Visit Your Local PBS Station PBS Home PBS Home Programs A-Z TV Schedules Watch Video Donate Shop PBS Search PBS
I, Cringely - The Survival of the Nerdiest with Robert X. Cringely
Search I,Cringely:

The Pulpit
The Pulpit

<< [ License to Pay ]   |  Good News!  |   [ Plug Me In ] >>

Weekly Column

Good News!: Tired of Negativism, Bob Looks Into the Future and Sees Ubiquitous Computing That Not Even Microsoft Can Control

Status: [CLOSED]
By Robert X. Cringely

What the high-tech world needs this week is some good news, and I am determined to provide that. It is in the nature of the news business, I suppose, to dwell on the negative. Just read the newspaper and it is filled with death, crime, accidents, and almost nothing positive. Read the business section, too, and there is little but gloom. But high-tech, in contrast, is constructed almost entirely from optimistic vibes, otherwise these businesses would never have been started. So let's dwell, then, on some positive news.

First, there is the ubiquity of 802.11 WiFi networking. This was a long time coming, but with Intel's departure last week from the HomeRF business, it is evident that 802.11 is not only the dominant wireless networking technology, but that it will soon be everywhere. A year from now, I think it will be rare for a new computer to be shipped without a built-in WiFi adapter, enabling ad hoc wireless networking to become the norm for homes and perhaps even for businesses.

This is great news! It is great news because standardization leads to lower cost leads to larger markets leads to new applications leads to better business. Just the decision to build WiFi into every computer will grow the industry and lead to dramatic changes in network services. The question will not be "Is your home networked?" but "How fast is your home network?" Bring a WiFi device into your home, and it will just plain work. Forget about pulling cable forever.

My home, by the way, finally has complete WiFi coverage. I was never able to make a single access point do the job because of the looming influence of a large, stainless steel KitchenAid refrigerator in my kitchen. The fridge cast an RF shadow that kept me from doing what every nerd wants to do — e-mail from bed. Well, this week I installed a pair of Linksys PLEBR10 powerline bridges to extend my network over the electrical wiring to a second access point in my kitchen. It works beautifully, though I had to return one of the bridges that was DOA and get one that worked.

More good news: XtremeSpectrum has started shipping samples of its UltraWideBand (UWB) chipset with full production slated for early next year. You'll recall I wrote several months ago about UWB, which is an unlicensed high speed data service that is purely digital with no analog components at all. This means that UWB — unlike even WiFi — can take full advantage of Moore's Law. Over time, just changing to new semiconductor manufacturing processes will make UWB faster, cheaper, and consuming less power. The XtremeSpectrum chip, which is the first UWB product to ship aimed at customers like you and me, carries 100 megabits-per-second for a distance up to 10 meters. Think of this as Bluetooth on steroids. Better still, think of it as the eventual Bluetooth killer.

UWB will do for home entertainment what WiFi is about to do for home computer networking. Computers are simple compared to wiring up your stereo system or home theater. When we talk about computer network cabling it means CAT5 and almost nothing else. But a look behind my AV receiver shows RCA plugs, optical plugs, DIN plugs, mini stereo plugs, and others that I can't even identify because they are hidden behind a tumbleweed of twisted speaker wire.

But UWB will change all that. Buy a new component, plug it in, and it will discover the rest of the entertainment network and simply install itself. Look behind your AV receiver and all you'll find is a powerstrip and surge protector. Of course this will take time. The XtremeSpectrum chip costs $19.95 in volume and you need one for each device, so it is still too expensive for true mass consumption. But then it takes time to design new components that have this capability. If Moore's Law holds true, three years from new UWB will be down to less than five dollars, and will be included in just about every device. The eventual elimination of all those plugs can save that much per device in reduced manufacturing cost alone.

Of course, UWB means having to buy all new stuff, which is why it is good news to consumer electronics manufacturers. And to folks like me who just LOVE to buy new stuff.

There is a probable intersection of these two happy story lines — WiFi and UWB. Three years from now, UWB will be running at 400 megabits-per-second or better with a low cost that might make it a WiFi successor. There is the issue of range, sure, but we are likely to see higher range UWB products that XtremeSpectum's current 10 meters. And as WiFi gets faster, its range is getting shorter, too. My guess is that the ultimate home network will be a UWB mesh. Perhaps the individual device range will be still limited to 10 meters, but there will be enough bandwidth available for every UWB node to act as a repeater, extending the network wherever people are in the house. Video, audio and data will go everywhere, with every device simply finding its place on the Net.

Mesh networking is another source of good news with application far beyond out UWB home networks. I predict mesh will be the big networking story for the second half of this decade. For those who aren't familiar with the concept, a mesh network turns every node into a router and repeater, and data more or less finds its way from where it is kept to where it is needed. Mesh networks form spontaneously and adjust as nodes come and go. Mesh networks are not only powerful, they are resilient, and they make very efficient use of bandwidth. My favorite mesh software this week is MobileMesh, an Open Source project you can find in the Links of the Week.

Where mesh networking will really explode is when it reaches critical mass and mesh software embraces a variety of media types (MobileMesh is nearly there already). That's when we will be able to extend the concept of an ad hoc network until it literally becomes the Internet. Let me explain. A major limitation of current networking technology is that it can only use resources it knows about, but a heterogeneous mesh network goes further. Through a process called Border Discovery, a mesh network can find more efficient paths to remote resources. So rather than being limited to your home network, the mesh can reach out over the Internet (or whatever we'll call it then) to remote resources that won't appear remote at all. This is done by creating tunnels that are really more like wormholes, cutting seamlessly from UWB to DSL to Wifi and back, and making your brother's CD collection in New Jersey into an extension of your own.

With a fully-realized mesh network, non-private data (this column, a Nat King Cole CD, or a copy of "Debbie Does Dallas") resides where it resides and the issue becomes less one of location than ownership. This is the Digital Rights Management endgame that all the current industry players are missing. In their determination to not only control all the money, but to also administer a centralized distribution system, they don't see the power of leaving data where it lies. With five or five thousand or five million copies of an audio file or a movie already on the network, why do I have to get my copy from Warner Brothers, rather than from my neighbor? The issue finally comes down to compensation, and turning Warner Brothers from a distribution empire with hundreds of employees to a royalty acceptance web site with no employees means that royalties could drop by an order of magnitude and profits could stay the same.

Think of it this way. I just bought a "Lord of the Rings" DVD at Fry's Electronics for $16.95. That $16.95 has to support not only the movie production, but also an immense manufacturing, distribution, and marketing organization that at the end of the day probably yields two dollars or less in pure profit to the intellectual property owner. So why not cut out that manufacturing, distribution, and marketing operation — and its associated administrative overhead — and instead just hurl a copy of the movie onto the Net, let it propagate as demand dictates, with that same two dollars making its way back to the film makers from every subsequent owner?

That's where we are headed, to a system where Microsoft doesn't control access to media as much as content controls its own use, and only the content creators get paid. And when it all comes together a decade from now, we'll see that for the very reasons I just described it was inevitable.

Comments from the Tribe

Status: [CLOSED] read all comments (0)