Passing the TestWritten by Michael Hickins Bluetooth SIG members work together to ensure the ongoing success of the specification and to improve the consumer experience. That’s why the Bluetooth SIG Qualification Programs have taken on such great importance and why members have asked the Bluetooth SIG to expand the programs’ scope even further, particularly in the area of product interoperability. The Bluetooth specification provides a baseline for interoperability but does not always define how devices need to be designed, because of conflicts with external standards or because certain feature sets have evolved beyond the specification’s original scope. As a result, a device may comply with Bluetooth wireless standards but fail to interoperate seamlessly with other manufacturers’ Bluetooth enabled designs and products. “There’s no hard line written into the specification as far as how flexible manufacturers and their products have to be,” says Brandon Nott, technical director for the Bluetooth SIG. Interoperability is a particular challenge because of the potential for disruption to the Bluetooth technology brand. When consumers encounter a problem, “they won’t know which device is weaker and will discount the technology itself, and that affects all of our members significantly,” says Nott. “Yes, the Bluetooth brand has been established, but frankly, we haven’t quite delivered yet on the vision of being seamless and easy to use in all cases.” Device-to-Device Testing
Over the years, the Bluetooth SIG has developed a series of testing programs to help improve interoperability. One of the oldest programs is device-to-device (D2D) testing, in which member-donated products are tested against each other. However, D2D can test only a small subset of the total number of Bluetooth enabled products on the market. A more challenging problem is the emergence of so-called “golden devices,” which implement a reasonable set of features very well. In practice, such products can actually limit broad-scale interoperability because manufacturers may cut back on their own device-to-device interoperability testing, relying instead on results from testing against a single golden device. “Manufacturers can get a false sense of security that their products are fully interoperable when they may in fact only be interoperable with very robust implementations,” Nott explains. Profile Tuning Suite
To offset this problem, the Bluetooth SIG introduced the Profile Tuning Suite(PTS) in 2005 to automate and standardize testing and “to close the gap between conformance (with the Bluetooth specification) and the ability to fully interoperate,”says Nott. The impact on product interoperability directly attributed to this automated testing suite is significant. Since the PTS became mandatory in 2006, interoperability failure rates have dropped to less than 1 percent. 
Testing Events
PTS successes aside, the Bluetooth SIG understands that nothing compares tohands-on testing in a controlled environment. Three times a year, UnPlugFests(UPFs) allow members to test their products in a confidential setting. “It’s very helpful, especially for products that include newer specs, and UPFs are the only places to find yet-to-be-released products to test against,” Nott says. “UPF events are safe havens for Bluetooth interoperability testing,” says Mani Elangovan, program manager at the Bluetooth SIG, who has attended and managed 17 UPFs over the past10 years. UPFs have delivered morethan 50,000 hours of interoperability testing against more than 4,000 product implementations. Elangovan estimates that thousands of interoperability issues have been identified and resolved at the events. In addition to UPFs, the Bluetooth SIG also sponsors smaller specialty interoperability events, such as the one for automotive interoperability held earlier this year. These events allow interested manufacturers to test industry-specific profiles and applications. Optional Test Cases
There is still a need for additional testing opportunities and program enhancements, but adding new requirements is not necessarily the answer. “With the expanding use cases of Bluetooth technology and the volume of products, it is increasingly difficult to address the interoperability problems that prevail, and the Bluetooth SIG staff is keenly aware of the implications of introducing new quality oriented test cases into the current testing program,” says Elangovan. In response, the Bluetooth SIG has introduced a series of 30 voluntary test cases. “We’re very conscious of not making anyone’s life more complicated or creating unnecessary overhead, but the idea is that if we can enhance the PTS, members can get a lot of value in exchange for a little extra effort,” explains Nott. The test cases were added to the Test Case Reference List, and many also will be added to the PTS. Unlike tests in the qualification program, the optional test cases were not put through formal approval. Rather, members submit test cases and have opportunity to review and comment. “It’s a new process for doing this, and I expect it’s going to be a little controversial to start with,” says Nott. But the program has the full support of the Bluetooth SIG’s chairman of the board, Ian Macnamara, who says the optional tests are a critical next step for Bluetooth technology. “The Bluetooth SIG isn’t a standards organization – it’s about delivering a best-in-class, interoperability-tested solution,” Macnamara says. “We’ve done an excellent job of delivering a very well-proven specification at the chipset level. Now we need to focus on the application level, which will help the customer experience and make life easier for retailers and operators.” If the Bluetooth SIG doesn’t police itself, Macnamara warns, other organizations will step into the breach, leading to a fragmentation of standards bodies and regulations. “The Bluetooth SIG has all the methods and tools in place, butit hasn’t pushed these optional tests until now,” he says. Nott agrees that the time has come to increase testing opportunities, saying, “If members can do broader testing on the front end, they can drastically reduce support costs and perform necessary release updates, all of which has a direct correlation to dollars.” New York–based Michael Hickins has written about technology and business for Executive Technology, eWEEK, InformationWeek and BNET.
|