Every generation of wireless begins long before it arrives, in arguments about what it should be. The technology that reaches a phone a decade from now is being shaped, today, in committee rooms most people never see, by engineers and companies and governments whose disagreements are technical on the surface and strategic underneath. The fight over sixth-generation wireless is underway, years before any consumer will use it, and the things being decided — about spectrum, about architecture, about who builds what — will shape the next decade of how the world connects. The argument is about more than speed, though speed is the part that gets the headlines.

The temptation, when a new generation is discussed, is to frame it as a faster version of the old one. That framing sells handsets, but it misdescribes what the generations actually do. Each generation of wireless has been less about raw speed than about the things that speed enables when it is paired with other capabilities: lower latency, more devices, new frequencies, new architectures. The jump from fourth to fifth generation was not, for most users, perceptibly faster. It was, for the industries that built on it, a different kind of network. The sixth generation will be the same, and the question worth asking is not how fast it goes but what it makes possible that was not possible before.

The spectrum argument nobody watches

The fight that will matter most is the least visible: the allocation of the radio frequencies the new network will use. Spectrum is finite, finite things are contested, and the frequencies best suited to sixth-generation wireless are, inconveniently, the ones already used by other things — satellites, scientific instruments, the quiet services that never make headlines until someone proposes to talk over them. The argument over who gets which frequencies is technical, tedious, and consequential in ways that outlast the technologies that prompted it. The decisions made in the next two years will determine what the network can do, and the people making them are not the ones who will use the result.

The architectural questions are no less consequential. The next generation of wireless is likely to be less a single network than a fabric of networks — terrestrial towers, satellite constellations, and indoor systems cooperating in ways the current ones do not. The engineering is hard; the politics of who operates which layer, and who is responsible when they fail, is harder. The companies that win the argument over architecture will shape an industry for a decade, and the companies that lose will spend that decade adapting to a network they did not design.

Why a generation that does not exist yet matters now

The objection that the sixth generation is years away, and therefore not worth the attention, misunderstands how these technologies come into being. The standards set today determine what can be built tomorrow, and the companies that participate in setting them acquire advantages — intellectual property, design influence, lead time — that compound across the decade of deployment. The generation that reaches consumers in the next decade is being won, in committee rooms and patent filings, right now. The consumers who will use it are not in the room. The companies that will profit from it very much are.

The governments, too, have noticed, and the strategic dimension of wireless has become impossible to ignore. The nation that sets the standards shapes the industry that builds to them, and the industry that builds to them shapes the security of the networks the world depends on. The argument over sixth-generation wireless is, beneath its technical surface, an argument about influence, and the participants are increasingly aware of it. The next two years of boring meetings will matter more than a decade of product launches, and almost no one outside the meetings is paying attention.

The generation before the generation

The wireless generations arrive on a schedule that the industry insists upon and the technology rarely respects. The fifth generation is still being built out, unevenly, across the world, and the places that have it are the minority. The sixth generation will arrive into a world in which its predecessor is unfinished, and the relationship between them — what the new one must do that the old one could not, and whether the old one's gaps justify the new one's expense — will be contested for the entire decade between now and its arrival. The argument is already old. The network it produces is not, and the things being decided in its absence are the ones that will matter most when it finally arrives.