Wireless: What's Left to Do?

Let's kick off the second half of the year with an interesting thought experiment, which is basically at attempt to determine the point at which wireless becomes boring.

Farpoint Group is about to celebrate its 17^th birthday, and, if anything, I can report to you that the rate of innovation in wireless today is as high as it has ever been. This is important, because, as new technologies mature, their rate of innovation begins to decline and we're left with no particular reason to buy a newer edition of whatever. Anybody care about their kitchen stove? Anybody running out to buy a new one because of some nifty new feature? No? No surprise here, although I believe that energy consumption is going to become such a big issue that you might just put this one back on your front burner (so to speak), although, barring a real breakthrough in efficiency, certainly nowhere your top priority.

How about your PC? Anything new there? OK, anything new there in the last five years? Vista, you say? Well, you already know how I feel about that one. But unless the styling of the case or dual cores really excite you, the innovation rate in computers is low and will likely stay that way.

Why? Because we have (OK, almost) all we need in PCs. We have most of the apps (or at least most of the features) required for personal and business productivity. We've got the cloud and the Web. The PC itself is about to become irrelevant as we move to platform-independent computing. Checked Microsoft's stock performance this decade? Buying Yahoo (and I still think they'll get it) isn't going to help, but such is another indicator that the traditional computer and software industries, to use an energy analogy, are past peak.

But, ignoring stock prices for the moment, wireless continues to advance at a remarkable rate. In the past year or two we've seen the commercialization of 802.11n, WiMAX, HSPA, and LTE should pop shortly. So many new mobile subscriber units are announced every month that it's impossible to keep up. Ditto mobile software, especially the latest crop of desktop-class browsers and open-source operating environments. We're seeing advances in power-related challenges, like reducing processor power consumption and even in batteries and related components themselves. Given all this, is there any end to this remarkable period of innovation in sight?

Well, believe it or not, yes, there is. And I think the answer to the question I posed above is 3-5 years. Tops. Beyond .11n will come 60 GHz. and 1 Gbps, but propagation at that frequency is tricky so 60 GHz. won't replace .11n - and it thus remains pretty much as we know it today. The battle between WiMAX and LTE will be joined, with LTE capturing the vast majority of new deals here, but WiMAX will still gain respectable market share, especially outside the US. Subscriber units of all forms will continue to advance, but I don't expect any radical new approaches to succeed. In short, in 3-5 years, we'll have all we need. More investment will yield diminishing returns, and that money will need to go elsewhere.

Where? In fact, the biggest challenge we face to mobile ubiquity is the long-term (there's still no issue in the short term) effects of the current global economic slowdown, with growth fundamentally constrained by energy supply and demand. But we will fix that problem as well, absolutely, although such will take much longer than 3-5 years - my guess, again, is we'll have a decent direction in 10-15 years at best. In fact, if I were about to enter college today, and still a nerd, I'd study energy and materials science. I'd learn all I could about raw materials and how to exploit them. I'd study international relations and international business. I'd learn Chinese. But I wouldn't spend much time on EE or CS (or, gulp, even wireless), all of which I dearly love. For best results (or at least the most interesting path), always go where the rate of innovation is highest. We'll shortly have almost all we need in wireless. I wish I could say the same for energy.