Google’s DIY Project Ara won’t fly in the real world

If you’re old enough to have danced to Devo, you also likely remember the build-your-own-PC phenomenon that geeks like me were into, spending many weekends at parts emporia like Fry’s Electronics to get that hard drive, this power supply, and the other motherboard. Dell took that concept and made a mint selling computers customized by you online, then built by the company. The 1980s were a blast!

In a return to the past, Google has been working on Project Ara, an effort to create a standard for modular smartphones that could ship as soon as a year from now. Hardware makers would provide a standard chassis, called an endoskeleton, to which you’d insert various modules and snap a screen. If you want a faster processor, you could take out and replace the existing one — same for the battery, graphics processor, cameras, power connector, and sensor array. You could even add capabilities such as ports and dual-SIM readers by placing them in empty blocks or replacing a component with a new one that served several functions.

[ More on InfoWorld: 10 cool Bluetooth devices for iPhone or Android. | Keep up on key mobile developments and insights with the Mobilize newsletter. ]

You can see why the DIY and Maker crowds are enthused by the idea. It’s Legos for techies — a return to the build-your-own-PC or Heathkit eras a few decades ago, when we stayed in our basements and off the streets.

To be sure, Google has some fascinating ideas in Project Ara, an outgrowth of military research into modular manufacturing for battlefield and espionage devices. The hot-swappable modules — called Phonebloks — are kept together through an electromagnetic lock, so they don’t accidentally pop out while you’re jogging. And there’s a universal bus called UniPro that provides Ethernet-like speeds of up to 10Gbps distributed across the various components. A universal bus is of course necessary, since you could be connecting anything into the device.

These technologies are unproven, and they haven’t been used in any consumer-scale manufacturing. However, they clearly could be relevant in various kinds of field systems, especially those requiring a device that’s hard to replace but has a long, serviceable lifetime. Think oil rigs, space stations, nuclear submarines, robotic factories, and perhaps even factory car stereos.

But they seem to be — to put it mildly — overkill for smartphones and other such personal devices. Modularity comes with several prices, which is why consumer products rarely are modular.

Recall that the build-your-own-PC mania did not survive the popularization of the PC — it was a geek phenomenon in a few cities. Modularity is simply too complicated for most folks. They want a phone, TV, car, computer — not a project. We all love choice, except when we have too much of it.

Modularity also reduces optimization. It’s no accident that the build-your-own-PC movement didn’t make it into laptops, beyond the ability to replace a hard drive with a DVD writer or perhaps put in a larger-capacity battery. What made laptops, then smartphones and tablets, deliver on mobility while being highly functional was their highly optimized, tight integration — custom, holistically engineered designs.

Although optimization can cause its own issues, on balance it’s how manufacturers can best push the envelope of functionality. A generic bus becomes a lowest common denominator — even if the first version, like UniPro, is overengineered to allow for future needs. Think of how many modern devices are hampered because they still have USB 2.0 ports or 802.11b Wi-Fi. It takes years for that stuff to go away.

Ironically, while Project Ara would let such components be swapped, its underlying bus would become the new legacy that slows down all other parts. Translation: You’ll need a new phone anyhow, so why not choose one optimized from the get-go?

A modular chassis is also a barrier to optimization. The size of modules becomes fixed, limiting flexibility in battery size and shape, for example — those irreplaceable batteries people love to hate give you more juice because they can take up more room and don’t have to fit in a specific shape. A modular chassis also eats up space. It’s no accident that Google’s Project Ara prototypes are substantially larger and heavier than the typical smartphone — they need the room and weight for both the endoskeleton and the standard-size modules, which for any particular purpose could easily be oversized.

Modularity is wasteful about the very things people really want optimized a mobile device: size and weight.

It’s great that the U.S. military and Google are exploring modular technologies and notions like a multipurpose bus and an electromagnetically “glued” assembly. I have no doubt that some of this technology will show up in all sorts of devices — and they could be used to improve the manufacturing process of the optimized, highy integrated smartphones we prefer today.

But I don’t believe that outside a few geeky neighborhoods that we’ll see Project Ara smartphones in public use. The rest of the world will buy powerful devices that they can use, not configure.

This article, “Google’s DIY Project Ara won’t fly in the real world,” was originally published at InfoWorld.com. Read more of Galen Gruman’s Mobile Edge blog and follow the latest developments in mobile technology at InfoWorld.com. Follow Galen’s mobile musings on Twitter at MobileGalen. For the latest business technology news, follow InfoWorld.com on Twitter.