 Good morning, everyone. Thank you, Amy. I resonate a lot with a lot of your observations. What I'd like to do today is really talk in the broadest sense of the digital transformation that Vodafone's undergoing. Because in a lot of ways, the network transformation is just a part of that. It's fundamentally altering from the hardware constructs that we've been to a software and eventually to a data construct in the future. So as we work for our customers and really figure out, how do we deliver the world's best gigabit network in a digital-first environment and really looking quite constructively at our past and our future and saying, it must be much simpler. And as I talk through today, we'll highlight a few areas where we've proactively and fairly aggressively gone after simplicity really just because we recognize that our customers increasingly declare that they will happily swap out speed and immediacy for the complexity and robustness that we might have prioritized in past times. To embrace not just a 5G transformation, but everything that's happening in the digitalization of the world and bringing everything online, we continue like everyone else in this room to be faced with, how do you drive the capacities and scales that actually cause us to implement our entire capacity over and over again year after year again to keep up with scale growth, but actually do so and crush the cost on a per unit of bandwidth so that we can continue to bring on the solutions that our customers need. And ultimately, the most important thing is the transformation to all digital so that everything is available in our customers' hands the moment they want it and in the future before they realize that they want it is really what we're trying to accomplish going forward so that our first focus is the customer experience. Now I'd like to highlight just a few of the things and where we stand today and then talk about how we've gotten there. We don't often come out and share kind of where we are in the journey but I'd like to take just a couple of moments to reflect on what we've done. I think one of the things that we targeted two or three years ago as we said to make this happen it's actually quite straightforward to do this on an instance by instance basis but to move the entirety of our customer base onto a software defined software controlled infrastructure is actually much harder and the way to do that is not by going after the niche use cases but to actually go after our primary businesses. So we put an early on focus of innovating in our services business as we built our new TV platform as we scaled our IoT platform as we transformed to Volte as we built 5G all of that has been migrating customers to the virtualized infrastructure. We would have migrated to cloud but as you know we're not there yet and this talk will cover just a bit of that but ultimately this is an important step to it and we'll talk just a bit more about it. We have built and really hats off to my operations and delivery teams. They've built in partnership with our vendors a huge infrastructure where we're now in excess of 30% of our network functions industrialized into the cloud. By the end of our fiscal year we'll be in excess of 50% and these are the crown jewels of the operations at Vodafone, the packet core, the voice core, et cetera. In one of our largest countries we actually have implemented in our passing traffic at capacities in excess of six terabits across virtualized infrastructure which is just more than twice what we ever passed on our traditional and it continues to hum along as well and we do it in a unified fashion such that we serve all customers seamlessly on that hybrid world because it's never the case that you get to build Greenfield when you have scale like ours, okay. And then as I said before, 5G non standalone as we've launched this summer across six countries has all been launched on virtualized infrastructure. As we move forward into 5G standalone in the near future we're expecting and counting our vendors to have built that in cloud native and so we've got a lot of work to do to make sure that our infrastructure scales the physical, the virtual and the cloud in a consistent way, okay. And that the automation that we build actually automates end to end across that because our customers don't want to know and we don't want them to know about the underlying piping, okay. So that's where we are today but how we got there has been a bit of a journey and I'd like to spend a little bit of time because it calls out that we're actually going to insist on what I say we need to do in the future, okay. So we started like everyone else in the room probably with we have the luxury of experimenting and inventing across numerous countries and so what we did is we actually had different interests and priorities in each country driving different virtualization strategies but what that led us to is that led us actually validate for almost every network function that we could virtualize it, develop the operating processes and really confirm that that strategy would work end to end, okay. Then we said wait a minute doing these siloed implementations is not going to take us at all where we want to it's not going in fact it's probably going to add more cost, we'll add more complexity. So that's when we backed up and we said we need to simplify quite significantly and so we went from a very multi-vendor centric to a much more homogenized domain specific area where we went from best of breed to best of suite and we would actually build aggregates especially around our big complex VNFs of the IMS and of the APC and as we went then to build scale we industrialized working predominantly with VMware we industrialized the VIM, the infrastructure management and then we put in place the best of suite solutions around packet core and voice core that led us build scale and as I like to think of it that secured the business case, right? That led us get more savings but more importantly that led us finally build scale operations to really drive the transformation without holding ourselves hostage to the world that we talked about now of automation and beyond, okay? So where we stand now is with the majority of our well with nearly the majority of our traffic in we're now at a place where we need where we've got the scale that demands and requires that we now automate, okay? And so we're very much at the part of our journey that Amy talked about in terms of moving to true automation where we actually have the scale in scale out that we've always dreamed about but more importantly we have that working in a cohesive fashion and a service creating fashion that lets us innovate and drive more services to our customers in a faster way. Streamlining because we've chosen to simplify, okay? And there are some really important things that we need to get to. Our next steps, what that then will situate us for is to be able to, now that we have automation, telemetry and most importantly, control, that then lets us bring in some of the work that we've been doing in our digital arena and in our more traditional ops teams of bringing in the machine learning, the automation as well as the predictive to be able to finally build the automation side of it. To move us to a learning approach and an insights driven approach to how we scale our operation, serve our customers and so forth. So that's where we are with our next steps coming then in after the automation folding in the learning. Now we've built also in parallel quite a substantial digital capability in terms of how we serve our customers. A lot of it was achieved by moving from our internal private cloud into the public cloud for a lot of our applications and we were in fact one of, if not the only carrier that actually went through the iPhone launch and continued to keep up with scale in this last iteration because we've relied on and built on the solutions available to us in the public cloud. Because we've embraced the DevOps tooling we've now been able to automate and we're releasing fixes, improvements and accelerators into our code base on a daily basis. Now these are things that we've seen our competitors do for some time but it's quite remarkable to finally see our own developers taking these controls in hand and being able to tailor the services that we deliver to our customers. And we're beginning to see some really compelling use cases of integrating learning and analytics. Most of our customer offers today are now made not by our marketeers but by the customization of our machine learning on top of the best capabilities of our marketeers. And then lastly, we've built reuse of our chatbot applications and others by finally standardizing on our digital experience library which is our API layer for our digital and that's really giving us confidence to push for the simplification and streamlining through adoption of common APIs into the network space as well. So these are some of the proof points that actually have seemed obvious for quite some time but for whatever reason we've resisted. And now we begin to see actually it's driving huge change and big improvements in terms of how our customers work with and secure services from Vodafone that gives us confidence now to push into how do we address the critical challenges that we face? Because one of the hardest things that we are confronted with today is while we've industrialized and are working in scaling it is still the case that if you look at how we've onboarded our VNFs it still looks too much like an artisan activity and not an industrial activity, okay? It is still too much the case as you would expect of us we require that each of our VNF providers commit to delivering a certain scale throughput performance and availability off of a given infrastructure, right? Seems reasonable to make the shift. What we didn't or what we underestimated was then the requirement that that would place on them to want to get into the details of the underlying hardware and software assets and configuration of the VIM, et cetera. So we look at how do we actually do that in an automated world where we need to bring in software updates potentially on a monthly or weekly basis, bring in new services, onboard new elements, move to the edge, right? None of these things will happen well in an artisan world, right? It will be beautiful, but it will be slow. It will work, but it will be too expensive, okay? And so it's really important that we work through how do we industrialize? And as we learned in terms of how did we industrialize the implementation of virtualization across the breadth of services that we offer, we expect we'll have to proactively simplify. So we welcome the activities that the rest of the industry is banding with us and CNTT and other activities to really help drive that simplicity and the consistency, right? Because you look at companies like AWS and why are they so successful? It's because they essentially hang out the placard and say, build to this and I'll guarantee that you get the performance, right? We need to have the equivalent and that's what we're trying to do, okay? And we'll talk just a bit more about that. The VNF support of network orchestration, we never imagined that the last thing to actually emerge would have been the way to talk from the VNF to the orchestrator, okay? But in fact, that's where we are. We've had orchestration strategies, funding and everything else, sat ready to go and then we find we actually have to foster a lot of collaboration between orchestrator vendor and VNF vendor, and by the way, also when it's the same company that you do that in an open and consistent way, okay? So driving the Etsy interfaces around sole X through five and so forth, making sure that we've got everything that we need in terms of that linkage between VNF, VNFM and orchestrator has come much, much later in the game than I had expected, but we have to drive it, okay? And then lastly, we've spent far, far too long in the virtualization world, okay? The cloud native world is moving at a pace that we can hardly imagine, and we haven't in fact begun to industrialize and move in that direction as an industry, okay? And why is cloud native going to be important to us? Because in fact, cloud native is the first assurance that we have of the ability to scale, of the ability to fully automate, but most importantly, of being able to distribute the application space into whichever geography that we want. And we're quite interested and insistent on that because what we see is that unlocks the edge, okay? And the edge for us may not be as distributed or as many as others, but what we do believe is it's a new era in telecommunications when we offer guaranteed latencies at less than 10 milliseconds, okay? When we offer the ability of applications to be real time in the truest sense, okay? And when we can do so at scale and at pace because we've automated, and without cloud native, I question whether or not we can get there. It is almost certainly impossible for us to get to the layers of, to the level of reliability and consistency of availability that we would need to get to without cloud native as well, okay? Now, three challenges, if we solve these three, I think we're well on our way if not 90% of the way there, okay? And I don't think I'm oversimplifying. The solution space for us, it comes down to some important design choices, some critical collaborations, it's not enough for anyone, regardless of scale, to simply declare. We have to partner with this, we have to partner with our vendors and with the other carriers around the world to make sure that we can get to where we need to get to in terms of streamlining our own infrastructure. And this is a hard nut to get over sometimes because infrastructure is something that we've never had to collaborate on. And in fact, we fought over it and competed with it historically, okay? And this is why we also invest quite a bit of hope and expectation around the CNTT initiative. So I ask everyone in the room to please embrace, help us find out what's shortcoming still in it, but let's find a way to solve it because as with the decision, we find that the decision that we took about three years ago to move from best of breed to best of suite has brought us an integration task that was actually manageable. We're looking at what we're doing in terms of how do we scale. And when you look especially down the road towards a now refactored into cloud native VNF, without having a higher degree of consistency and stability, there is almost no way that we will be able to operationalize either the infrastructure or the services, but maybe more importantly, we question whether or not the VNF vendors can actually have any type of reliable business model or ability to deliver so long as each of us give them a very specific different deployment target. Maybe that's what Amazon taught us above all. It was all frustrating in the earliest days of Amazon because you only got what they gave, but in reality that turns out to be a blessing if you look at the diversity of innovation that they've spawned. And so now maybe we as Telco can learn something from that and with the reference architectures and patterns of CNTT deliver similar simplicity. We have to get consistent in terms of what is the standard that we require of the VNFs and what is the standard that we will require also of the orchestrators. One of the things that has been quite surprising to me is how many times I've actually had to really urge, coerce, demand, whatever you want to describe it, what I thought was quite simple, which is if a VNF vendor, you have to work with multiple orchestrators. If an orchestrator, you have to orchestrate multiple VNFs. Seems really straightforward, but we aren't yet there. So a lot of work has to come in really driving the Etsy interfaces. But remembering, actually all of this is irrelevant unless we can actually integrate to the service layer above us. So actually bringing in now interoperability in our case with TMF, okay? But if I can have anything, please, it is that set of Etsy interfaces that let my orchestrator and my VNFs talk in a consistent way. Super, super simple, but it's stunning when you look at roadmaps, how far out sometimes they are. So please bring those roadmaps in, okay? And then lastly, while I said before, I may not have made the point well enough, if the building scale stage was really landing the business case, the automation stage is what lands the customer experience. And until we have that customer experience that is born out of full automation, cloud native, immediacy, real time, telemetry, ability then to expose the whole thing to the power of machine learning for optimization, tailoring, and prediction, we will struggle to serve our customers as well as they require it, right? The hyperscalers have done a fantastic job of raising the bar to our customers and we live it, right? Each of us in the room is a product of that as well. And we as telco carriers and service providers have to be much more aggressive at driving all the way through to the point that actually the insights of our customers optimize our networks. It's no longer an engineering task. It now becomes an automation task. And that is what cloud native will enable us to become while driving availability, cost, and reliability. So would really appreciate everyone's help to help us get these three things addressed because then I think we begin to finally fly. Thank you.