Children Of A Stupid Network
By Mike Masnick, Wed Mar 23 03:00:00 GMT 2005

It's not just the "smarts" that are moving down the network, but bandwidth as well. Having a ton of bandwidth at the end points may represent an opportunity, but will anyone do anything with it?


For years, people have been talking about the rise of the "stupid network" where the "smarts" in the network move outwards to the end points, rather than being managed centrally. This is the basis of the open Internet -- and hopefully -- an open mobile Internet as well.

However, it's not just CPU power that has moved its way down the network, but bandwidth as well. Specifically with Wi-Fi, there is often much more bandwidth at the end points than along the next stage of the network. It has been this way for quite some time. While people rushed out to buy 802.11g equipment over 802.11b equipment because of the higher advertised speed, many buyers didn't realize that doing so gave them no actual advantage in surfing the Internet at home -- as they were still constrained by the bottleneck of their broadband connection, often a tiny fraction of the possible speeds offered on the local area network. While there certainly are some who network together computer and, increasingly, consumer electronics, to make use of this higher local area bandwidth, much of it still goes wasted.

This is making some point out that all that wasted bandwidth is an opportunity. The ability to build ad hoc local area networks, possibly even disconnected from the rest of the Internet, utilizing all of that excess bandwidth is an area that not too many have paid attention to just yet. However, the speeds are only going to increase, which should make it even more attractive. The argument made in the article linked here is that 802.11n, with speeds of around 100 megabits per second, will make this possible -- but that might not be the case. Already with 802.11g, speeds are half that (in theory; reality is much slower) and there aren't that many applications taking advantage of it. The speed alone isn't going to convince anyone.

However, one way to think about it might be to compare it to the PC world. For years, PCs grew ever more powerful (explaining the "smarts" at the end of the network). Most of those cycles were completely wasted -- but eventually people recognized that as an opportunity and build specific applications to take advantage of them. One of the first such apps was distributed.net's effort to see how long it took to break certain security algorithms. This was quickly followed by things like SETI@Home, which makes use of spare cycles to go through space telescope data in search of any indication of extra terrestrials.

The distributed.net project began in 1997, long after many people had computers in their homes with wasted cycles. It's now been quite a few years when people have had excess local area bandwidth for ad hoc networks. The real challenge will be coming up with the creative applications that can make use of all that excess ad hoc bandwidth. However, as an untapped resource, it's there and apparently growing all the time.