 Rwy'n cael ddod. Rwy'n cael ei ddau? Rwy'n cael ddau. OK, o ran y cyflwygoedd ymlaen, Paul oedd yn fawr i gael y cwmhiliadau a'u byddau a'r bwysigol. A gael y cwmhiliadau yn y cyflwygoedd, mae'n gweithio'r gweithio y byddai'r cyflwygoedd ar hynny'n gweithio ar y cwmhiliadau. Felly, dyfodd yn fyddi. Mae'r rydw i'n rhan oedd ei wneud i ddisgolion ymlaen. Ymgyngor y cyfrifyn yw'r cyfrifyn i mi dyn nhw'n gweithi'r fawr. Yn gyweithio cyfrifyn ar gyfer yn ymgyrch, ein data hyn o fawr, a Gw Creator, sony Erickson ac rhai. Mae'n wneud y ddweud yr olygu. Felly pan hynny, mae'r ddweud drwy fawr, mae'r ddiwylliant, gwi'n gwzglas o'r cyfrifyn, tra ynglyn chyrnod o'r cyfrifyn, mae'r ddweud cyfrifyn, tw富'r rhwng yma y gallwn o ddesod, ar ffrwng y dŷl pan fydd rhan o beth bob un o wahanol. Fyddwch ddechrau chi'n cael y pwrwyddaeth sy'n dweud am yr adnodd yma, ond rhan o beth mae'n rhan o beth o rhaid. So fel mae'n ddegol i'r bwysig, Myth ddaen nhw'r cyfrifiadau yma sy'r cyfrifwyr ar y dent o ddod o rheidiol a'r merchandise a'r content yn nhw yw dechrau yn ymweld. Rwy'n byr ar gweithio eu ddweithio mewn ddweith o wahanol maen nhw. felly, roi i'r fflawn o ran y cyffredin ni'n fath o'r users yma? Rhywbeth y netdroff yw'r ffordd yw'r ffordd mewn ymgyrchu'r cyffredin o ffordd yw'r ffordd yw'r belyd, a ydych chi'n dawn o ffordd â bryb ac chi'n ddigonio ar y cyfroed gyda'u ffroed. Ond ycon y cyfroed y gallwch yn ffordd y gyrparu arall o'n ymgyrchu pethau sy'n ddiweddill yn siart fynd i chi oedd o'n wirneidol gael yma o singyntoedd o'r cystafell cymaint, a o'r cyflwyno. Wrth gwrs, mae angen i'r cwrnod yn cwrtol hwnnw mewn cwrnod yng Nghymru, cael ymちrch yn gyfer o'r logo'u gyd yn y blaen y byddiadau ond mae'r cyflwyno yn dda maen nhw i chi o'r cyflwyno a'r cyflwyno yn dhydrwng ac i skrwpan o'r cystafell. A yn y gyrdd rai wneud, ac rai gyrddio'r â'r gwybod frynghig yn oed mewn Apple, ond mae e'n gwneud o thysydd mae'n cymaint Siola wasghaf ar oed saying that any operator wanted to sellter irrespectiving of the fact that it completely is unbranded. But this wasn't anything new. Nokia have themselves been launching a set of services under the OV brand to do things like, you know, syncing you contacts and backing up your phone, mapping all the sort of stuff that Network operators were kind of out there doing, and Nokia are now trying to offer that to their customers as well. So the battle for the customers is fiercer for players than ever, ond yn ymdrygu o'r newydd. Bydd eich ddweud y clyweddau y press, sy'n ddweud, a this is how things break down as far as devices. Apple is doing loads of stuff with iPhones, a toodd anodd iPhones, a toodd androids, a toodd anodd, they're interesting, but they're still small. And then there's everyone else, and that's all the people who've got the old devices. And we've heard this a few times already today. This problem in the industry, we call it fragmentation, the fact that there's lots and lots of different devices out there, ac yw'n dweud y bydd yn gweithio. Yr ysgol ydy'r lleol sydd yn siaradion beth rydych chi'n mynd i'r gweithio. Mae yna dweud cysyllt y lleol, ac rydyn ni'n gwneud fy modol ond y Ffigur Unedig yn ymgyrchol o yr unigol gyda y year. Felly, mae yma efallai ffigur yn ymgyrchol o'r unigol. Cymru yn 95% ddefnyddio Ffigur Unedig yn amser Nwcian, Samson, Sonairexon a Mojarola. Rydyn ni'n dwylo'r gwirioneddau bod ni'n dweud I don't think this contradicts some of the research that we've just seen around ownership in universities because you might expect university students to be more inclined to have the latest devices. But if you're going out to the population at large, this is what was out there in January in the UK. So you might think, well, okay, that makes it very, very simple in, doesn't it? You don't do Apple and Google just yet. If you want to get reach, you go to everyone else. But within that big majority of the pie chart, you've got a huge amount of variety. There's a load of different capabilities. You've got different screen sizes and shapes. You've got different keypad layouts. Some of them don't have keypads. Some of them have got GPS, some have cameras. You've got different types of network, different operating systems. There's a huge amount of variance within that. And when you're making products for mobile devices, you have to account for this variance. So you can't target that 95% as though it's a single device. You can't even just target Nokia handsets, just say, well, we'll handle Nokia's because they've got three or four different families of devices and within those families, there's lots of differences. So if you dig into these figures a little bit more, so this is the depressing slide now. If you want to reach 70% of UK mobile owners in January, you have to be on 375 different devices. Well, okay. What you could do is you could take a lowest common denominator approach, just do something which you know every phone can handle, and that's probably text messaging or voice. And there's lots of interesting things you can do with text messaging. It's still underexploited as a medium because that's not what my business is about. I'm not going to talk to you about that today. So okay, so if you say to someone, particularly five years ago, if you said to someone, well, how many phones do you want your application or service to be on? They'd say, well, all of them, of course. If you say, well, that's prohibitively expensive. 70%. 70% is a nice reasonable sounding number. No, that's most people. Yeah, that sounds okay. But as you can see, a vast number of devices. And that's eight different manufacturers up there. Now, in practice, actually, it's not quite that bad because a lot of mobile phones only differ very, very slightly. For instance, we've got a really quite depressing habit in the industry of making mobile phones for women by taking an existing mobile phone and making it pink. Even more depressingly, that works really well. But okay, if you just combine all of these, actually, they're kind of the same phone, but in different cases, into what we call families, you've still got 70 of those families. That's still 70 different devices that you really ought to buy them in and test on them and all that sort of stuff. If they're 100 quid each on eBay, and we spend a lot of time on eBay, that's seven grand, and people don't have a seven grand budget when you're trying to do a mobile project, that's ridiculous. And even if it takes an hour to do that, that's two weeks of someone just testing it, nothing but testing, and it often takes more than an hour. Okay, so you think, well, let's just take the best. Let's just get the top device, the few devices that give us the best reach. Well, the top one on our list gives you about 4%, and 4% of an addressable audience is nothing, whatever you're doing. If you take all of the iPhones together and match them up, it's 3.6%. So even there, it's not all that much. And when you're down near the bottom of that list, these things are distributed on power law curves, so the top few devices are really, really popular, and then you've got that sort of long tail. Now, as you go down that tail, you find yourself spending an incremental amount of effort, maybe an hour of testing and buying in the device, to do something that gets you a third of a percent of your audience. I mean, at that point, you're sort of doing it, thinking, well, we need to get to this magic 70% number, but is this really worthwhile commercially? What's the return going to be for the effort you're putting in there? There's a silver lining with all this, and that's that iPhones taught us as an industry a really, really good lesson. And it's taught us that you can reach, and it's okay to reach a small and very, very active audience by very, very tightly targeting them, and you can maximise a commercial return there. Now, most of my customers are interested in selling things to people, apps and services. So for them, most of the time, there are a few exceptions, and we'll talk about one, but most of the time customers are interested in just maximising the revenue that they get back in, it's not so much about reach. For educational organisations, I think it might well be different, and for some of our customers, like the BBC, they have a mandate to reach a wide audience, but most of them, it's okay to go to a very, very small audience and just get what you can from there. And also, we should bear in mind that there are lots of companies out there, mine included, but there are certainly plenty of others who have, over the last 10 years, gotten a lot better at dealing with this problem, have evolved strategies and technologies and approaches that help mitigate it to a great degree. So, there's a big hairy problem there around fragmentation and mobile devices. We've also got some problems as an industry around tariffs and billing. We've got two sorts of customers, pay upfront or pay monthly ones, and it's a cliché that the pay-as-you-go people are all the poor people that don't use data services. It's not quite true, but tariffs are extremely confusing, aren't they? I don't understand what tariff I'm on, particularly. You've got what, in minutes, inclusive minutes, roaming minutes, text messages, bolt-ons, text messages that are outside of your bolt-ons, flat rate data, data by the minute, data by the megabyte, and some people claim that these are deliberately obfuscated and made confusing so you can't really make a rational choice between them. And all that leads to a quite well-known phenomenon in the industry of bill shock. Anyone heard of that one? Bill shock? It's where you get a new mobile phone and it does lots of whizzy stuff and you go, oh, this is great, and you're sort of wandering around using the maps all the time, showing your friends going, oh, that's fantastic, look at that. Oh, I can download games, all that sort of thing. And then you get your bill at the end of the first month and you get that 300-pound bill and you never do any of that fancy stuff again on a mobile phone for the rest of your life. And it's something that happened. I mean, look, iPhone, one of the really fantastic things about it, ignore the device completely. You can't get an iPhone without getting a flat rate data tariff, so you're protected from all that sort of stuff. And just Apple forcing that through for the industry was in itself incredibly valuable move. And mobile billing in general. Mobile billing's kind of a bit strange because it's about the most convenient form of billing there is. Sending a text message or having a text message sent to you and then either comes off your credit or on your bill at the end of the month, you can't get much more convenient than that even with sort of credit cards. But at the same time, it's got a really, really bad reputation thanks to some nasty sort of fairly shoddy scandals around TV voting and SMS and some things like the crazy frog, which you're probably still trying to forget, all that sort of stuff, that kind of thing. So it's got a really poor reputation with customers and they're kind of a little bit worried about it, but it's still incredibly convenient and it's kind of there, there to be used. So that's the world that I've been living in for the last 10 years. My name's Tom Kim. I'm the Managing Director of Future Platforms and we design, build and launch products for mobile phones. We do stuff with a particular emphasis on collaboration, participation, playfulness. And some of what we do is educational. We're working with a couple of divisions of the BBC right now, Bitesize, and I think Raw, I think they're now called SkillsWires or something like that. But a lot of what we do is not education in any way at all. We help our customers kind of deal with a lot of these problems around fragmentation and ensuring that your end users don't have bill shock or that sort of stuff. And we've been doing that for about a decade, which means that we've been through a lot, lots and lots of changes in the industry. When we started out, we were doing apps for palm pilots and wapsites and no one talks about either of those things anymore. We've made it all for a lot of mistakes over the last 10 years, most of which we've learned from. And we've learned a lot about how to do this stuff properly. So what I'm going to talk about now is a little bit of our approach to, yeah, there are some really badly Photoshop people in that picture. We did it for the local paper and a couple of the guys weren't in work that day, so we Photoshopped them in. And then you get a call from a journalist saying, you know that newspapers get in trouble for this sort of thing, don't you? So I'm going to talk a little bit about how we approach building software for mobile phones and show you a couple of examples of the sorts of things that we've done that sort of give you a feel for that. So it's not enough to build products the right way. You have to make sure that you're actually building the right thing in the first place. And this is where we see design coming in and a well-honed design process. And Apple have done a great job of demonstrating the value that design can bring to products. Just after the iPhone launched, you couldn't sit in a meeting with someone from a network operator without someone standing up and saying user experience is key as if that was some sort of new discovery that Apple had dug out from behind the curtain a few months before. I mean, when I first started working in mobile in 1999, one of the stats I remember being bandied around even back then was that Nokia phones were really, really popular with operators because people that own Nokia phones sent twice as many text messages because it was just easier to do it on those devices than it was on a Sony Ericsson or Ericsson that was back then, all that sort of stuff. So, ease of use and design has always been important for the industry. It's just getting a lot more press these days. And we tend to start our design process by getting everyone together physically in the same room. And that's stakeholders from our customers, all the different members of our team, from designers, developers who know a lot about the constraints and opportunities of what can be done, testers who are the poor guys who have to sort of sit there at the end of the day going, no, it doesn't work, no, it doesn't work and understand a great deal about the pain points of mobile apps and services. We get all of these folks in and we'll typically run a day-long workshop where we talk about business objectives where we look at the audience that we're designing for, we look at any research we've got, go through a fairly structured format and we come out at the end of that with an idea of what the scope of the project is going to be, how we're going to approach it, priorities, because you always end up with a shopping list that's way bigger than what you can actually afford in these sorts of things, all that sort of stuff. We then co-locate our designers with the teams that are actually building stuff. So again, you've got just constant feedback all the way through the process between the two and so it's not that you're designing a beautiful product on paper and then you take it over to the guys who are going to build it, you look at it and go, that's not going to work. You avoid that sort of stuff and we involve our customers very, very closely as well. We tend to get them, we tend to actually insist on having them down at least every couple of weeks to actually see demonstrations of the products as it goes on. There was a quote earlier about IT projects used to take three years and now it's got to be weeks. Yeah, absolutely. And so you need to be getting regular feedback all the way through the process from the very, very beginning to do this right. We tend to avoid doing anything digital for a little while as well. We do a lot of work with sketches and paper and index cards and that sort of stuff. So we'll mock up, you can see the sort of, very, very vague notions of what a user interface for a product might be. And we can then use those mock ups in testing processes, put it in the front of users, theorise about them, all that sort of stuff. The really nice thing about pen and paper is that if you've just spent a few seconds jotting down an idea and it's a rubbish idea, you throw it away. If you spent a day designing something pixel-perfect in a graphics package, you're much less likely to throw it away even if it's back in your mind, you're thinking, oh, that's not so good. Because you've invested a lot more time and effort in it. And in those early stages, you want to do a lot of investigation and throwing out the bad stuff. So we tend to be quite light on documentation and heavy on conversation. Lots of time spent around whiteboards, photos, video, that sort of stuff. And we also make a point of trying to get stuff out there as soon as possible. If some of the stuff is probably familiar if you guys have come across agile methodologies or Scrum or anything like that. But we make a habit of getting products into end users' hands as quickly as we can, testing. At the beginning of a project is when you know least about what it is that you're trying to build. And you gradually learn more and more and more. And the cone of uncertainty, as they call it, sort of narrows towards the end. So you want to get stuff out into people's hands as quickly as you can so that you can start learning as quick as you can. So a couple of little case studies. Anyone seen these before? You're probably familiar with these if you sort of grew up in the UK in the last 30 years. This company, they're called Puzzler Media. They're the UK market leader in puzzle magazines. They're about 30 years old now, 35 years old, I think. They sell about one and a half million magazines a month, something like that. It's a big business for such a crappy logo, isn't it? They're really, really 70s. I mean, the cover's classic, sort of Scandinavian model, barn owl sort of thing. I've been told that they tried changing them in sales drops so they just went back. They really, really know their audience quite well. It's a sort of an endearing naffness. They approached us about five years ago. They were going through a management buyout and they were looking at all sorts of different opportunities for their business. Interactive TV, mobile, online, all sorts of stuff. I've got a great little print business who want to do more with it. And they made a really quite strategic investment in mobile in that they didn't just dip their toes into the water and try it out. They sort of really, really went for it. And they've seen some good returns there. So it turns out puzzles are really, really great mobile content. They're small, they're really familiar. Anyone not know what a crossword is? No. They've got mass appeal and you need to refresh them regularly. These little chunks of content that deliver a little bit of fun and then you have to get a new one. Interesting. So we went through the classic sort of design phase I just talked about. We actually watched people playing crosswords in Sudoku for hours and hours. Everyone plays crossword the same way. Everyone gets a few clues then joins up the things on the grid. Sudoku, certainly a few years ago, everyone played it differently because it was new. People writing outside the grid, drawing to the sides of it, drawing inside the grid, all sorts of different things. And that completely affected the way that we produced these things for phones. And we did some prototyping, tested these prototypes out real users, launched Sudoku. What they did, they made a kind of strange decision early on or an unusual decision early on. Most people, when they're doing mobile apps, they'll produce, make mobile apps, make the files, give them someone else to sell and say, just give me a check at the end of the month. There you go, orange. Take these and see what you can sell. Put them on your portal, whatever. Puzzle did something a little bit different. They decided to own their own distribution. And what this meant was that whenever anyone buys a puzzle, a little signal comes into them and they package it up for that customer and send off the file or that sort of stuff. It's more of an investment for them to do that up front, but it gave them a lot of analytics and useful information. And they were doing something completely new that no one had done before, so they needed that info. So which means we get sort of interesting things like these little graphs out the back. So we were hearing earlier about the most popular time to go online, being about sort of 11 o'clock at night. Well, that's pretty well what we saw. When we first launched this service, we thought our target audience was going to be commuters on the train up to work. Get on the train, text off, get your puzzle, play it for 20 minutes. Seem to make sense. Puzzle took out a load of advertising in the national press next to Crosswords and Sudoku, thinking, you know, well, that's where our audience are. Absolute rubbish. People puzzled mainly, well, that might be on the way home, sort of 6.30, 7 o'clock, and then 11 o'clock at night. And you can see when they all have dinner, I quite like that. Completely changed the way that they approached selling their puzzles and they wouldn't have known this at all if they'd just shifted the files off to someone else to sell for them. Similarly, all the operators in the UK, certainly the big for that then, kind of had roughly the same sorts of audience numbers, so we thought we'd see sort of equivalent traffic from all of them. The blue line is O2. We saw about double the traffic from O2 than we did from anywhere else. We think that was because they spent a lot of time back then educating their customers about mobile gaming. So who do you think puzzler went to to do the first portal deal with? Clearly O2, because they could drive the most traffic. Again, they wouldn't have known that. It made a big difference to their business. They've changed their models loads over the years. They've gone from doing direct to consumer stuff themselves, from doing press ads. They did some co-branding deals with ITV and the Daily Mail, where they'd been using daily mail branded puzzles and sold those and they wouldn't really bought them, so they stopped doing that. They did lots of one-off purchases. They moved to a subscription model. Get your sign up and get a puzzle every day. And that's worked really, really well for them. And what's also worked really, really well, despite what you might hear in the press, is deals with operators. So they've been on the operator portals of T-Mobile and O2. They're launching on to another one any second now. And that's driven hideous amounts of traffic. Now, those guys have been incredibly painful to work with. It can take months and months of commercial meetings and integration work and all that sort of stuff. But once you're on there, they're delivering really significant traffic for these guys anyway. They're also quite unusual in mobile gaming. They've got a very, very low price point. I think 50p of time. Most mobile gaming companies are trying to sell their puzzles for as much as they can each download. These guys are trying to build a long-term relationship with their customers. Completely different model. And the results now, they're doing... I can't give you the exact figure, but they're doing well in excess of 100,000 paid downloads a month. And I don't mean 101,000 or 105,000. I mean well in excess of that. And it's a kind of equivalent sort of commercial scale for them as a magazine title is another way of putting it. It's kind of like they're adding a magazine to their sort of stabila titles. And they're just launching another one right now. So that's an example of someone that's spent a long time working in mobile to reach a really wide audience. This isn't iPhone and Android. This is normal phones. Sony X and Nokia's, all those sorts of things. So the next project, it's a bit more education. We've done a few education things over the years. And this is more about the design process. It's quite an unusual one. Channel 4 learning have a mandate to educate 15 to 19-year-olds. But the slots that they've set aside for this audience aren't the sort of slots that 15 to 19-year-olds are really watching TV on. Sort of late morning kind of thing. So they realised this and they thought, well okay, where are our audience then? The people that we're trying to reach to teach about life skills. Well they're online and they're playing games and they're on their mobiles. Okay, well we'll move the budgets over there. And this is a piece of concept work that they commissioned from us. It's not gone as far as production, I have to say, but it illustrates some aspects of the design process. And the inspiration from it was a South African experiment or South African sort of teaching method where they get kids in the classroom and they give them all a test tube of clear liquid. And the kids sort of socialise throughout the day and whenever they've socialised with someone, they sort of exchange a bit of fluid in the test tubes. And at the end of the day, they put this marker die into the fluid. And it shows that where a couple of the test tubes were contaminated at the beginning of the day, now all of them are. And it's a sort of analogy for the network effects and the spread of STDs, that sort of thing. And so we wanted to use sort of simple two-player games to educate teenagers about the risks of STDs in a kind of way that isn't preachy. So you play a game with someone, it's quite fun, you play over Bluetooth, so you're sort of nearby, it's quite fun, but you run a risk of catching something nasty. The analogy is quite obvious. Most of our team are quite a long way away from being teenagers, unfortunately. So we started out by looking at the available research and Channel 4 commissioned a hell of a lot of research to assist with this sort of stuff because they're doing a lot of projects in this area. And identifying who our audience were. So we crystallised the audience and the research into personas. And I think there were five of them and these are four of them. And we refer back to these individuals who don't actually exist throughout the design process. There's a terrible temptation. If you start thinking of the person you're designing for as the user, what does that mean? There's nothing there that you can really hang sort of thoughts or ideas on. If you actually personify them and then refer back to that, would Charlene use her mobile phone in this sort of way? You can ask yourself those questions and get almost secure in answers even if they don't actually exist. So it's a technique that we use quite a lot in our design process. And then we went through, I mentioned earlier to do some sketching. And these are some of the screens from the game. Just sketch them out on index cards. So we've got a concept and an audience. We've got the cards written out and we've got a flow and a structure for the game. And it's very, very quick. It's very easy, it's very physical. And once we had these, we could then do our first testing. So what we did was we went into a local school and we sat down with a group of teenagers and we talked through the product with them. You know, saying to me, what would you think would happen if you pressed this button? No, I think I'd then get a screen that takes me through to show me the game. Okay, yeah, that's what we've got here. Sort of checking our assumptions and validating them and learning a little bit about what they thought about the products in the process. And, you know, we learned quite a lot. We hadn't anticipated the need for privacy. They wanted to play this game quietly, particularly this game, this sort of game. And if they caught a disease in the course of playing the game, they wanted to be able to read up about it without anyone knowing. And that was kind of interesting because we thought that we would avoid having too much information actually within the game. We didn't want to come across too preachy. We didn't want it to be a, you know, play the safe sex game on your phone kit. But it turns out a lot of them wanted a lot more information about the various diseases in the product, you know. So later designs put that back in there. And anyone who's got teenage children, cover your eyes now. You won't like this one. We found that we had one little piece of feedback which might worry you. So perhaps designing sex education games for teenagers is a bit of an extreme case. But we do find this with much more everyday products as well like Crosswords and Sudoku. We and our clients, we're not the end users of the things that we design. And our collective understanding of these end users often falls well short. So, you know, if we do a great job of building the wrong thing, then we've failed. And this sort of approach makes sure that we build the right thing. Or at least we find out as early as possible that we were going to do the wrong thing and can it quickly then before it's gone into any sort of expensive production process. I've been asked to finish with a few little predictions. I think one of them in particular has kind of been covered off already. But here's what I think we're going to see over the next few years. Five years ago, I would regularly meet people and I'd explained to them about this fragmentation thing about all different phones they've got to cover. They'd say, oh, that's awful. I'm going to come back to mobile in a few years when Apple and Google have sorted it all out for me. And it's not getting better, it's getting worse. Samsung have launched their own platform, Bada. You've got Nokia and Intel getting together with their Mego stuff. Nokia have launched Qt, which is another sort of UI framework. You've got Hewdet Packard buying palm. The Apple platform is a bit fragmented. Some of their mobile phones aren't even phones. The iPod Touch, for instance. If that's not fragmentation, what is? All this sort of stuff. It's getting worse. And it's just one of those things about the industry that's a bit different from the web and everywhere else. I grew up working on the web. I remember what a pain it was to do, oh, two versions of Netscape and two versions of Internet Explorer for that website. And now, you know, we're launching products that we've actually tested on 300 different phones. It's just one of the things that's different. The web promises to sort that out for us. And, you know, it's looking good there. If you're doing stuff today for a wide audience, it's not an option right now. I think we're going to see simpler cheaper tariffs, particularly on data. We're seeing lots and lots of folks getting very, very interested in using data. And the minute you go flat rate, use a lot more of it. That creates problems for operators in terms of handling capacity and all that sort of stuff. But there's opportunities there as well for all sorts of different ways of selling it. You look at some of the very innovative things that Vodafone have done with tariffs. Some of the sort of interesting free-roaming stuff or their day-long promotions, which I think they've done on Valentine's Day, where I'm encouraging people to buy a day's worth of data. There's lots of interesting stuff to be done there. And again, back at Apple. No one's really noticing the sort of tariffing stuff they're doing with the iPad. There's sort of no loyalty to an operator using a special sort of SIM so you can't use it never in any other sort of device and you're paying £25 a month or something for data. Well, that's interesting, isn't it? That's adding to someone's monthly spend for data for a device that they're predominantly used at home or in the office, so it won't use the networks very much. You can see that there's lots more work to be done there in that kind of area. I think we'll see tariffs which are simpler to understand in particular for end users. And the last one I think has been covered off to death almost today already, commoditisation of access, the idea that people will get online not using the kit that you give them that you have to fund, but using their own kit, using their own mobile phones, their own laptops, all that sort of stuff. And that, I'm guessing, changes the economics of some things for the education industry. So, thanks very much for staying away through that. Cheers. Thanks, Tom, that was excellent. Stay there because there might be some questions. We've got a few minutes for questions, so anyone got a question over here? Just hang on a moment for the mic. Kelvin Gann from the University of Bath. I was talking to you earlier during lunch. You were talking about fragmentation. Do you see HTML5 as a solution to that? Why, yes. HTML5 is currently the best way that we've found of doing apps that work across Android and iPhone. But I think it's the only way of doing them unless you buy an expensive commercial product from someone. So it's been really, really good for that sort of stuff. You can't do everything you might want to in an app. I'll give you an example. We've just launched something recently, an Android reader for the Guardian newspaper. It wakes up in the middle of the night and grabs your newspaper. So it's there on your phone for you in the morning. So when you're on the way to work, it doesn't matter if you're on the tube and you've got no coverage. You can still read the paper. Couldn't do that sort of stuff with HTML5. Some of the gaming apps and gaming is tremendously popular on iPhone and Android. You couldn't do those with HTML5. But for the stuff you can do, it's great. I've just remembered the author of that book, I recommend it, Mark Pilgrim's Dive into HTML5. Brian. Sorry. Flipped side to that question. Is it now time to be getting rid of our Flash developers? Steve says yes. So is Flash or should Flash be on the way out? Should we be looking at migration, not just on a mobile environment with the iPhone issues but on the desktop as well? I guess it depends what you're using it for. I mean, Flash has never really delivered a large audience on mobile. I mean, compared to the overall size of the mobile. Mobile is great because you can talk about how you've got millions of users. But that's actually a tiny, tiny percentage. So you get big numbers banded around all the time. Flash has never delivered an audience that's worth addressing, in my opinion, on mobile, and despite repeatedly promising to. So it's not should it go away, it's never been there. And I'm sure they'd like to change that. On the desktop, I couldn't really comment. There's all sorts of different uses for it. Video, online games, mini-clips, sites like that. I couldn't imagine them existing at the moment without Flash. But it depends. One there. I don't think he's ever going to get to the scale, obviously, of the phone platform. But what do you think the timescale is going to be as far as tablet PCs and the arrival of the iPad and Android are going to come out with, you know, alternative products and what have you? Where can we see that sort of splintering it even further? So sorry, you're asking about the timeframe before these things are coming out? Yeah, I mean, how long do you think it's going to take from sort of now to a sort of saturated marketplace of similar products and similar platforms, but then you've got to rescale up that you can't just use the same apps on a phone, you've got to do tablet versions and what have you? Well, I think people are getting smarter about designing things that scale. If you look at a lot of the Apple guidelines for interfaces, they actually make sense of the sort of tablet scale as well. There are a few metaphors that kind of change between the two, like the multi-paying stuff you get in mail, but actually a lot of the apps scale up reasonably well. As far as when we'd actually reach saturation, well, okay, what's the replacement cycle for a netbook? I'll probably look at how long it is between people replacing netbooks because that's where tablets are really, really competing. And also probably for folks that are kind of less involved with technology, like my parents are ideal iPad or I shouldn't say iPad, ideal tablet users of some sort. They've got no real need for a keyboard or anything like that. So I'd say, I don't know, significant numbers within say five years. I'm a big believer in the format. I'm more ambivalent about which particular device might win. Google will be having a Chrome tablet out quite soon, I'll be full. Okay, last question. Yeah, sorry, I don't want to be this to death, but I'm intrigued by the fragmentation point. I mean, you know, I've tried myself and I congratulate you on staying afloat for 10 years in this world. How, I mean, is Java a factor in your lives? Or do you write apps for the native operating systems? The puzzler example I gave, that's all Java. Yeah, okay. I mean, what we do has changed a few times over the years. I mean, most of what we're approaching to do right now today is iPhone, iPad and Android. But over the last five years, most of our work has been Java. Okay, you don't write Symbian apps or anything. We did a little bit once, but the audience for them is not so big for us. There aren't that many people commissioning Symbian apps, so we haven't made much of a business there. Okay, thank you. Okay, can we just say thank you to Tom again?