The tech industry has to prove that smart devices are a smart investment.
Smart devices can do many things. Some of them are useful, some are questionable, and some are just plain ridiculous. But the real issue for semiconductor and system companies isn’t whether people will use them. It’s whether they will work long enough and well enough to warrant the extra cost.
The reality is that very few people use all of the features in any device, or even within a single software application or operating system that runs on that device. But all of those features have to work consistently and well enough when they are used, and they have to be current enough that people are willing to buy them in the first place.
Consider automotive infotainment systems. In a June report, JD Power found that for the third consecutive year, this is where the most problems were reported. The cars worked fine as a means of transportation. What people did while driving did not.
This is a big issue for car companies, because brand loyalty depends on user experience. If a driver has to repeat a comment five times to a voice recognition system, or the car makes the wrong choice and distracts the driver, the result can be anything from extreme annoyance to disastrous.
Put in perspective, though, cars are much further along than smart appliances, and brand loyalty is equally affected if a smart washing machine or refrigerator malfuntions. At least there are standards for communications and security in the automotive sector. In the home, there are still questions about what networking protocols will be win out and how different networks can transfer data. Will WiFi, ZigBee, Bluetooth, and powerline all peacefully co-exist?
Even more disturbing is the state of security, which didn’t exist in many of the early smart devices (and many still on the market). In some cases, such as smart meters, consumers didn’t have a choice. The utility company installed them so it didn’t have to send out meter readers. But it’s not hard to figure out that if the heat or air conditioning system is turned down or off, then the building is unoccupied—or that if you can hack into one smart meter, you can hack into all of them.
It’s easy to blame the disaggregation of the electronics industry for this fragmented approach to problems, but the reality is that it doesn’t get better with fewer companies. Big companies historically have tried to turn proprietary standards into de facto standards because it’s less work on their part. This has some ugly consequences, because there are very few markets with just one giant company. Remember Betamax vs. VHS and Blu-ray vs. HD DVD. How about the Unix wars, OS2 vs. Windows, or the DEC Rainbow, which was 98% compatible with other PCs?