Long read: 5G and Mixed Reality

I am taking part in two roundtable discussions this week, each focused on a specific technology or set of technologies that stands to have a major impact on our world over the next few years. Though I now spend more time talking about business and culture than I do technology, I remain convinced that it has been one of the biggest drivers of change in our world over recent years and will continue to play a primary role in shifting our future. Understanding the potential impact of these technologies – positive and negative – is therefore critical.Here then – in part as an exercise to organise my own thoughts in advance of these discussions – are my thoughts on two questions:

• What is the role of 5G in the future economic success (or otherwise) of the UK?
• What is the potential of AR and VR technologies, particularly in the property realm, and how/when will this potential be realised?

5G: poor marketing

I doubt I am the first to make this analogy, but 5G is a lot like HS2. OK, no-one mistakenly believes that 5G is responsible for causing pandemics. But each is a large-scale infrastructure project, the motivations for which are widely misunderstood.The naming of HS2 is one of the worst marketing decisions ever. Because HS2 is not about speed. It is about capacity. We just do not have the tracks available to move enough trains up and down the country – or across it for that matter. HS2 does not greatly accelerate the journey from London to Manchester (the most pertinent route to me). But it does put high speed trains on a separate track from local stopping services, allowing more of those to operate at a more consistent rhythm. It adds capacity.Most of the headlines about 5G have been about speed of one form or another. How fast you can download a film. The low latency connections that will apparently be critical to self-driving cars (something about which I am a little sceptical). It is understandable that this has been the message as this is what might sell it to individual users keen to be ahead of the curve. But really, 5G is also about capacity of one kind or another.

Increasing efficiency

The first kind of efficiency that matters, is spectral efficiency. There are only so many frequencies on which you can usefully carry information from point A to point B. A subset of these are suitable for mobile devices. With demand for connectivity growing constantly((OpenReach recorded 10 PetaBytes per hour flowing over its network in May, a large leap in a constantly growing figure driven by lockdown - https://www.ispreview.co.uk/index.php/2020/05/openreach-records-10-petabyte-peak-in-uk-internet-traffic.html)), the challenge is on to carry more and more data over the available spectrum. 5G uses new spectrum but it also makes much better use of it, topping out at 30 bits per second (bps) per Hertz (Hz) according to the CTIA ((https://www.telecompetitor.com/ctia-5g-will-provide-big-spectral-efficiency-gains/#:~:text=5G%20networks%20will%20provide%20major,%2DAdvanced%20technology%2C%20CTIA%20said.)) versus the 15 bps/Hz limit of 4G.This is closely connected to the second kind of efficiency: simultaneous connections. We keep adding more and more devices to the network. The compound annual growth rate of the number of smartphones averaged 93% from the early 80s through 2017. There are now an estimated 22bn connected devices – predicted to be 50bn by 2030. 4G is seriously limited in the number of devices it can connect in a given area, its so-called connection density. While 4G networks can connect around 2000 devices per square kilometre, 5G devices can theoretically connect up to a million. If we want to add more smart devices to the network (and some examples of where we might are below), then we need a more efficient network.Of course, all this data must go somewhere once it reaches the base station from your devices. This is called backhaul. As part of the 5G rollout, operators will be looking at how they deal with your data, introducing more computing at the edge of networks for example, and caching copies of popular content so your request doesn’t need to flow across the network.

Too much juice

Then there is the issue of energy efficiency. 4G networks consume a *lot* of juice. Speaking to LightReading((https://www.lightreading.com/asia-pacific/operators-starting-to-face-up-to-5g-power-cost-/d/d-id/755255)), Jake Saunders, the managing director at ABI Research, said a typical 4G cell might draw six kilowatts in power, rising to perhaps nine kilowatts at peak periods. Multiply that by the roughly 23,000 base stations in the UK((https://en.wikipedia.org/wiki/Telecommunications_towers_in_the_United_Kingdom)) and you get a draw on the grid somewhere between 138 and 207 megawatts.Huawei estimates a 5G base station might consume 300% to 350% of a 4G base station, once operating across all the available frequencies. This sounds bad, until you consider the alternative. Even if 4G could support the continuing growth in demand, its power consumption would be orders of magnitude greater than 5G. According to Orange((https://hellofuture.orange.com/en/5g-energy-efficiency-by-design/)), more efficient 5G technologies are expected to divide the energy consumption per gigabit transported by a factor of 10 compared to 4G once they reach maturity by 2025, and then by a factor of 20 by 2030 .So, if we want to keep expanding the number of devices on the network, and the richness of the media we consume, we need to move to 5G.

Slowing progress

The obvious rejoinder from some will be: do we really need these increases? Or should we stick with 4G? While I think there are technological avenues we ought to avoid (for example, some geo-engineering responses to climate change, certain weapons technologies), I don’t think there is a very strong argument for suppressing the development of general purpose technologies like networks. Just like roads and rail before them, networks are not without their risks and challenges. But they are the platform for other forms of progress: scientific, economic, medical and social. Stopping our move to 5G would not only undermine these efforts, it would place us at a global disadvantage.

Delays on the line

Which leads neatly to one of the issues under discussion at the roundtable event: the delay in our 5G rollout caused by the UK government’s decision to prevent Huawei supplying new equipment, and forcing operators to remove what they had already deployed.Personally, I have always been deeply sceptical about the specific threat presented by Huawei’s technology. The telecoms equipment industry, like all high-tech industries, has long and complex supply chains. There are many opportunities for hostile entities to try to interfere with the equipment before or when it is in place. And it is so complex that maintaining security is a constant battle. Do I believe that Chinese government agencies would try to access UK networks through Huawei’s technology? It seems likely. That’s what spies do. But is there any greater risk from us using Huawei’s technology to anyone else’s? That, I find hard to believe.I have been saying this publicly for a couple of years now, though until the last few months have had no actual relationship with Huawei aside from going to a phone launch about five years ago where I walked away with a free device (as did everyone else in the room). Huawei are paying me to be part of this roundtable event, as any other company would have to. But my opinions are fairly well documented long before this point and haven’t changed at all.

GDP impact

Whether or not you believe Huawei’s equipment presents a material risk to our network security, we now have another issue to contend with. Back in 2017, the UK government predicted that once rolled out, 5G would play a key role in 5-6% of UK GDP((https://assets.publishing.service.gov.uk/government/uploads/system/uploads/attachment_data/file/582640/FCCG_Interim_Report.pdf)). If its rollout is delayed, what effect will that have on the economy?Research commissioned by Huawei from Assembly Research has attempted to quantify that effect. It suggests the economic hit will be around £18.2bn. The delay could also “jeopardise £108bn of economic benefit to the UK and the creation of 350,000 jobs in regions outside London and the South-East over the next decade, putting at risk the Government's 'levelling up' agenda.”If you want to understand where those figures come from, then I recommend reading the research (not available to link to as I write this). But I have my own take on this, particularly with regards to where the economic benefits of 5G come from. For me there are two distinct drivers to look at, one of productivity, one of growth.

Hello? Can you hear me?

While our broadband providers have coped admirably through lockdown, I think most people would acknowledge now that our connectivity problems are far from solved. How many calls have you been on over the last few months where people’s video and audio were disrupted, or they dropped in and out of the conversation because of poor connectivity? People are still wandering around their houses trying to get a better signal or working from strange locations because their Wi-Fi or 4G just does not give them a reliable connection elsewhere.Lots of people have tried to quantify the cost of this poor connectivity. In January, Zen Internet suggested SMEs could be losing 72 minutes per day((https://www.computerweekly.com/news/252476441/Poor-UK-internet-connectivity-and-technology-cited-as-hampering-UK-SME-productivity)). Clearly, Zen has something to gain from such statements. But I do not think they are probably that far out. Even if Zen’s estimate is out by a factor of ten, that is half an hour of lost productivity each week, for each employee. That adds up.We need connectivity that stops being a question. You should never, ever, need to think about whether you are connected, only about the application.

Tech-driven growth

The question of growth is more interesting perhaps, but also more speculative. What future applications does 5G support that we do not have today? The simplest answer is just more: as noted above there are hard limits on the number of things we can connect via 4G. But the more fun answer comes from thinking about how you might use the specific characteristics of 5G: ubiquity, latency, bandwidth as required.The first obvious answer is things. Right now, connecting things to the network – particularly remote or mobile things – is still a bit fiddly. Imagine if you could take a thing out of its box and it just worked, with no configuration, for maybe five years before it needed recharging. Think of the applications in manufacturing, fitness, care, logistics, security, entertainment, toys. 5G combined with new battery technologies, passive energy harvesting, and low energy computing should finally open up the potential of the so-called Internet of Things market by stripping away the hassle and fuss and making it easier to deploy connected stuff in the field((Note, early 5G devices have been very power hungry because they are maintaining simultaneous connections to 3 and 4G networks to sustain coverage. This situation will improve dramatically with future generations of the technology and improved 5G coverage.)).The second answer is mixed reality. I have long argued that it is likely we will make a shift from smart phones to smart glasses or headsets. These devices will need ubiquitous rich connectivity, and low latency. Which makes for a neat segue to the second part of this very long blog post.

Living in mixed reality

Right now, we live in two discrete worlds: physical and digital. They touch in various places, like crossing points between dimensions. But they are fundamentally separate. We interact in different ways with these two worlds, one directly and the other indirectly, mediated by a mouse, keyboard, games controller or touchscreen. Or perhaps a voice assistant. We behave differently, communicating online sentiments and in styles we would not dream of replicating in the physical world.These two worlds are separate and distinct, but they are moving closer together. Sixty years ago, the digital world was alien, unintuitive, and required deep expertise to navigate. Those who visited the digital realm were like astronauts exploring an alien space. Today, we can feel our way through the digital world, a space that is more and more familiar. Tomorrow, I think the two worlds begin to truly merge.

Alternatives to dystopia

There are many dystopian visions for this future. Of humans separated from each other by screens spread across their field of vision. Of flashing pop-ups and garish neon interruptions to our field of vision. Notifications pinging endlessly in our ears. But while I think this is a risk in the interim stage, ultimately I can see a much more positive vision for our future cyborg selves.Human senses are incredibly plastic. We can learn to process all sorts of data coming through as sound, touch, taste, sight and smell. You can see this in the work on sensory substitution of the celebrity neuroscientist David Eagleman((https://www.eagleman.com/www/www/research/sensory-substitution)). Or you can see it in the way we learn not just to drive cars, but to sense them. After a while driving a car you stop being conscious of most of your actions. They become automatic. But for them to become automatic, your brain must be unconsciously processing sensory inputs: it’s the tone of the engine and the change in acceleration that tells you when to change up a gear, not the numbers on the rev counter.As we start to live in a permanent mixed reality environment, I think the same thing will happen. We’ll start to ignore the dashboard analogue: the pop-ups, notifications and formal interfaces, and start to rely on more subtle signals. I think the future of mixed reality is like having super powers, or what the original Book of the Future called ‘Extra Sensory Perception’.Imagine all the rich information to which a smart headset might have access translated into streams of sensory information that your brain can interpret subconsciously. Subtle hues of colour, vibrations, sounds, temperatures, electrical stimulation perhaps. A sixth sense for what is happening in your social circle or the stock market. Subtle indicators to point out bargains, risks, or potential dates.The digital world in this scenario becomes translucent: something that colours our interactions with the physical world but does not block them. I think it has the potential to make us more human, not less.There is a huge amount of technological development and interaction design to be done here. But I see this as a more likely outcome than the dystopian futures so often presented.VR, by contrast, will always be an escape from reality. Or at the least, a vehicle into other realities: someone else’s or perhaps our own future. In lockdown, all of those options sound more appealing than ever. But as an escape, they are only somewhere we could, or should, spend a small amount of time. Most of the time we have to face reality, however moderated.So, bringing this back to today and the topic of the roundtable event: how will these technologies, VR and AR, affect the property sector.Right now, I still think the advantage conferred by VR in exploring a property is marginal. It is a very solitary experience, which buying a property rarely is. That is not to say it cannot add value – particularly when the property is not yet built. But the core of the decision making is still likely to be made based on data, photographs, discussion, and a good measure of gut feel.AR has a much greater role to play, though perhaps less immediately. AR will be a critical tool for everyone in the property value chain. It will be used to capture images and plans of land, overlay existing properties with virtual remodels, highlight critical information from utilities to crime, inform contractors etc. We are just not there yet with the hardware, the interaction design, or frankly, the network.It is perhaps fitting that it is in a property industry context – a business that deals specifically with the built environment – that we have this conversation. But tomorrow’s world is one where physical and digital are largely indistinguishable.