|
Guest Opinion
November 2000
Look
out
We’re seeing the conversion of embedded devices from inward-focused islands of functionality to outward-leaning ‘net citizens.’
Mark Sigal, Rapid Logic
We hear the term "embedded Internet" bandied about as if it were some ubiquitous technology phrase that doesn't need defining. It seems to drum up two trains of thought. The first, sexy—where we contemplate countless potential intersections of the global network-of-networks with the specialized capabilities of embedded devices. The second, an oxymoron—for how do we reconcile the very "hidden" nature of embedded devices with the ubiquity and rich diversity of the public Internet? However, two technology developments will soon change the future of the embedded Internet from market-speak to market reality.
First, the advent of affordable DSL and cable-modem services has helped the Internet evolve from low bandwidth and intermittent to "always-on" and reasonably "fat." These changes have made the Internet look more like a reliable utility, opening the door to a richer set of applications. The next-generation home-entertainment system, powered by Internet-leaning Sony PlayStations, TiVo-like movie recorders, and Napster-style music jukeboxes, is just one simple example.
The second development is actually the key enabler of the embedded Internet. We're seeing the conversion of embedded devices from reliable but inward-focused islands of functionality to outward-leaning "net citizens" that can be discovered, described, and subscribed to in a standardized fashion. If the first generation of embedded Internet devices were "set it and forget" types of products, the next generation of devices will be much more promiscuous. A key trait of such devices—central to their reason for being—will be their ability to be reconfigured, or customized, for different users or automated daemons, often maintaining multiple personalities simultaneously.
One of the hottest sections of the embedded Internet, for example, is VoDSL (voice over DSL). Essentially, VoDSL, involves the retrofitting of telephony services, such as voice, call waiting, conference calling, and follow-me services, over DSL connections.
It's a simple enough concept. But the actual process—achieving connectivity, enforcing guaranteed quality-of-service requirements, creating secure "virtual circuits," billing for such services, and provisioning the configurations necessary to make such services actually work reliably—is extremely complex. Now imagine scaling that model to hundreds, thousands, or millions of callers, each of whom has specialized service requirements. For example, bronze customers may be willing to sacrifice quality or security for price, whereas platinum customers won't. You begin to realize there is a lot of infrastructure that suddenly needs to be optimized to controlling and synchronizing behavior across lots of devices.
As the embedded Internet moves from market-speak to market reality, the next wave is about creating a surplus of network services. How such services are provisioned is the center of the equation. Tomorrow's challenge (and opportunity) will be to evolve service development and deployment from the "Model T" era (where you can have any flavor of DSL, so long as it's vanilla) to the "My Yahoo" era, where every service is completely customizable.
Author information
Mark Sigal is President and CEO of Rapid Logic.
|
