How to ruin the Internet of things: Tie up with a carrier
Source: Galen Gruman
It's bad enough that a smartphone is tied to a specific network. Imagine what happens if your car, fridge, alarm, or other long-lived equipment is
The tech industry is going gaga over the Internet of things, a nebulous set of technologies that make everything from refrigerators to body sensors and wearables communicate with each other and with apps on your smartphone, tablet, PC, and Internet services. One example of the Internet of things mania is Cisco Systems chairman John Chambers' declaration that the market will be worth $19 trillion, a figure larger than the entire U.S. gross domestic product.
But if you look past the hype, there is a lot of exciting, useful technology -- much of it mobile -- being developed under the "Internet of things" rubric. What scares me is that if the crazy hype doesn't kill the Internet of things in its tracks, the cellular carriers might do it in.
The cellular carriers are salivating over the Internet of things, as they see it creating huge demand for cellular data services over their 3G and 4G networks, as devices all try to talk to each other for reasons both smart and stupid. Of course, they want to own as much as that service as possible. In fact, the carriers have been eyeing this market for years, preparing to own as much of it as possible.
So you have, for example, Audi and Tesla both agreeing to make their future cars run their embedded computing systems over AT&T's U.S. cellular network. You have alarm companies tying up with AT&T, Sprint, T-Mobile, or Verizon to provide communication between alarm systems and the monitoring office over cellular, as landline phones continue to disappear from homes. AT&T and General Electric are also looking at ways to embed machine-to-machine (M2M) cellular radios in industrial equipment, transmitting through the AT&T network, of course.
Any buyer in their right mind should recoil in horror at the thought. Any device or service provider should too.
Imagine your car being tied to a specific carrier for its lifespan, which these days is easily a decade. Now think about the inconsistencies in your carrier's cellular coverage: It's great here, terrible there. That coverage quality changes over time even in the same location. At my house, for example, the Verizon voice service that has worked reasonably well for a decade no longer can keep a call active for more than five minutes, and Verizon has no clue why that changed, much less what to do about it.
Do you really want your car, refrigerator, alarm system, and so on to be stuck with a single provider, no matter where you live or drive? It's bad enough with smartphones and tablets, but imagine that situation with something you can't easily replace.
Some neighbors experienced this issue firsthand with their alarm service, which was tied to T-Mobile. Problem is, T-Mobile's signal in the central part of San Francisco where I live ranges from poor to nonexistent. At my neighbor's house, it is nonexistent, so my neighbor's alarm can't connect with the monitoring service. The alarm company of course can't swap out the radio or SIM card to change carrier -- its deal is with T-Mobile, and T-Mobile only. As a result, my neighbors had to pay for a landline they otherwise don't need, on top of several hundred dollars more for the replacement equipment and service to connect the alarm to it.
Another example: My local farmer's market is on a main shopping street, and some of the purveyors use a Square reader to process credit and debit transactions. But a butcher whose iPhones are on the Sprint network can't get a reliable signal at or near his stall, so his farmer's market staff have taken to walking customers to the street with the butcher's iPhone and Square reader to process the transactions. Lucky for him, chances are small the phone will get ripped off, and the phone and its reader can be moved to where the signal is. If you're driving down a highway or wiring in an alarm system, you don't have those options.
We of course have long put up with this nonsense on our smartphones, which in the United States can be legally locked to a specific carrier, although the feds have begun rethinking that issue. Cellullar tablets, hotpots, and cellular PCs have the same tie-ups, which may explain why they sell poorly. For businesses with employees in multiple locations, the carrier tie-up can be a huge problem, and it's one advantage of the BYOD phenomenon, since employees tend to buy devices they know will work where they live, travel, and work. But many companies -- delivery firms and field support, for example -- that use specialized equipment don't have such flexibility.
Yes, service providers such as DataXoom offer multicarrier service to businesses, so you can have a mix of devices under one data plan, but the smartphones themselves are tied to a specific carrier. For iPads, it's unclear: DataXoom tells me Apple says the devices are "configured" for a specific network, but won't commit to them working with a different carrier's SIM. Some bloggers say they've been able to swap SIMs in iPads without incident, though I should note that the Sprint model uses different radio frequencies than the other models because Sprint works with unique LTE frequencies. The carriers' use of different frequencies can in effect lock a device to that carrier as well, because the device's radio supports only that carrier's spectrum range. (Making multicarrier radios is hard, so the practice is limited to pricier devices.)
Other devices supported by DataXoom, such as some Lenovo laptops that use Qualcomm's Gobi radio chips, aren't tied to a specifi carrier, and need just a SIM swap to change the cellular network they work on. That's how it should work for all devices and certainly for all embedded systems (that is, Internet of things devices): The SIM and/or radio module should be field-replaceable.
But such freedom to swap networks works against each carrier's own self-interest, so we'll continue to see tie-ups such as that between Audi and AT&T. The only way to stop that unworkable fragmentation is to not buy anything tied to a specific carrier. We may have lost that battle for smartphones, but let's not lose it for everything else. If carriers manage to tie the Internet of things to their networks, the Internet of things simply won't work, and it will go the way of other badly executed ideas like the Web TV and Wi-Fi telephony.
| }
|