Unlicensed Spectrum, Open Standards and Wi-Fi Bathtubs

Over the weekend, an article from the Washington Post featured the following description of a new, high-end bathtub from Kohler:

Kohler's new deep-soaking tub has underwater speakers and colored lights, but it doesn't disturb your bliss with whirlpool jets, bubbles or effervescence. The acrylic walls of the new "Fountainhead VibrAcoustic" tub vibrate gently with the music while you soak in calm water imbued with colored light. If you tire of ethereal spa music, you can use the tub's WiFi feature to tap into your computer's Internet radio or your iTunes playlist.

As the author admits, “I never thought I would be writing about a Wi-Fi bathtub.”  And I am willing to bet the FCC never thought about or anticipated a Wi-Fi bathtub when it made the so called “garbage bands” of 900 MHz, 2.4 GHz and 5.8 GHz available for unlicensed communication purposes. In 1985, Wi-Fi had yet to be invented.

Such an unimaginable and unanticipated outcome leads me to think of an observation from David Reed, Jerome Saltzer and David Clark in their comment on Active Networking and End to End Arguments. As the authors provide in reference to appropriate network design, the lower layers of a network "should support the widest possible variety of services and functions, so as to permit applications that cannot be anticipated [emphasis added]." This fundamental network design principle is among the key foundations that have allowed innovation to flourish on the Internet. In the same way, by opening the bands to not just one, but any and all entrepreneurs, the FCC created the structural flexibility and openness to permit applications that could not be anticipated. Thus, allowing IEEE and technology companies in 1997 to develop the open 802.11 technology that would later become known as Wi-Fi.

Of course the second part of the Wi-Fi success story are the open 802.11 specifications or standards, designed to foster compatibility among devices and products from different manufacturers. As Mike Marcus provides:

Companies began building 802.11b-compatible devices. But the specification was so long and complex—it filled 400 pages—that compatibility problems persisted. So in August 1999, six companies—Intersil, 3Com, Nokia, Aironet (since purchased by Cisco), Symbol and Lucent (which has since spun off its components division to form Agere Systems)—got together to create the Wireless Ethernet Compatibility Alliance (WECA). The idea was that this body would certify that products from different vendors were truly compatible with each other.

This in turn led to the flexible design and mass production of Wi-Fi chips that are now found in everything from laptop computers to “Fountainhead VibrAcoustic” bathtubs.

Contrast this with licensed spectrum, such as that controlled by wireless companies like Verizon or AT&T. In this licensed context, any future innovation or use of the spectrum must go through the licensee – fundamentally shaping what the spectrum will be used for (in this case wireless voice and data communications) and ensuring that any entrepreneur seeking to build a device, application or provide a service on that spectrum will have to navigate an often risk-adverse corporate hierarchy. Add in wireless companies overwhelming preference for closed (proprietary) equipment standards, and you have a formidable “bottleneck” for innovation.

Early users and engineers of the Internet and its precursor networks recognized the importance of not limiting innovation and development by building-in structural “bottlenecks” or biases into a network. Much in the same way, unlicensed spectrum and open standards limit structural “bottlenecks” to innovation providing an open and flexible framework driven by creativity, entrepreneurship, and technology.