Mobile Communications

What's the point of 5G?

The mobile industry has a message for us. Now that the fourth generation of standards is here, it’s time to start thinking about 5G. And one thing makes the headlines: speed. My, we’re already up to 1 terabit a second!

Thanks. We’ll call back. The one thing we don’t need on mobile data is more speed. It’s useless, dinosaur thinking. It would be useful, a bit, if the price per gigabyte was hugely lower and we could switch wholesale from wired connections. It would be more useful if at the same time the coverage was so good we could give up on Wi-Fi. But much better coverage and much better pricing would be stupendously better with or without another hiking of the headline throughput.

Neither price nor coverage are dependent on a generational change, but they’re the two most important improvements we need from our mobile systems.

The whole generation game has stopped being anything but a marketing exercise. Once, it made a lot of sense. The first generation of cellular phones were analogue and had no data capabilities to speak of. Next on the conveyer belt was the switch to digital. Truly generational: 2G gave us GSM: roaming, SMS, far more reliable voice, the ability to change handsets without changing numbers, and the first smartphones. With 3G, there was much faster data - but by then, the system had already started to break down. Before 3G happened, incremental standards like EDGE had already boosted the data speeds and engineers already talked of 2.5G, just not to the users.

However, 3G was still very significant. It was the first time the mobile networks had been engineered to do data as a primary service - and, indeed, voice functionality hasn’t really changed since GSM. But the data was good enough for video and high quality streamed audio.

4G turns up. Superfast! You can get Wi-Fi-like speeds, and indeed you could use it as a wired broadband replacement. If, that is, you were made of money, You could get 50Mbps or more, which, on a contract with a 2GB monthly cap, would run at full tilt for a whole six minutes. Or less.

There’s no point in going a hundred times faster again only to run out of data in four seconds flat. That barely qualifies as a cuddly toy.

And that’s the problem with 5G. It’s implicitly recognised by the industry, which talks internally of a post-mobile-broadband world such as massive machine connectivity, low latency links, and “extreme use case scenarios”. There’s also a bunch of technologies to make better use of spectrum, reduce power requirements and so on, all of which provide incremental benefits to users and indeed will be rolled out incrementally. There isn’t going to be a generational change. It’s not on the cards.

The toupee is slipping. In what was described as a ‘lively’ debate at the Future of Wireless conference in Cambridge in late June, the delegates from industry voted against the need for anything called “5G” at all. So if the people designing the technology don’t want it, the users don’t need it and nobody can really define it, what’s the point of it?

Cash. Each previous generation has seen massive new deployments of handsets and infrastructure, all paid for by us, the users who have been sold - with decreasing truthfulness - a story of incredible new mobile joy. It’s been great: you can now get a decent smartphone on a decent contract for a few quid a month. Thanks for that. But, like the old Generation Game TV show, it’s as over as a Brucie Bonus.

The sooner the marketing side of the industry stops playing games with us and takes a sober look at adding real value from new technology without the addiction to big bang upgrades, the more likely we are to get somewhere worthwhile. That means picking real metrics for coverage, reliability and quality of service, and competing on those. Use the new technologies to make things better, then sell on truth.

There is a lot of work yet to do. We want them to do it, and we’ll pay a fair price to make it happen. Just please, don’t call it 5G.


« Could the lack of cow dung be a problem in urban India?


Connected cars: The dangers posed by hackers »
Rupert Goodwins

Rupert Goodwins expected to be an engineer, but journalism happened. As an engineer, he worked in defence, for Sinclair Research and Amstrad, in startups and for himself. First appearing in print in 1982 and online in 1984,  he's written about all aspects of technology in business for most of the UK nationals and tech magazines, and was most recently editor of ZDNet UK. Tries to solve more problems... See More

  • twt
  • Mail


Do you think your smartphone is making you a workaholic?