If you’re planning to build a smart home around Apple’s HomeKit platform, be prepared for higher prices, more clutter, and fewer choices than other platforms.
That’s because unlike competitors such as SmartThings and Wink, Apple doesn’t support the most popular smart home wireless protocols through a central connection hub. Instead, some HomeKit products such as smart light bulbs and dimmers require their own bridge hardware to connect with Apple devices, adding cost and complexity. And because of Apple’s security requirements for third-party HomeKit bridges, some companies may charge more for HomeKit-compatible hardware or avoid the platform entirely.
All of these problems could theoretically be solved by an official HomeKit hub, capable of managing multiple smart home devices through a single connection point. But for Apple, following other smart home players into the hub business would also bring trade-offs in complexity and functionality. As we approach the third anniversary of Apple’s HomeKit announcement–and the company’s WWDC conference, which could bring HomeKit news, approaches–it’s worth revisiting whether a HomeKit hub really makes sense for Apple.
When Apple announced HomeKit in 2014, the company seemed to be betting on Wi-Fi and Bluetooth as its wireless protocols of choice. Because iPhones, iPads, and Apple TV boxes already have Wi-Fi and Bluetooth radios inside, they can connect directly to certain HomeKit devices with no additional bridges or hubs. Apple TVs and iPads even have some hub-like features, such as being able to relay a Bluetooth signal onto a Wi-Fi network, and enabling remote device control from outside the home.
But Wi-Fi and Bluetooth aren’t always ideal for smart-home products. Wi-Fi tends to be power hungry, making it a poor fit for battery-operated sensors, doorbells, and locks. Bluetooth is more power efficient but has a shorter range, and can’t connect to devices outside the house on its own. Although an upcoming version of Bluetooth will allow many devices to form a single “mesh” network, device makers say they’re uncertain about when this will launch, and how well it will work in smart home applications.
For those reasons, smart home device makers often turn to a couple of well-established smart home protocols known as ZigBee and Z-Wave. Both operate at lower power than Wi-Fi because they’re optimized for sending smart home commands rather than a firehose of data, but they have longer range than Bluetooth. They can even extend their range further by relaying a signal across a chain of devices.
“There is a strong reason for companies making light bulbs, like us, to use a protocol like ZigBee, which is basically designed for these whole-home networks,” says George Yianni, head of home systems technology for Philips Lighting, maker of the popular Hue smart bulbs.
The big trade-off is that Hue bulbs require a middleman device to relay commands between the ZigBee radios inside each bulb and the Wi-Fi radios inside phones, tablets, laptops, and other general-purpose computing devices. A Philips Hue bridge sells for $60 by itself, or for $70 in a bundle with two basic Hue bulbs (normally $15 apiece). Lutron also offers a bridge for its Caseta smart dimmers, priced at $80 by itself or $89 with one dimmer and remote.
With other platforms such as Wink and Samsung’s SmartThings, multiple bridges aren’t entirely necessary. Both systems offer $100 hubs that connect with any ZigBee or Z-Wave device, so users can connect to bulbs, switches, buttons, and sensors from a variety of companies. While the up-front cost is greater than buying a bridge for a single smart home product, the cost over time can be cheaper as you add more devices to your home.
Requiring multiple bridges creates complications for HomeKit beyond the added cost. To become certified for HomeKit, each bridge or standalone device must have a licensed MFi chip from Apple, which provides encryption and authentication. While Apple’s high security standards are laudable, they place a bigger burden on device makers, one that might not be necessary with a central MFi-approved hub.
That burden helps explain why Keen Home, a maker of smart air vents, has so far avoided putting an MFi chip in its own bridge, which is required for users who don’t have their own hubs. Will McLeod, Keen’s cofounder and chief of product, says that although the company and its manufacturing partner are MFi-licensed, including the chip would have raised costs for all consumers, including those who don’t want to use HomeKit. Keen is now considering whether to release a separate HomeKit bridge, which would cost more than the current version.
“It’s a non-trivial change, and it’s something that we would have to pass onto consumers, but it might also be something that people are willing to pay more for,” McLeod says.
The MFi chip requirement also creates trade-offs for device makers who stick it out with Bluetooth or Wi-Fi for their HomeKit products. Fibaro, for instance, used Bluetooth instead of Z-Wave in the HomeKit versions of its water leak and motion sensors. The Bluetooth and MFi chips put a bigger strain on battery life than Z-Wave would have, prompting Fibaro to strip away some features such as more frequent temperature readings and more fine-grained light sensitivity controls.
“It’s something that was a challenge for us,” says Rich Bira, Fibaro USA’s general manager. “Some manufacturers, if they don’t do the algorithms correctly, or they don’t spend a lot of time on it, the battery can be a bit worse.”
Waves Of Change
Although it may seem that Apple could solve all these problems by releasing a HomeKit hub, it wouldn’t be a panacea. The up-front cost for a multi-purpose hub would likely be higher than a single-purpose bridge, and could require users to educate themselves about wireless protocols that aren’t in their vernacular.
“I definitely understand where Apple is coming from,” Keen Home’s Will McLeod says. “It’s less for a consumer to understand and learn about.”
McLeod also points out a problem with all multi-purpose hub devices: They don’t support all the things that a device maker might want to do. For example, a platform like HomeKit might not anticipate that a smoke detector should talk to a set of smart vents, telling them to shut off in case of a fire.
“We realize, because we’re deep in the heart of it, that the No. 1 cause of death in a fire is smoke inhalation, and if you can close off the vents when a smoke detector detects smoke, you can prevent that from spreading throughout the home,” McLeod says. “That’s definitely something we want to build, but that’s not something that becomes baked in on a third-party’s platform that’s not thinking about intricacies like that.”
Several other vendors I spoke to also bemoaned the lack of control afforded by generalist hubs. Philips’ George Yianni, for instance, notes that lighting controls are more responsive through the company’s own bridge, and that certain effects such as fading in with the sunrise are more tightly coordinated. Although Philips is open to the idea of putting all its features and integrations onto a third-party hub, the device would need to have enough resources to duplicate everything its own bridge can do, and would require strong business agreements to ensure that Hue support doesn’t get abandoned or fragmented.
“Today, one of the reasons we’re able to keep pushing new features and functionality is we’re able to ensure that all our bridges in the field are running the same software,” Yianni says.
Perhaps the way forward, then, is to approach the hub as an optional add-on for enthusiasts and professionals, rather than a central component of the HomeKit experience. In January, Z-Wave demonstrated a proof-of-concept hub that could map HomeKit controls onto Z-Wave devices. Raoul Wijgergangs, vice president of the Z-Wave business unit at Sigma Designs, which makes the protocol’s hardware modules, says that one company is already working on a HomeKit-enabled Z-Wave hub for later this year. (He won’t give more details, but confirms that it won’t be an Apple product.)
“I think that smart home has proven to be more complex for Apple than they initially realized,” Wijgergangs says. “I’ve experienced them taking a different stand over time toward the more established technologies in this place, and also the more established players and service providers.”
To clear the path for a Z-Wave hub, Sigma spent two years working with Apple on a new security framework, which is now required for all new Z-Wave devices. The upcoming HomeKit hub will have its own MFi chip, and will support all those devices automatically, Wijgergangs says. Although this may create new complexities for consumers, he notes that 80% of the current smart home market is professionally installed. Z-Wave is prominent in that market, so a HomeKit hub could help Apple extend its reach in ways that weren’t possible with just bridges, Bluetooth, and Wi-Fi.
“They are taking more of an approach to embrace some of the values and essentials of this industry,” Wijgergangs says, “rather than take the more traditional Apple stance of, ‘We know it better, we’re going to change everything, and we’re Apple, so the world will adhere to the way we view it.'”