It was going to be the next big thing and its promise seemed boundless. A few years ago, Bluetooth was hailed with almost the same enthusiasm that met dot-coms in their heyday. It was going to be a ubiquitous technology and there appeared to be no limit to what it could do.
We were going to be able to pay for a soft drink at a vending machine with the touch of a button on our cell phones. The moment we entered a hotel, Bluetooth would spring into action and tell the hotel computer who we were and what we wanted – smoking or not. And when we were going out shopping, Bluetooth would be there as well. As we walked down the street, it was going to allow vendors to push coupons and other deals to us through our cell phones or PDAs. If we passed a record store, it would inform us that a new single was available for download in MP3 format – for a small fee.
“If you believed the hype, Bluetooth was going to do everything except cure male pattern baldness and somebody was probably working on that,” said Frank Geiger, the wireless products line manager for 3Com in Salt Lake City.
So far, the hype has proved to be just that. But if you believe the vendors who are still playing in the Bluetooth field, it hasn’t yet gone the way of Beta videos either.
To be sure, some still contend that Bluetooth will one day be performing the complicated programming tricks once promised of it, but they concede that those are many years away. Hopes for it are now much more sobered.
Vendors say, and many analysts agree, that Bluetooth’s primary role is as a cable-replacement technology. Where cables are awkward, unseemly or undesirable as a means to connect two devices, Bluetooth, a low-powered, short-range radio frequency technology designed to connect devices wirelessly, can step in.
Simon Ellis, who works with the mobile products group at Intel Corp in Santa Clara, Calif., describe it as being analogous to USB (universal serial bus) technology.
USB is a ubiquitous way to connect devices to each other, be it a mouse, keyboard or camera to a PC, or a headset to a cell phone.
USB technology appears to be very simple on the surface, but it has a complex underbelly, Ellis said. For users it’s easy – they buy a mouse and they plug it into their PC. But the computer has to understand what the device is, go and get the driver for the device, install the device and then make it available to the appropriate applications.
But USB has one obvious limitation, Ellis said. It has cables.
“So Bluetooth is really the same technology without the cable.”
He wants to make the technology as stable and ubiquitous as USB, but says this will take time – as it does with all such technologies. It’s usually when a technology enters its five- to 10-year period that applications really get developed for it, Ellis said. Bluetooth is at about the four to four-and-a-half year mark, he said.
But while Bluetooth enables wireless connectivity, it doesn’t compete with that other wireless standard – 802.11b, Ellis said. Many of his peers agree.
“I think what the public suffers from more than anything is a misunderstanding of the two technologies and that one has to win and one has to lose,” said Stephen Orr, director of sales for Motorola Canada in Markham, Ont.
Bluetooth is generally viewed as a cable replacement technology, whereas 802.11 is considered a network extension technology. Although 802.11 and Bluetooth share the same unlicenced radio spectrum – 2.4GHz , they work in different ways. 802.11 is the power-hungry workhorse with a longer range. Bluetooth is a lower-cost technology that works at short distances – typically about 10 metres – and requires less power. Bluetooth runs at 1Mb, though the actual throughput is closer to about 700Kb, since the protocol layer eats up some of the bandwidth. PDAs, with their small batteries, could probably more easily support Bluetooth.
802.11 is a TCP/IP or voice over IP technology. Bluetooth is not natively IP, and that means it’s not a LAN technology, said Chris Kozup, a senior research analyst at The Meta Group in San Francisco.
Whereas 802.11 might be the ideal way to link your laptop, with its relatively large battery to the company Ethernet, 802.11 wouldn’t make much sense in a mouse. Here, Bluetooth, with its simpler protocol layer and its ability to survive on low battery power, would make much more sense. A mouse doesn’t need to talk to the network, and it would take a fairly sophisticated computer inside it for a mouse to be able to do so.
However, not everyone believes that 802.11 and Bluetooth have clearly defined and separate roles.
Mark Quigley, a research director at the Yankee Group, Canada, an Ottawa-based analyst firm, says that the two technologies do compete. You could put Bluetooth in airports or hotels to accomplish the same thing – Internet access and e-mail.
“802.11 has a little more market share and mind share,” he said.
For a while, there was confusion about what Bluetooth’s role would be, Kozup said.
“We’ve seen a lot of people back off and say this is really just a cable-replacement technology, not a LAN.”
When Bluetooth first came out, some companies tried to use and market it as a LAN technology, he said.
CeBIT Hanover, a technology tradeshow, once tried to do a large-scale trial of Bluetooth and roll it out across the entire showroom floor as a wireless LAN, but it failed miserably, Kozup said. There were devices with different Bluetooth standards present, and they weren’t interoperable with one another.
This isn’t a LAN technology, he said, and companies shouldn’t try to use it as such.
Where Bluetooth and 802.11 do compete is at the technology level, Kozup said. They operate at the same frequency and so the closer the devices are next to each other, the more they will interfere with each other, he said. They will cause an increase in bit error rates.
Not surprisingly, many vendors disagree with this assessment.
There is some amount of interference because of the shared frequency, but no more than a microwave would interfere with a 2.4GHz phone, 3Com’s Geiger said.
“It may degrade slightly, but nothing the user would notice.”
Some companies trying to achieve the co-existence of Bluetooth and 802.11 have used multiplexing to deal with any interference between the two, Kozup said. So for one second, a device will be in Bluetooth mode, for the next, it’ll be in 802.11 mode. This will affect throughput but may or may not visibly affect performance depending on the application. If you’re syncing your Palm with your computer, you won’t notice, but if you’re using a mouse, even a small delay can be noticeable, he said.
Both technologies were designed to prevent interference, Ellis said. Bluetooth hops among 79 different frequencies at 1MHz intervals. 802.11, depending on the iteration, also uses either frequency hopping spread spectrum or direct sequence spread spectrum.
Whether or not the two technologies interfere with each other, Bluetooth still has some problems of its own to overcome.
Primarily, Kozup said, Bluetooth needs a governing body to oversee its progress and actively do stability testing in order to make sure different iterations of Bluetooth work together. Though the Bluetooth Special Interest Group (SIG) was created to supposedly fill this role, it lacks the weight necessary to fill the gap because membership is on an opt-in basis. This is one of the main reasons we haven’t seen wide-scale adoption of the technology, Kozup said.
The price also has to come down to below US$5 a chip, he said, adding it still has some ways to go to achieve that.
Ellis, who is on the Bluetooth SIG board of directors, disagrees with Kozup’s assessment, saying that the technology is currently below the US$5 range now, which is what the SIG’s goal was.
The SIG has “unplugged fests” periodically where vendors can come and test the interoperability of their products with others. Initially, there were some problems with Bluetooth interoperability and different versions couldn’t always talk to each other.
This was a result of ambiguity in the Bluetooth specification, which has since been cleared up, Geiger said.
“The standard is definitely stabilized and has reached the point where we can manufacture products that are compliant,” Motorola’s Orr said.
Security isn’t a concern either, he said.
“Wireless inherent in its nature, because of broadcast, poses intricacies or unique security issues. Bluetooth resolves and gets away from a lot of those, particularly the ones you hear about with wireless LAN,” Geiger said.
Users can control which devices have the ability to connect to each other through authentication. So if your cell phone is near your Bluetooth-enabled headset, they will recognize each other, but will ignore your colleague’s Bluetooth PDA.
Also, Bluetooth is not intended to have a wide broadcast area, so someone attempting to grab information from the air would have to be in close proximity to the user. The information is encrypted using 128-bit encryption.
Kozup agreed that the authentication technology is there, but said Bluetooth is lagging on the encryption side of things. This is inevitable because the kinds of devices that Bluetooth sits on simply don’t have the processing power to do complicated encryption algorithms, he said.
Many e-business types of applications were first envisioned for Bluetooth during the hype phase. But developers thinking about getting into the Bluetooth field would probably be better off making applications for the consumer market, as many now see it as primarily a consumer technology.
The biggest driver of Bluetooth will probably be the cell phone, Ellis said. But it’s not enough; the Bluetooth device will need something to talk to. Bluetooth headsets are also available and ideal for the technology, as it was designed from the beginning to carry voice. Bluetooth cars are also on the horizon, and they will allow drivers to use their phones without using their hands.