Di2: in defence of the wires

Everyone of a certain age remembers the Fisher Space Pen, a multi-million dollar project to create a biro using pressurised ink that would work in space. And it did. The irony, of course, was that normal biros don't stop working in space: the ink is drawn down by capillary action, not gravity. The moral? Don't spend a big wad of cash developing something new when the original doesn't need improving.

That's a criticism that's been aimed at both the yearly incremental differences in top-end road gear, and also the switch to electronic shifting that Di2 represents, I think wrongly on both counts. Okay, the differences between groupsets are pretty small year on year, and you'd soon spend your inheritance buying Dura-Ace every 12 months. But you've only got to look at the bigger picture – say two groupsets ten years apart – to see that vast improvements are being made by these small triumphs. It's only two decades since STI and Ergo were just glints in the eyes of their respective design teams. We may not like the system, but it has moved us on.

By a similar logic, there's no reason to shout down electronic shifting just because wires work nicely. Who could have imagined, in 1990, what a revolution integrated shifting would be? I'm sure we thought we wouldn't need it. Who'd go back to down tube shifters now? Electronic shifting opens up new avenues to explore, some of which are hardly new: I remember testing a bike equipped with automatic electronic shifting back in 1999. You wouldn't want that original Auto-D system on your race bike, for sure, but that's not to say a cadence-driven automatic system wouldn't have significant merits. How often are we told that keeping a certain cadence is key? Why else do we buy computers to measure it?

So roll on Di2 says I, and all its children. but keep the wires. Why? Well, Shimano rightly assert that wireless uses more power, and that it's less reliable. The latter point is a key one for me. I use all sorts of wireless devices – bike computers, mobile phones, HRMs, laptops – and they're none of them 100% reliable. If your bike computer or HRM gives you a rogue reading, you can shrug it off. If you're in a bunch sprint and your bike starts making phantom gear changes it's a different story: that could cost you the stage win. Not that I'll ever be in a bunch sprint or in danger of taking a stage win, but it's the pro scene that drives all this innovation, as it is in every other sport, and we benefit from the trickle down.

But more simply than that, the system isn't improved by dint of its being wireless. It's a static setup, with fixed distances between the components: wires work brilliantly in those conditions. I love having a laptop with a wireless broadband connection that can follow me round the house, but the desktop is still tethered to the router like it's always been. Why? it doesn't move. why would it need to be wireless? Until I'm shown a genuine benefit it is, like the Space Pen, a pointless innovation for that application. I feel the same way about bike computers, by the way. except on mountain bikes, when the distances between the components are no longer fixed. By switching to the ether we gain nothing except unreliability. There'll always be insulating tape and zip ties in my toolbox...