 We've had a very good day. We had a lot of discussion. We went through at least six sheets of paper and for half the day, we kind of just got to know each other and what everybody was doing. We tried out the new super duper coffee machine with the tablet on it, which is very good. No, the first half of the day was to understand the aspects of each other's projects and then find some common ground between them. And once we'd done that, we realized that we were using perhaps three different wireless standards. Did I say standards? We were using the sub one gigahertz low power, which is the ISM, so either 433 megahertz or 868 megahertz. That forms the backbone of the open energy monitors transmit units to their data concentrator. 433 is great inside the home because being a lower frequency, it has good penetration of concrete and other dense building materials, much better so than some of the 2.4 gigahertz stuff. So we've got low power sub one gigahertz. We are also doing things potentially with wifi in terms of the ESP8266 and also to throw into the mix, I was keen to bring in some Bluetooth low energy devices. Now the great thing about BLE is it's supported by both flavors of smartphone and it's very easy to use your Android or your Apple smartphone to control devices over BLE and also to extract data from devices using BLE. But also BLE, you can have a peer to peer type network. So the idea is to have some sort of a Bluetooth low energy thermostats which is either a room based thermostat or an individual based thermostat, say on a pendant that can be worn perhaps by a senior citizen that can interact with the whole ecosystem. So to give you an example, normally this thing attempts to be smart, if you've got a mechanical TRV at home you're used to putting it somewhere between frost and maximum and then it maintains the temperature around here and this will do the same thing. But normally it observes, it also observes your occupancy the room and tries to cut the heat when you're not around and when you're not likely to be back, which is fine if you all agree on what that temperature should be. So this measures the air temperature around it to drive that. But if you could then have, can you have you got your dongle the Bluetooth little pendant thing? My voluptuous assistant will show the BLE. So if you gave this to an individual to wear about themselves or to put, no the little Bluetooth one, that thing, oops, or throw on the floor, put on the coffee table near them then that takes over and it's thermostat. It's temperature sensor which would drive this instead. So that's quite a novel idea that as you walk into the room your temperature profile then what drives the device? And also if you're sitting on a sofa only about 600 or 1,000 millimetres off the floor the temperature down here is a lot lower than it is up here. So having a personal thermostat that you have close to you means that your little bubble of heat is going to better be reflected by the overall control system. So if, for example, a senior citizen who's spending a lot of time fairly immobile, perhaps sitting down in a chair for a long period of time it means that they have the maximum level of comfort. And being a personal thermostat they would have some sort of a button on it where they can increase the heat level to suit their requirements. So that could all be done using Bluetooth low energy. Can we check in Laura as well? All right, okay, I learned a lot about Laura. I actually went out with a girl in 1980 called Laura but that's another story. This is, what's the acronym? Long range, low power, yes. This is quite a new initiative which allows kind of sort of kilometer distance type or multi-kilometer distance connectivity between a low bandwidth network and a device. So the way I sort of rationalized it's Twitter for IoT devices where they're sending 140 characters up to the cloud and they're maybe getting four character expletives back again. That was my understanding. Is that about right? So IBM, who were actually really helpful to us a few months ago and got us on Laura, it's really the smart city end of the internet of things. So curiously it's on the same band and using effectively very similar radios to these ones. But these, so BLA gets you within a room, sort of 10 meters, these will get you sort of 100 meters and Laura, which is in the same band as this is, I am, ISM, unlicensed stuff, will get you well one or two kilometers in an urban area and maybe 10 kilometers in a rural area. So you can see there's a whole span of different radio devices. There's not just one radio internet for things, but the Laura one, we are putting sensors in bus shelters in London. If you see a couple of dodgy looking guys putting things with wires in them in seats in London buses, that's okay, that's us. I'm very disappointed we haven't been arrested yet, by the way. And so the idea is we can go to the bus operator and we can say, yes, it really is possible to put a low power radio that will last for several months into one of the existing voids in the bus shelter in our case to sense the presence of people. And so it's, as you said, well, four different, five different radio things. So we were talking about bringing some of them together. Right, okay. We realized that we're dealing with multiple radio standards. There was no one size fits all. So we then thought about, well, what about the data packets that we're trying to send between these different radio standards? So the last couple of hours of our discussion today was basically to look at different packet formats, some of which Damon has already put a significant amount of work into that are not only authenticated but secure. So within, let's say, a typical packet that we can send over, say the RFM69, ISM, Chunsever, we can get a fully authenticated and secure packet that has a 32-byte payload. And so we discussed ways in which we could pull all this lot together and get it in to the, get it compatible with the Open Energy Monitor, Emon Pi, which is their base station cum gateway. So just as a little example, IoT tends to be done. Security is done, if at all, as an afterthought and often badly. Security's a cost, not a benefit in many people's minds. Now, my view is, if we're gonna put this on 100 radiators, well, you know, that's too bad. If you're gonna put it on 400 million radiators, we really don't want it to be possible for board teenagers to turn people's heating off till it breaks. My nightmare scenario, never mind the privacy of burglars being able to tell when you're out, is also the board teenagers sitting on the 17th floor of a tower block, completely disabling the heating and nursing home next door in the middle of winter. So you've got to get security right. And so this thing is now spitting out, I'm sure you're all feeling the radio waves here, is now sending out stats about the room it thinks it's in and the radiator thinks it's controlling every four minutes in a completely secured frame format, which, so as you say, it has an authenticated element, which is the things like the IDs and the structure of the pack and what type it is, and a completely secure payload protected with, now I haven't written this myself, don't worry, it's not a homebrew thing, it's AESGCM, your browser can talk that as well. And that, we've had a number of pairs of, a number of sets of eyes, look at the implementation and the design and so on. So I'm sure there will be amendments to be made, but this, I think, would be tough for the NSA to break. And on that subject, by the way, I don't know if you remember little prism program that where they were eavesdropping on lots of people's telephone conversations and so on. That little prism logo, they stole from my dad. So we've got a connection in the NSA already. So the point is that if we were to, if the packet format we've come out with this is also usable for OEM, then suddenly OEM's internal wireless communications are safe enough that no one is going to be able to mess with them, either to turn things on or off or whatever in an unauthenticated way or to eavesdrop on the traffic. So that is the aim in getting the authentication and the encryption. Of course, you can just hit any spies with pieces of equipment that you've got there. And also because some of us are still into hardware, we had a very good session at a Google hangout with Martin Harazanov, who's based in Sophia in Bulgaria. Martin is rapidly becoming the expert at porting applications onto the ESP8266. And by means of a video link, he was able to give us his presentation. Martin some months ago produced this rather good three-channel relay board. One of our intentions is to build, this has already got the 8266 in, our intention is to put a little board in this area, which we can put the RFM69 and also a Bluetooth low-energy module, probably the RFduino, which uses the NRF51822 device. Build that into here. This goes in place of your central heating controller and this acts as the bridge between the various networks. So that's going to be a little ongoing project. How are we doing? So our proposals, we came out with eight things, most of which we've covered. So we're going to do the relay board with BLE and the RFM. We're going to have a continuing investigation into the NRF51 and 52 Bluetooth low-energy devices. The microcontroller board, the Y-Node 5 that I did for the open inverter, that's going to be made available. We've had some sponsorship from Ragworm and they've produced a number of PCBs for us. Unfortunately, they couldn't be here today. Hopefully I'll get the boards tomorrow. So perhaps at the next OSHUG meeting, we're going to be able to hand some of those out for anybody who wants to play with them. And we've decided that this Open Energy Interoperability Group is going to meet again during OSHCAMP, which is in early September up in Hebden Bridge.