 Mae'r ffordd yw'r cyfnod o'r 18 o'r ffordd yma yn ei wneud yn fawr o'r BBC i helpu'r hyn o'r cyfnod o'r microbyt. Yn ymgyrch yn oed yn November 2014, mae'r ffordd yn ymgyrch yn ymgyrch. A'r ffordd yma, mae'n rhaid i'r mwyaf o'r cyfnod o'r cyfnod o'r cyfnod o'r cyfnod o'r microbyt. Mae'r ffordd yn eu empethu sytheader Gruffudd yma, ond dysgu eu hwn sydd yn ymgyrch, ac mae ymgyrch €'r ffordd o'r cyfnod o'r cyfnod o'r cyfnod o'r cyfeirا�'r cyfnod. Mae gen i yn mynd i hynny oherwydd mae e'n... Dyna ydy'r gwrdd hynny dyfod o'r d Sultan Argynyr. Wrth cariat ddiwylo hwnnw ei ddim sydd yn gudd gwneud o'r ddinfaith ar gyfl어�anyaid buべ addurau esŵu mewn gwein. Be'r oedd yn ymddangos i'r iddynt ychydig i'r cyflwydu. Ie, yn ymdryd wedi'i ei wneud hon i'r cyflym cyflwydu. Felly... ...yna, efallai'n ddigwyddiaeth i chi i'r cyflwydu? Mewn ymdryg ar y gyflwydu hynny ymddangos, ymdrygu Howerd Baker, yw'r cyflwydu yma yn yr Oedd-BBC. Ac yn ymdrygu'r 21 ymdrygu. I don't know how many of you saw the presentation yesterday, a quick show of David Allen and I found it really, really interesting, the parallels and the differences between the BBC micro development of what was getting on for 40 years ago. And some of the same hurdles that had to be coped with internally within the BBC from this. Things that are, you know the wonderful thing other listing magazines are available, the BBC have to comply with a very level playing field, particularly in the year that Microbit was coming about and of course it was the Royal Charter renewal year and I don't understand why but there seems to be quite a few people with political gripes against the BBC. I don't, I tend not to follow most of that stuff but they had to not favour any commercial opportunity. The other thing was they were not allowed to spend any licence payers money on, it is very ring fenced on what they can spend money on and what they can't spend money on. So apart from funding a few people to allow some of their time on it, they were allowed to make programmes, they were allowed to make help create some of the content from the education budget. But they weren't allowed to put any money into the actual delivery of the hardware itself. So my first involvement was I got called in, a fair few of you will know Alexandra Deschamps on Ceno, the person who most things IoT flow. She'd been working with BBC R&D for quite a while advising on IoT stuff and when this project came onto her radar they said they needed someone with an experience of high volume manufacturing cost sensitive toy type stuff which is what they called me in. So my background is I worked one of the big consultancies back in the late 90s doing high volume cost sensitive stuff so it was toys, consumer end of medical, things like counters for aspirin halers that were kind of rattled off in 10 million off. So I had a, as we'll see, the micro bit is or isn't a toy. The definition of toy or targeted at children is open to interpretation shall we say. So the first involvement was helping the BBC understand who they might be able to go out and ask for a million pounds each which is quite a fun thing to do. Bless them they didn't particularly understand a great deal about the semiconductor landscape. So kind of first explanation is arm don't make chips, arm design chips, but that's actually played. It kind of measured in quite nicely because by being on a being having a centring on arm they could then invite various different silicon vendors to contribute some chips each. And we thought it was possibly more likely that you might be able to get 300,000 chips from one vendor and 300,000 from someone else and 300,000 from someone else. Oh, quarter in time. OK, so this is, when I first went up and chatted to them and Howard, I completely got it. I understood where he was coming from and as I said to them when I went up there, if I gave my kids, if I took a Raspberry Pi, I have an 11 and a 12 year old that you'll see in a minute. So they're the right age group for, they were sort of the target audience anyway. But if I took a Raspberry Pi and I attached the very best, brightest screen that I could lay my hands on and I put the best, capacitive, projected, capacitive touch stuff on it and I nursed every last bit of horsepower out of the GPU. My kids would go, yeah, but it's not an iPad because that's what they've grown up with. They haven't grown up with an iPad, I hasten to add, but they've grown up with cheap Android things that are never really quite work and they resent me for it. But kids today, they have grown up with, the majority have grown up with touch interfaces. Toddlers will pick up a toy and try and swipe it. It's just instinctive to them to do this sort of thing. My experience has been that trying to get my kids interested in things like scratch, when they're doing things that go on on the other side of a piece of glass, it's a virtual world that they're very familiar with. But whatever they do, their first efforts, and probably the first five, ten years of their efforts, if they go into video game programming or whatever, are going to be disappointing compared with what they've grown up with. There's a certain gratification in making the catch run around on the screen and then you get imaginative and turn it into something more dodgy. But there is a very different world that lives on the other side of the piece of glass for me. So the counter of that is that my kids occasionally stick their head into my spare bedroom lab and they'll see a little servo motor twitching or something like that. And they go, oh, that's cool, what's that? And they're intrigued by these things. It's not something that they've really come across before. I think we're sort of... I think it's something that we're all on board with, that electronics kits and getting your hands in there and sometimes letting the smoke out is a really great place to start, that great excitement from the first time something works and great mystery when the thing doesn't work or stops working. So this is actually an example of something that was put together by a seven-year-old just recently and it ended up using the micro bit with touch sensors, buzzer, flashing LED on Rudolph's nose. Down there. And it is, if something is physical, even if your mind then enhances that, the imagination, it enhances it far more effectively than you're trying to create that via graphics on a screen. It's certainly not a... I have great respect for the Raspberry Pi, so I'll end Eben Upton, came along to some of the early meetings with the consortium. So it's very much addressing a different aspect. But to me, this is real hardware. It's Arduino style rather than operating systems and that kind of programming. So yeah, hands up confession. I'm strange, I'm a geek. The strange thing really is that I'm a creative engineer. At least I can say myself so. I'd say most of you here are more creative and probably more engineering than me. I won't say you're stranger than me, but we have to accept that we are a few percent of a few percent and people who come along to this are a few percent of those who are bothered enough to do this in their own time. So we are, if you like, the real elite or to borrow a phrase from one of my old bosses, Adrian Woodward. He used to explain bell curves and he talked about his three Sigma shoes. He had a pair of red driving boots that he wore all the time. He drove those because he had a TVR and he enjoyed the extra excitement of touching the pedals and blah blah blah. But he always said, yeah, I'm 99.7% of the way up that bell curve in that. So whenever we were trying to design products for the general public, we had to detach ourselves from our mindset and remember that, yeah, we are a bit strange. So going back to the BBC micro days of 40 years ago, people like us or the equivalent people as us then were capable of getting a load of seven four series logic chips, soldering them together and creating the things that we create. But it was the democratisation, if you like, of the ZX81s in my case, the VIC-20s and the BBC micro. So it really took that and opened it up for most people to get involved. And it's a lot of the scene around the kind of industry that we're involved today, but I think really harks back to the early 80s that drew us into it, but also got UK involved in that kind of thing. So, and this, one of the partners, Tech Will Save Us, helped us get our heads around what should be in and what should be out when we were doing the cost reduction. And they said divide the kids into four. There are the currently capable and the less capable. And there are the currently interested and the less interested. And I remember Beatrice in Block B there was someone who already had ideas for what they wanted to do and wanted to get involved with this kind of thing and had some kind of capability understanding of what to do. And the key point is Microbit is not for Beatrice. Beatrice can go out and buy an Arduino. Such things already exist. Beatrice can lash things on to a Raspberry Pi or alternative products were already out there for that. That's probably perhaps three, four percent of the kids, if that perhaps more like one percent. The key point was to take those who were interested but not capable or capable but not interested and enthuse them or enable them and get the, if you like, maybe the perhaps, we were hopeful that maybe it was probably something like 40 percent lived somewhere in the A and D sectors there. We've moved them in. And we had to accept that there were going to be the seas who didn't know what they were supposed to do, weren't interested and they were probably spinning them across the playground or something like that. But there was the opportunity there for them to move into one of the other categories and then into where we would like them to be. So making that real, Sam and Izzy, my kids, they're both into football actually so don't take gender stereotypes here. But Izzy's an engineer. Thanks to Chelsea here last year, she learned to solder while she hadn't come into my room in the previous three years for me to show how to solder, I don't know, but dad knows nothing. Sam, I try to inspire into this thing, but he's having nothing to do with engineering. Incidentally, yes, he did touch him and yes, we did get a free cake and we scored from it. But making this real, I said to Sam, what would interest you? What can you do with your football with this? I say, maybe you put a pedometer in your boot and you can see how many steps you take during a football match. Give them to your friends and see how much running each of you does and compare at the end. Yeah, whatever. But about 20 minutes later, he came back and he said, actually, would I be able to put one on each shin pad and count how many times during a match I play the ball with my left foot and with my right foot? Because Sam's actually really two-footed. One of the things he's really proud about is that all of his mates can only play left wing or right wing. He wanted to show off to his mates that he was really two-footed. That's the thing. There's so many example programmes out there for heart rate monitors, pedometers, that sort of thing. Whatever we try and dream up for kids, the majority of the ideas are going to inspire them. It's actually the kids' ideas themselves that need to be enabled. I think that it goes beyond kids. I'll confess that most of my interest in getting involved with microbit was a bit selfish of course. I see a lot of entrepreneurial ideas around the accelerators, the startups. And so many of them, particularly in IoT, where you've got so many disciplines required to even get to a prototype stage, they come from a very small select group of generally engineering-minded people. What I saw was that beyond 10 years ago, prior to 10 years ago, you had to be one of the people who I'd either studied a couple of subjects at university and combined them, or you had to have grown up with the practical electronics, wireless world and so on. In my perception, Arduino came along. I know there were other alternatives for that, like PIC, basic stamp and those things, but Arduino was kind of a step change in that generally, finger in the air 10 times as many people were now able to make some sort of prototype because you didn't have to pick up a soldering iron and the programming language was more accessible. And that's great because a lot of techie-minded artists, the maker movement were suddenly able to express their own ideas without having to persuade an electronic C-type person to help them out. However, you did still generally have to install something on your computer. You had to go through configuration steps and you had to be reasonably familiar or accepting of square brackets, curly brackets, semicolons at the end of the lines, those sorts of things. What I hope will be happening with the next step from that with things like scratch style blocks, programming, that sort of thing, but by making it more accessible, we might end up with a further tenfold increase in the people who are able to do things. So, for example, the start-ups that I work with, most of them will reluctantly try and get into a bit of Arduino, but there's a whole load more who really can't get their heads around that. It's going to be the plumbers, the electricians, the horse owners, the people with in-depth knowledge of the application that come up with the best ideas. If I can get them trying out their first ideas on something as simple as this, if they're not patronised by reading a tutorial written for an 11-year-old, then the best ideas will come through. I'm happy to sit here and take good ones into production. I do believe in IoT, but I don't think it's the fridge ordering milk. It's not going to be 50 billion thermostats. We were doing remote thermostats back in the 90s, and my boss at the time said, you could do all this 20 years ago before. Actually, most people aren't that bothered about turning their heating on before they leave the comfort of their office. There is a killer app out there somewhere. I don't know what it is, but it's going to be something that we haven't dreamt up. I'm prepared to bet that in 10 years' time we'll be here going, oh yeah, that's where the 50 billion devices are, but none of us at this stage have seen what it is. That's why we want, hopefully, 1,000 times as many people out there trying their ideas. Particularly, I want my kids to be thinking of the technology as a tool, rather than waiting to be spoon-fed by whatever, waiting for someone else to write an app that behaves in the way that they want to use it. Could you refresh your screen? How am I doing? I'll press on. The actual making of the mic a bit. This is Johnny Austin from ARM, one of the embed team. Probably the key guy from ARM's side in actually getting things during the development. His colleague Chris Stiles actually laid the board out because although ARM were on board with it, it was largely down to people to find their own time around their other work to progress with this. Chris laid out the first board on a flight back from Shanghai. There's actually a picture of him at about 37,000 feet over Moscow, a capture of his seat back from Shanghai. I finished the layout. Will anyone hear claim to have laid out a PCB, and it being perfect first time and didn't have any revisions? The first board, there was the classic thing of the USB connector being on the wrong layer, so the five pins were reversed. That board, using the fast turnaround people that they'd used before, took two weeks. I said, okay, let's see if we can do this board a bit faster. We then shrank it down to, this was the first microbit to be used, so Chris was on another flight back from Shanghai and did the same again. He revised it a little bit over breakfast, and at 8.30 in the morning it was still data on his PC. Just south of Cambridge in Safran Walden, there's a board layout on the board, a board production company, RAK, who we happened to know from 20 odd years ago. This is where the real people connection comes in, because my board layout guy, Genesis Pro Systems, happens to ride a bike and is a big enthusiast and the owner of RAK is also a cycling enthusiast. It meant that John could go down to RAK and stand there at the end of the machine and talk about who was going to do well in the Tour de France this year or whatever. Then as the boards came out, grabbed them, hopped in his car, drove up about 10 miles to the north side of Cambridge to another little company, EFS, who specialised in being able to hand-solder non-hand-solderable components. They sat there and they ran up two boards with loads of little 0402 capacitors and the kind of things that if you cough, they fly across the bench. The key chips on there were sort of 32 or 48 pad QFNs with only the connections underneath. So their specialism is doing that fine work and then popped it back down to John Lancaster who took it in, squirted the code in and it was back on in Arm's lab at 1830 running firmware. I'm a big fan of Shenzhen having worked with Chinese toy manufacturers for many years. They do things great, but I do just want to rile against a little bit the incubators and accelerators who say you can't do any of this here. The only place you can do it is Shenzhen, which is not true. A few of you might know Adrian Godwin, so we had a quick pop into a local EMC facility that we were allowed to use ourselves rather than submitting for tests, do a quick bit of radio testing. Adrian ended up having to drive most of the way around East Anglia at one point, getting some radio testing done because there was one little peak that was probably going to fail the radio tests on the first one. Actually, that came about. The first batch that were manufactured in China and came over, there was a component that it was decided would be better not on there. Walkbury, a company in Peterborough who specialised in rework, were again very accommodating. I phoned them up and said, do you think you could take a component off some boards for us? I said, what's some? I said 26,000. They said, pop them up. I was blown away that they'd be set up and ready to do this. I didn't think rework happened like that in those volumes in the UK. They said, it's nothing special. It's really quite common for people to get things manufactured and the samples all come through fine. The shipping container gets cleared from Felix Stodox. The bright-eyed young entrepreneur opens up the doors of the shipping container and goes, what on earth has gone on several time zones away? Misunderstandings happen. Walkbury's main business is doing rework rectification of things that units don't have to get sent back for another six weeks on a ship. One of the things I really enjoyed about this process was jumping back. It was about 15 years since I got out of really working on toys. The working at a million off is great. You kind of go, shall we have this component on or shall we not? Then you kind of go, okay. So one million times that component is about the price of a car or one million times that component is about the price of my house. By this point, the consortium members didn't have any more money that they could raid from their CSR budgets. It really was quite critical that we had to get these million units made within the existing budget. There was a whole load of extra pressure because a certain... I won't say because but part of it was a certain Mr Clarkson decided to really, really wanted a steak sandwich or steak and that had quite a budget impact because the Top Gear franchise going off was a big expense to the BBC apparently. I also remember sitting in one of the meetings and going, okay, so how things are going to go on on the production line? And they said, oh yeah, it'll be all right, it'll be all right. And I sort of sat there and did a quick mental bit of maths in my head and I went, okay, one million units divided by four weeks, divided by 80 hours, allowing for there being far more working hours in China than there are in the UK, divided by 60 minutes, divided by actually it comes out at about one per second. And then we said, oh actually, we think there is a problem there because it's going to take about 28 seconds to program each unit. So we applied some of the old thinking, doing clever things like having the boards test themselves but then actually say, yes, I'm programmed and I'm working properly over Bluetooth because that would in itself test the Bluetooth. In the end, it turned out that one of the most likely failure points was missoldering of the USB connector, which there didn't seem to be any convenient way of testing the USB connector other than plugging it in anyway. In true China style, I think they took a sit, people about 50 wide and plug in every single unit. We had to squirt firmware onto the devices anyway, so we could have done it with better nails but it turned out that many hands was the way around that. OK, I can go on and do a little bit about a... No, I'll just skip to the results I think. This is the thing I find most gratifying. The key one is that first one, that one year on the percentage of girls who said they would choose computer science as a subject option rose from 23% to 39%. Going back to it not being for the Beatrice sector, it's that 90% said that they had an expectation that anyone can get in and code rather than it being something that you have to leave to other people to do for you. The other important slide is the micro bit is now no longer anything to do with the BBC. The BBC were involved for a year and then it's now handed over to the micro bit educational foundation and it's being rolled out around the rest of the world. Croatia, there was a crowdfunding effort there that bought micro bits for a thousand of their primary schools. I think it was about $300,000 they raised to buy micro bits for that. But the micro bit educational foundation is very few people. Johnny Austin that you saw kneeling on the floor before is doing most of the stuff there on a day-to-day basis. I popped him a quick message and said, hey, are there any calls to action that I could put out here? His message is please make accessories, add-ons for it is what they would do. Techmicrobit.org is the place to go for all of the information on that sort of thing. Make lessons that enthuse kids, David from Duckbot yesterday. That's the thing that excites kids, it's things catching fire. Obviously that's not something that BBC could do but those are the kind of things that actually get kids interesting, sticking them on rockets, putting them on weather balloons and so on. Put those cool projects, they've got a place at Hackster, is where they're putting things. If you're on the software side, improve the dial, the runtime, the editors and particularly if a simulator for MicroPython would be hugely appreciated. The design is now open source. The reference design using a relatively available radio module because one of the difficult bits that most people would struggle with is actually laying out the RF section of the board. This reference design is available and unfortunately it was originally created in Altium and is a very expensive seat to have. It's now been transferred to KeyCAD and some of the lower cost tools. On-going work on getting the reference designs translated into widely available tools is something that they really want. I think that's probably about my... I'll take a few questions if you like.