Design challenges and slow vertical sector implementation should force ZigBee developers to keep their expectations in check.
For the backers of ZigBee technology, this is a week to celebrate. After years of talk and work, the ZigBee Alliance has completed its specification, a move that will allow developers to take the word "pre" off their ZigBee chips and systems.
Developers and ZigBee backers, however, shouldn't celebrate too long. While today marks a big step for the ZigBee community, it's also the first step in the long ZigBee adoption process a process filled with technical and business challenges and a process requiring extreme patience by ZigBee backers and developers.
For the past few years, developers and marketing folks have talked about the low-cost networking capabilities it will provide and, in the long run, the sensor networks it will allow end users to set up. But, you have to wonder whether ZigBee developers will be able to live up to that promise in the short term.
For years, companies such as Cirronet and Aerocomm have been rolling low-cost 900 MHz and 2.4 GHz radios that developers can use to bring wireless networking capabilities to industrial applications one of the key markets for the ZigBee sector. At the same time, large security companies like ADT have tapped proprietary RF architectures for providing wireless links in their home security systems another big potential market for the ZigBee sector. Now that ZigBee has been adopted, should we expect companies to drop these investments and move to a new, yet unproven wireless technology? Of course ZigBee backers will say yes, but the reality, like with all other wireless technologies, is that industrial and security system developers will be cautious with their initial ZigBee rollouts. So it will win some adoption, but true mass adoption will take time.
There's also a budding question in the sector about whether ZigBee is overkill in applications like industrial, security, and home automation. Several weeks ago, when talking with Cypress, Product Marketing Manager Carl Brasek questioned whether the capabilities provided by ZigBee (in particular, mesh networking) were overkill for some of the applications. "We'll go to a customer and they are not ready to go to ZigBee yet," Brasek said.
OK, so you have to be careful in how you take Brasek's comments since Cypress is pitching its own low-cost networking option called WirelessUSB. But, Brasek makes a good point about the capabilities provided by ZigBee. For years, companies have tried to push point-to-point wireless networking solutions into industrial, remote meter reading, and other vertical markets. And while some companies in these vertical sectors have adopted wireless schemes, many more have shied away from the technology in favor of other embedded networking options. Now that the ZigBee spec has been ratified, designers in these verticals will not only have to deal with another wireless option but also pitches about mesh networking capabilities, which for most may be a long-term option but overkill today.
The other challenge will lie in whether a large or standalone chip company can actually make a business out of ZigBee. Before the standard was even complete, Microchip started giving its ZigBee stack away for free to developers using its PICmicro microcontrollers. When asked about the move, Microchip said that it was in the business of selling microcontrollers and that ZigBee is a basic block it had to provide to its customers. If you read between the lines, Microchip sees ZigBee as a baseline technology it has to provide to customers but not one it can draw revenue from not a good sign for a relatively new technology.
If ZigBee radios turn out to be as cheap as promised, chip developers may also face problems on the integration front. If the radios are cheap, there's no doubt in my mind that the radio will also get embedded into the microcontroller as integration efforts intensify. That means the potential for selling a standalone ZigBee RF chip will also disappear, again hurting chip manufacturers.
Of course, the above concerns don't mean that ZigBee can't succeed in the sector. Right now, with a fresh spec in hand, tons of hype will start to hit the wireless market about ZigBee technology. It happened with Bluetooth. It happened with 3G. It happened with Wi-Fi. And it will certainly happen with ZigBee.
But after the hype starts to dissipate and real designs start to hit the market, there's no doubt that ZigBee has the potential to be a disruptive technology in the vertical sectors. And, if mesh networking truly takes off in sectors like gaming, then ZigBee could be a huge success, making it more attractive for chip and system companies to implement the technology.
ZigBee developers and backers, however, have to keep their enthusiasm in check. The initial hype around ZigBee will be nice but don't expect the technology to have a big impact for the next few years. To be a success, designers have to address the challenges above as well as the ones that will emerge when ZigBee moves from whiteboards to real-life deployment.
Even more important, ZigBee members also have to be patient. While the consumer sector moves at a lightning fast pace, vertical sectors, such as industrial, move much slower. So gaining adoption will take quite a bit of time and, in turn, require extreme patience.