 Welcome to ITU Telecom World 2017 from Busan in the Republic of Korea. I'm very pleased to be joining the studio today by Dean Bubley who's the founder and director of disruptive analysis. Dean, welcome to the studio. Thank you. And I'd like to start off by talking about digital smart transformation. It's the theme here at ITU Telecom World. What does it mean to you if anything? Right. I have to say I'm not a big fan of some of the buzzwords. So you won't usually hear me talking about the word digital and you won't hear me using the word transformation but I'm managing a panel to, moderating a panel tomorrow on reinventing telcos, which I think is a bit more descriptive. And I think that it's a, partly it's a technology thing and I think we'll talk about that in a minute, whether it's 5G or fiber or whatever. But it's also a cultural and process and organizational thing. And I think that one of the issues I see in the telecoms industry is getting around the move from being a sort of a traditional business that sells phone calls and access to one that is more like a software company where products are developed, they're conceived, they're prototyped, they iterate and they change over time. Sometimes they fail, sometimes they get replaced. Rather than the sort of top-down view that the telecoms world traditionally had is like here are some standards and one or two services, now everyone go and sell them. This is more like the other way around of see what works, see what customers, whether it's consumers or businesses want, improve them, the products, occasionally they'll have issues and eventually they may become standardized. So it's really transforming from top down to bottom up. Now you're moderating, as you said, you're moderating a session this week on reinventing telcos. Why have telcos in your opinion taken so long to react? I think there's a couple of things. One is they are institutionally hamstrung by some of the ways in which they are both set up, they're existing, some investments. In some cases they have a profile of employees that is geared towards engineers and lawyers, should we say, rather than designers and marketers and mavericks. And so I think there needs to be a sort of a bit of injection of risk taking in some cases, but also a new set of skills and new set of people that are more comfortable with less analysis and less defined propositions upfront and also perhaps different attitudes to their network access versus their services business. The other issue is, and I'm going to be probably quite controversial here, I think that some of the industry organizations need to change as well. And so I think the structure of processes of whether it's ITU or GSMA or 3GPP, while they're changing, I'm not sure they're changing fast enough. And I think that having one year or multi-year cycles for things in a world where products change on a weekly basis obviously brings challenges. The other side to this that I think that is the root cause of some of the problems to telcos is they talk about innovation, but they don't do any R&D. I think the operating with the largest R&D budget is NTT in Japan, which has 3 billion dollars. And then there's a number of other large operators that do R&D at decent levels, whether it's AT&T or BT or Orange or SKT here in Korea. But you compare that with the budgets that are put either by the big network vendors, as with Huawei recently, I think it's 10 billion dollars, 10 or 11 billion dollars. Samsung I think is 13 and sort of all Apple, Facebook, Google, they all do 5, 6, 8, 10 billion dollar R&D. So that means that any one of those does more R&D than all of the operators on the planet put together. And it's unsurprising then that where the innovation and where the new services and revenue comes from. And I think that that mindset of putting money into research and creating things needs to change rather than just relying on a couple of standards bodies or author vendors to invent new stuff for them. And in terms of 5G, obviously, Internet of Things and AI, how are these going to impact on telcos? Well, I could probably talk to a day on each of those, but I think that the interesting question is how they impact on society and how they affect telcos. IoT is the trickiest one in a way because I think it's going to be massively important, but I think a lot of the implementations will have relatively little to do with the telecoms industry because they will be done by industrial companies. A telecom company isn't going to be hosting the software to stop robots colliding inside a car manufacturing plant. Important or similarly to sort of do predictive maintenance on elevators. So I think that it's going to be quite hard for operators to get out of the connectivity part of the puzzle, which is a fairly small percentage, unless they develop deep domain expertise. And what I expect to see is operators have maybe one, two, three, four areas where they put money into, so you might find Telstra with mining or Orange with healthcare or Vodafone with vehicles or something like that. But they're not going to be able to cover everything. 5G, 5G is clearly having a huge effect on the general consciousness of mobile broadband and also IoT connectivity. I have some different views on 5G to a lot of people. I think that we're going to start seeing it initially for fixed access, probably here and also in the US and a couple of other places for the last 1, 200 meters from a fiber node. I think that we will then see it in sort of hot spots in urban areas for sort of more faster, better mobile broadband for probably smartphones predominantly. And that won't be different because you have small islands of 5G in an ocean of 4G. So there's no such thing there as a 5G business model. It doesn't make sense. I think the other use cases of around IoT, whether that's massive IoT or low latency, I think they are currently being overhyped and I think they will be later to market than people expect. I think the massive IoT, frankly, operators are only just rolling out NB IoT, which is the 4G version. So they're not going to rip it out two years later with a 5G one. So that's going to be pushed back. The other side, although we might get sort of 4.5, 4.9G and bits and pieces of them. The ultra low latency stuff is really interesting, whether that's for drones or headsets or AR headsets or something. The problem there is the commercials. Let's say it's be generous and say 1% of N devices need that low latency. Is that enough to essentially put the compromises on everything else at cost and complexity? Unless you're going to charge $1,000 a month ARPU for a VR headset, I don't know if the numbers stack up. I think there will be some important point use cases, so you can mention for public safety or for maybe private. I think there's a really interesting trend here about telcos or is it private 4G and 5G networks where you could perhaps have a private network on a manufacturing plant. So I think that's a possibility. Was it on AI? AI again has many, many touch points for telcos. Whether that's in terms of the internal operations, running the network, optimizing paths through the network or maintenance or whatever happening or planning of upgrades. There's some interesting stuff around the customer interface. Perhaps looking to make it easier for people to do customer service and either self-service or have a bot of some sort. And then there's the sort of the AI which relates to extracting data from perhaps unexpected sources. I saw an interesting thing yesterday from KT saying that they've realized that some of the patterns they see in the light on their optical fibres could indicate give early warning of earthquakes or other national, because the actual, the way it diffracts inside the fibre. And that's a really sort of off-beam use case because the sort of thing that if you have AI that's looking for anomalies to normal behavior, you might throw things up like that. You also have the poacher as well as the gamekeeper and the risk here is that the end-users AI might be smarter than the networks AI. So if my AI is trying to manage the amount I spend, it sort of pre-caches data over Wi-Fi when it knows it's cheap. It's sort of even frankly, negotiates my tariff on my behalf. As a customer, maybe I have an artificial intelligent mobile purchasing bot of my own, which actually drives a harder bargain than I could. Absolutely. We all need those, I think. And just really, I mean, in terms of the, what have you seen and what have you heard here at ITU, Telecom World that's particularly simulated your structure and imagination. Okay. I was on a panel or I was in a panel session yesterday that I was listening to on what seems like a quite dry topic, but is actually important, which is on spectrum management. And there seem to be, if not complete consensus, but at least grudging acceptance or enthusiasm for shared spectrum. And normally, you either have wireless bands which are dedicated, which is, for example, for cellular, or for satellite or broadcast, or you have unlicensed such as Wi-Fi. And there's an interesting middle ground because increasingly it's getting very difficult to offer new slices of dedicated spectrum for mobile broadband because pretty much every bit of spectrum there's something there already, even if it's not used very much. And to actually clear and replace, you know, even an isolated example of, I don't know, point-to-point microwave is actually getting very complicated. So there's new ways of dividing up spectrum so that that existing user, which could be the military for naval radar or something like that, which isn't going to move, to leave them protected, but other people can use the spectrum either in places or at times when it's free. And I think we're moving towards a world where that is going to be used for mobile broadband. It may be used to build on-premise enterprise networks using wireless for IoT. And I thought the panel was more positive towards that than I might have expected at this event. Now it's been a few years since you've been here but I know that it's obviously you taking the time out of your busy schedule to be here. I just wanted to find out why is this event important and what makes it stand out in your calendar? Well, firstly, it's obviously a slightly different audience to some of the events I go to. And I do everything from deep technical events to policy events. And this event is you do have a fairly sort of government and policy slant to it. But also it's interesting because particularly this being in South Korea, there are out on the show floor an awful lot of interesting startups. And one of the things I want to try and do today is grab an hour to wander around and sort of look at it. I mean I literally have seen everything from virtual reality headsets to a toy dolls which are actually robots for kids to navigation devices for container ships. I probably want to sort of drill into some of these a bit more. I think the sessions are interesting as well because they they are more international than some events. And so you do have a lot of people from parts of the world with different attitudes to telecoms and also to policy and governance. And sometimes it throws up some of the tensions as well as some of the agreements. And I think it's important to understand where that is because I think that's that's an issue for network developers and also analysts and consultants such as myself. Dean Bobby, thank you very much for being with us in the studio today and wish you the very best for the rest of the event and for the future. Great. Thank you. Thank you.