases that were able to handle traffic of scale from a wide variety of devices. Pachube essentially was one of the first kind of open platforms in the sense that anyone could sign up and use the platform, use its API to push data from devices or to pull data from the database to create analytics or other types of applications on top of the data.
Skip to 0 minutes and 52 secondsAnd in many senses, it was an attempt to break down silence, in other words, have a generalised horizontal platform for handling the data from all sorts of different device or thing. But in another sense, it became a silo in and of itself, because the things that we're pushing data to Pachube or pulling data from, only knew the existence of other things on the same platform. And we've seen this replicated now in many other IoT data repositories that actually whatever backend system is used for handling the data of a device limits their capacity for that device really to have any interaction with anything beyond that data repository.
Skip to 1 minute and 50 secondsAnd so what's interesting is that the situation is similar in many ways to the very early days of the Web when people got online and accessed the web through portals, through CompuServe, or AOL, or Demon.net, or even through an academic network-- and in a sense, only had visibility for the things that those networks curated or enabled them to see. And it was the search engines that came along and helped to really break down the silos between all of these networks and enable people to find stuff beyond the network that they were already on.
Skip to 2 minutes and 31 secondsAnd so that's essentially what we're trying to do with Thingful.net, because we're now in a situation where there are dozens, if not hundreds, of data repositories for things and sensors and what have you. But each of them only manages the devices that are on that platform. So what Thingful does is it goes out and searches across the data repositories. And what we're trying to do essentially is enable things to find each other that are beyond the network that they happen to be storing data to.
Skip to 3 minutes and 9 secondsThe reason that this was important was, on the one hand, because actually, with this kind of creation of so many silos of data, our concern on the one hand was that actually the market for data platforms was becoming commoditised. And on the other hand, the kind of siloing of data actually restricted the original dream of the Internet of Things, which is somehow that everything should be able to talk to everything else using the Internet as a backbone.
Skip to 3 minutes and 42 secondsAnd so what we're essentially trying to do is find a way to develop the kind of communication infrastructure you might think of it whereby things can find other things but only according to the terms under which the owner of those things allow themselves either to be found or for the data to be used. So by way of an example, let's say you have a heart rate monitor. You might want that data to be accessible to your doctor in real-time, but you'll make that data available to your grandmother with a one week delay, because you don't want to scare her every time your heart rate increases.
Skip to 4 minutes and 28 secondsYou might then make the same data available to a research group for some kind of analysis, but you want it anonymised. And you might make it available to some commercial organisation, but only if they pay you for it. And in all cases, we want to find a situation where the owner of that data is in control of where the data goes, what's done with it, and who has access to it, and even who has the capacity to find that it exists. We call this entitlement. And we think it's going to be an important part of the Internet of Things.
Skip to 5 minutes and 9 secondsThe idea that data generating devices and even human beings that are generating data need to be able to express who has entitlement to find or use their data. And in a sense, that's actually the longer term vision of Thingful, which is to be a transaction enablement service and entitlement framework for things to find each other and make use of each other's data.
Expert experience: platforms
We asked some of our experts to share their first hand experience with data and platforms.
In this first video Usman Haque, Founding Partner at Umbrellium & Founder/CEO of Thingful, talks through his experiences with Internet of Things platforms.
© King’s College London