 Morning everybody. OK. My talk will give a slightly complementary viewpoint to what we just heard from Tim. I've just celebrated my 30th year in industry. September 1986 I joined BBC research department as a graduate trainee, Now, since then I've had possibly 10 or 12 new jobs, and I've flitted around the industry a fair bit. I've got that stage in my career where I've seen most things, and I am now starting to recruit junior engineers to join development teams. Now, when I went through the system in the mid-80s, mae'r ei gadw o'r ddysgu'r ddifrwng yng nghyrchu yng Nghymru a'r ddifrwng yng Nghymru. Rydyn ni'n gweld â'r ddifrwng yng nghymru, ac yn y ffordd, rydyn ni'n gweld yn ei wneud yn y roedd y coda'r. Mae oedd yn ychydig yn ymdweud am y ddifrwng yng nghymru, mae'n gweld yn ei ddifrwng, ond o'r ddifrwng yng Nghymru, mae'r ydyn ni'n gweld y ddifrwng yng nghymru, o'r ffordd i'r cymdeithasol y gallwch chi'n gwybod ar hyn o'r pethau o'r pethau. A'r hoffa'r cymdeithasol wedi'i gwneud o'r llyfr o'r cyflwyno. Felly, y rhefnod ar y cyfnod o'r ffordd o'r cyflwyno o'r cyflwyno o'r pethau o'r pethau. I'm going to crack on through the first few slides. Some of it is open for question. I'm going to make a few statements and I'd like you to perhaps think about what has been said and perhaps question some of it. So what I will say is education fundamentally the ultimate in open processes? It might be free to use up to a certain level. Teaching materials are widely available. You can use them, you can update them, you can enhance them, you can re-issue them. It is an altruistic process that's really done to improve the wider interests of society. These days there's just so much material out there on the next that students have opportunity to engage in tutorial materials that, when I went through the system, it was impossible to get the information that you needed. Data sheets took three days in the post to arrive and when they did, your design ideas had moved on. The good thing is that if you want to know about something these days, you get online and somebody out there has done it. If you're trying to get started with a new microcontroller and you want to know how to set up the SPI ports or whatever, somewhere out there, somewhere on the planet, somebody will have written that bit of code and you cut and paste and you say, ah, so that's how you do it. So there's still people out there reading the data sheet and doing this stuff for the first time. So what's new in 35 years? Well, we've seen the advent of massive online courses. Gorilla learning, this is all that sort of grey market educational material that people are putting on tutorials and Hackaday and Instructables and there's a wealth of material and how just to do things out there and it's not come through the traditional educational groups. We've got open software and open hardware movements that's certainly open hardware. That's only something that's come together in the last, well, it's less than 10 years. We've got new platforms. There's just about every month there's some new Linux widget or an IoT thing that's interesting. If you troll Kickstarter, you see that someone's managed to get a mobile phone chip onto a thing about the size of your thumbnail and they're selling it for somewhere between $5 and $9. So we live in massively exciting times in terms of what's becoming available. It's relentless and how do you keep up with the technical change? Also, compared to when I went through the system or when I was born in 1965, the world had half the population that it has now and there are hundreds of millions of people entering the education system every October and they will all have their learning experience and they'll come out three or four years later and they will want to find gainful employment. We have to find ways of preparing that bright new talent for the world that's ahead of them I came out in 1986 when we still needed grunt engineers to work in the defence industry, but I didn't want to follow that route. I went and joined the BBC and I didn't want to just become some guy in a cubicle plodding away at some weapon system. Times are changing, so how do we prepare all that new talent for the digital revolution? Okay, so what's good? Well, a larger percentage of students can now access tertiary education and get their first degree. There's a lot more choice based on the American system. You can now pick and mix your course content and you can do a module of this and a module of that, whereas we, as I recall, it was just dictated. You will do this and maybe in your final year you might get a couple of options, but there was none of this so I'll start this course and I'll get some credits on that and I'll change and do that. So there is a huge amount of choice. I said earlier, information has never been so easy to access. You can guess it's on your laptop, you can guess it's on your tablet, it's on your smartphone, it's just so easy. And also undergraduates are travelling around the world to take part in this global revolution of education. They want to, particularly from Southeast Asia, they want to come to the western countries for a cultural and an academic education. And the knock-on effect of all these millennials coming and travelling to the west and getting what's effectively a western education, the knock-on effect of their society within the next 5, 10, 15 years is going to be enormous. What's not good? Well, massive increase in costs. I went through the system free of charge. I marched over Westminster Bridge and was set upon by police horses in November of 84 when we were fighting to retain the student grant. How times have changed. I don't want to be 21 years old and 40-something thousand pounds in debt. We never had that burden put on us. The evolution of standards. I interviewed many new graduates of electronic engineering who just can't use own's law. What are they teaching them out there? So, you know, the education system has to get aligned with the needs of industry. Sadly, you graduate, you've got your certificate, and you always mean that you're going to go straight into a fulfilling career. The graduate trainee schemes that so many of us embarked upon for 18 months when we first graduated where we were sent back to maybe a special training school, particularly at the BBC, we had Wood Norton where they taught us how to work TV cameras and work radio studios and men transmitters and all sorts of things. Those schemes have been eroded away to the point that they're no longer there. Sorry, I missed one. Right, massive online courses. One of these that I came across a couple of years ago is from Nan to Tetris. How many of you have encountered this? Right, okay. I suggest you Google it. It has to be the poster child of online learning. Shimon Shokan and his co-author Noam Nisan, 10 years ago they put together a book to try and get engineering and computer science students a better overall picture of how computers work. So, they started from first principles with Nandgate, and chapter by chapter they took you through a series of layers of hardware so you actually designed and implemented your own 16-bit computer. Now, all of the materials to do that were PC-based and there was a hardware description language and you could build your own gates and simulate them and you build up from simple sequential and common tutorial logic, you'd build shift registers and multiplexers and arithmetic logic units and program counters and after five chapters and maybe five days' hard work you'd got yourself the hardware design for a very simple 16-bit processor. I rattled through the first five chapters of that book in a few evenings and when I put that thing together and it's actually executed code you have that aha moment, so that's how it works. Then I hit a bit of a brick wall because the next seven chapters of the book were software starting with machine language for that processor and it was really back to school for me to get my head around some of the layers of software from the machine language to assembly language to compilers to virtual machines to high-level languages. This was a big hole in my education and I wouldn't have realised that unless I'd followed the Nandatetras course. Anyway, we can all go back to school at some time. With online study there's a vast wealth of new academic material becoming available and the good thing about online material is you can pick and choose and you can dip into it as you want. You may be holding down a full-time job and you just need refreshed and certain techniques and you've got the opportunity to mix your work and your study. We're likely to see a massive growth in this in the next 10 to 20 years. There's going to be a lot more of it. Here's Shimon Shoklin's book. That was him aged five and he's already got a primitive computing device there. If you do the course, I recommend that you buy the book. It's about £20 and it all goes to supporting this that I believe is a very worthy cause. Had we opened it up, well, open software and open hardware, they present tremendous commercial opportunities. Today's young people are no less motivated than we were when we were 16, 17, 18, if you saw Amy's talk yesterday evening at the festival, it was inspirational. The elder generation often puts the young generation down saying, well, they're not interested in the demotivating and all the rest of it, but they've got the same brains that we have. They've got new and different skills. They can type on touch screens without even looking at it and they are going to go on and they're going to become the new generation of engineers and scientists. We have to find ways of training those people so that they meet the needs of the new evolving digital industrial revolution. There's a need for affordable products and platforms with any open technology. There's a strong community and sharing ethos. These techniques are applicable both to the western world and to the developing world. If we can find the right products at a low enough price, there's no reason why this open education can't be a global phenomenon. It all helps to break down the geographical and cultural barriers. I'm currently working with a young lad who lives in Shenzhen. He recently came to the UK. It was his first time ever out of China. We drove around the UK on a tour of hackspaces at the end of June. He couldn't believe how green the countryside was. He couldn't believe the number of cows, sheep, horses and whatever in the fields. He had only ever got his ideas of what the UK is like from what he'd seen on Chinese websites and actually travelling, as so many young Chinese do, to get an education in the UK and North America. That opens up their ideas to the western ways in which things are done. That can only be a good thing if we're to have a better exchange between East and West. The emphasis should be on learning by doing. I'm a very practical guy and I've got certain skills that I picked up by watching and learning. It was a young Chinese lad who was changing microprocessors on a circuit board in the Shenzhen factory who taught me how to take a microcontroller off a board using a soldering iron with a massive bit on it and how quickly he could change a micro, get the old one off and get a new one on using very primitive tools. Once you've seen how people use these skills and you're able to copy them, it's tremendous. Finally, I think young people ought to be able to demand more from their £9,000 or if you're an overseas student, £16,000 per year. Okay, new platforms. We've got a mix of platforms and the open software community is coming up with packages that will run on multi-platform. There's a whole wealth of single-board computers like the Raspberry Pi, the Beaglebone Black that Tim mentioned. If you want something with a bit of grunt, there's the Ojoids and then there's IoT devices like Omega 2 and all of these little wireless connected devices. Basic engineering design tools, they should be opened up. They all share a sort of common graphics core and maybe there should be a concerted effort to try and make the tools compatible with each other and make it simple and it doesn't matter where you are, whether you're in the East or West, there will be a set of design tools that have been tried and tested and made to work. Let's get the tools that we need and make them accessible. ChiCAD is a start, but we need to follow that up in MechanicalCAD and we need to be able to work with systems that link neatly into laser cutting and CNCing and 3D printing. We need to be able to access low-cost global manufacturers in North America if you've got Osh Park for a PCB manufacturer. Students ought to be able to design things and send the design files off and a week later they come back from Shenzhen. If they could do that, there would be no stopping them. Here's Odroid. I think the fan gives it away that it's got quite a bit of grunt. A blatant plug of a new product of mine. Do you want to step up, sir? I'll start around the audience. Alan's going to be talking about My Storm later, so all I'm just going to say is it's a multi-headed beast and we're learning how to tame it. It's got a 32-bit arm and it's got a big FPGA and you can plug a pie in on the top connector. Do it yourself. Imagine a situation where engineering and computer science students are issued with a very cheap bit of hardware in Freshers Week and it's theirs to keep and build upon and basically it's probably a Linux platform with some FPGA and it allows personal experimentation anywhere. You can do it in your study room. You don't have to have access to a lab in order to tinker with hardware. You learn by doing. Students could build their own instruments like multimeters, digital oscilloscopes, logic analyzers, protocol sniffers. These devices, they're not overly complex and if you give them a platform that's got the processing power to do that, they can design their own, point them in the right direction and have the means that any designs that they come up with can be quickly replicated and shared around the campus or shared around the world. So one possibility is that they build up their own tablets or a laptop, for example something on the lines of Pi Top which is an open platform, it uses a Raspberry Pi, you put it together and there's a big hole in the middle which allows you to put your own hardware, for example an experimental breadboard or whatever. So there are some innovative ideas out there and how we can come up with innovative products that make this so much easier and accessible. Okay, five years ago there was MyDac which is a kind of data acquisition system. It's from National Instruments, it's £142. The breadboard that goes with it is £58. So you spent £200 and you've got what, 20 quids worth of hardware. So that needs to be opened up. Right, new languages, C++, they provide a framework. Students should be encouraged to code. New languages are constantly out there, JavaScript, Python, Lua, and then hardware description languages like Verilog, VHDL, an open one, a recent one called MyHDL and one from Nicholas Verth, Pascal Pioneer called Lola, logic language, I believe that's short for it. Okay, new approach, teach a man to fish. Engineering students, they have a common first year where they're given practical instruction in 3D printing and laser cutting and soldering and coding and sort of stuff that hackspaces do. So they can all have fun and generally get them motivated. The introductory year gives them access to basic mechanical and electronic engineering coding skills and those design tools that I spoke of earlier. Practical content, so every first year undergraduate can lay out the PCB or write an Arduino sketch or whatever. Then later in their career an industrial placement year I recently had a young lad join me from Vietnam and he came to me as a very, very shy 20-year-old. I threw him in at the deep ends. In his first week he designed and laid out an Arduino-based processor board and then I taught him how to solder it. It was all surface mount and eight months later he was laying out four-layer boards for me. His soldering skills, partially due to his acute eyesight were much better than mine are these days. How did you get it started? Well, Shimon Shokin's The Elements of Computing Design shows one successful approach. Open platforms, I've mentioned Beaglebone, pocket chip. There's more and more of them out there. Low-cost hardware are very on my storm. A collaborative effort to produce easily accessible and well-researched content. It's all very well-having platforms but platforms are useless without content. Then you need to build the strength of the community and the ecosystems around those platforms. Industrial placement, I think I heard this two months ago but if you spend a year in industry somewhere in the middle of your academic course you learn to work with a team rather than perhaps individual study. It improves your communication skills. It gives you a sense of purpose to reinforce your academic studies. It always looks good on your CV and my tutor, he went on to get a good job working for a satellite communication company in Hampshire. He was delighted with that. It can be an enjoyable experience to both the student and the mentor. What can be achieved? Can we modify the university system to allow for a greater practical content somewhere? An online network of affiliated course providers. We need to get industry involved. They know what they need in terms of what skills graduates should have. Maybe a sabbatical exchange, that's a two-way exchange programme so that industry and education they can swap roles for maybe a year at a time to learn a bit more about each other's environment and discipline. With the right education you can always stand out in the crowd. That's me. OK, thank you very much.