Dana Blankenhorn has been a business journalist for over 25 years and has covered the online world professionally since 1985. He founded the "Interactive Age Daily" for CMP Media, and has written for the Chicago Tribune, Advertising Age, and dozens of other publications over the years.
About this Site
Moore’s Law defines the history of technology. It held that the number of circuits etched on a given piece of silicon could double every 18 months as far as its author, Intel co-founder Gordon Moore, could see. Moore’s Law has spawned constant revolutions since then, not just in computing but in communications, in science, in a host of areas. Moore’s Law applies to radios, and to optical fiber, but there are some areas where it doesn’t apply. In this blog we’ll take a daily look at new implications of Moore’s Law in real time, as it rolls forward to create our future.
It’s part of the future, no doubt. It’s even part of the present.
But the assumptions that Internet traffic is growing mainly in response to video, that Internet-capable networks must give video 99% of their capacity, or that Internet Law must be changed to accommodate video are fictions.
The Video Fictions are relics of the pre-Internet age. They’re wrong for three reasons:
Video is passive -- When you’re watching a video you’re watching, you’re not interacting. The Internet is all about interaction. It’s about ideas. It’s about interruptibility. It’s about cutting your attention into as many pieces as you can, multi-tasking in order to do more. Video takes all your attention, and demand for it is limited by audience attention.
Video is expensive -- A quality blog item, like this one, can be created by one person in a few hours. A quality video takes the work of many people over many days, and bad video takes just as much time to make as good video. You can’t have both good video and interactive video. Good video just takes too long to make.
Video has plenty of channels – Most of your cable bill is taken up by worthless nonsense already. There isn’t enough quality programming to fill the DirecTv and Dish Network satellites. Broadcasting has worked for almost 90 years. All these deliver more programming at far less cost than the Internet ever could. The Internet, as a video medium, is best served for tiny niches, with low demand, and it already does this.
The assumption that “the future of the Internet is video” is driving just about all the stupidity we see among big companies and policymakers today.
There are video applications which have value on the Internet, but they don’t need the bandwidth or Quality of Service (QoS) up-sells of true video. Videoconferences are of value (sometimes) and video VOIP calls can be of value (to long-separated family members). But the idea that we need the Internet to watch the same TV that comes to us via satellite and cable is nonsense.
There are also some applications that can use QoS standards, and payments. Interactive games can use QoS, especially when players are going against one another in real time. Medical applications can use QoS, although those applications that really need it should be done in clinics or hospitals with ample bandwidth, not the home.
Meanwhile, there is an enormous, and growing bandwidth shortage in the average Internet home. I face it every day. Why?
As the first non-engineer to rise to the top at the chipmaker, he said he would push platforms, and communications, and low power, and change the corporate culture. He would also push more product internationally, and put consumer products (not just ingredients) under the Intel brand.
Whether he is moving rapidly enough remains an open question.
There are enormous claims of movement this week at CES, where Intel built a booth that would have made its former Comdex managers blush, and changed its slogan besides. But is Intel really leading? Or is it following?
What's with this getting in bed with Microsoft on a Media PC? Don't they know those things don't sell? Microsoft does.
Why is Intel continuing its campaign to pollute the developing world (this time Vietnam) rather than deal with the pollution problems caused by semiconductor manufacture?
The Always On medical market won a big endorsement today from a San Francisco research house, FocalPoint Group, which advised hospitals that the technology is ready to lower costs and improve care.
The study projects that more than $7 billion will be spent on wireless data applications in the United States by 2010. Technologies, including WiFi, RFID, cellular, and low-rate ZigBee modules will be used to improve asset tracking, patient monitoring, and emergency response situations. In each case, these technologies are being implemented in health environments today and are expected to restructure the ways in which hospitals are organized and to handle patient needs.
In terms of the work I've been writing about these are fairly primitive applications. Track the medicine in the pharmacy, give doctors on call instant access to records they can read, that kind of thing. We're not yet talking about wireless monitoring of patients, or following those patients after they leave the hospital.
Despite this we're talking about huge savings:
With the vast majority of hospitals relying on paper-based systems, more than 770,000 Americans are injured or killed every year from adverse drug events, costing between $1.5 billion to $5 billion annually. Personal digital assistants (PDAs) can improve the legibility of drug orders while patient-worn wireless bracelets communicate information about drug allergies or deadly drug interactions -- potentially saving the healthcare system billions. In addition, long-range wireless transmissions and electronic patient records provide more complete and continuous data feeds about patient history, health and activities.
America's biggest tech companies are focused today on the problem of creating, not technologies, but platforms.
Not that there's anything wrong with that. Intel and Microsoft and Cisco all rose to prominence with platforms. The first two had "WinTel," a marriage now on the rocks (Windows works fine with AMD, Intel will make Apple chips). Cisco had the Internet platform.
These companies changed the world. But the world is a funny place, a "what have you done for me lately" place. In business, it's a "what are you going to do for me next" sort of place.
Microsoft, like it or not, has defined a platform strategy. To that extent, Windows still works. The problem is seen most visibly at Cisco and Intel. Let's tackle Cisco first.
The report correctly identifies the biggest problem, user acceptance:
Concerns over privacy and data protection are widespread, particularly as sensors and smart tags can track a user’s movements, habits and preferences on a perpetual basis. Fears related to nanotechnology range from bio medical hazards to robotic control.
None of these are unreasonable fears. Addressing them requires acceptance of some very new, and important societal values:
Privacy
Personal control of personal data
These must be enforceable to have meaning. The technology and tools for all this have been around for years now, but the business has not gone anywhere because no country on the face of the Earth has yet accepted the fact that it must give up absolute rights to its citizens' data before people can trust the technology enough to use it.
There are cell phones, there are WiFi phones, there are cordless phones, and there are VOIP phones. But never the twain shall meet.
Now a universal wireless phone has come a big step closer, with news that WiSpry has perfected final elements of the chip design, and will work with Jazz Semiconductor to produce the necessary chips using CMOS technology.
The two firms' release on this subject is deser than a good bagel, but like such a bagel it's worth the effort. WiSpry calls its technology RF-MEMS, combining Radio Frequency (RF) technology with Micro-Electro-Mechanical Systems (MEMS), a combination of electronics and mechanics on one chip.
The result, in this case, is a one-chip tuner, like a radio that carries its own channel selector. The difference in this case is that the channels are different communications networks. Thus the same device can handle the frequencies used by cellular carriers and those used for WiFi, and more. In quantity, at no additional cost. Voila -- universal phone.
The "problem" is that an Always-On data appilcation (say a heart monitor which phones the doctor when there are potential problems) won't use the network very often. (ABI calls these applications M2M, for "machine to machine.") As a "user" a heart monitor is a $10/month account (maybe). The distribution channels used for cellular, which depend on people, can't deal with that kind of account. (I'd always thought you piggybacked on your individual account, but never mind...)
The direct route is for the carrier to create a gateway, as Orange has in the UK. The problem with that is the carrier is only going to deal with large-scale developers. They might sign a deal with an ADT (for security) or a Honeywell (for home management), but not with a small, local guy -- too much trouble. The indirect route is to have someone else run a gateway equivalent, allowing the creation of a Movile Virtual Network Operator (MVNO) to handle data services. The trouble with that is that the operator gets cut out of some of the revenue, which they find unacceptable.
The result is nothing. There are no big application developers because there are no small scale successes. And there aren't any small scale successes because the carriers won't do business with them directly.
There's a lot of hyperbole there. (Patrick Henry, right, was nothing if not hyperbolic.)
But the fact is that the tools and technologies needed to create a "hot zone" -- an area that can get 802.11 wireless coverage -- keeps going down.
There is no need for such zones to be defined by political boundaries. There is no need for there to be just one such network in an area. There are tons of places near me that have multiple networks in reach. That's the beauty of the unlicensed band.
What you need to deliver a HotZone to a corner, a neighborhood, or a development are:
Antenna arrays that let you serve your whole service area efficiently, by which I mean efficiently in terms of spectrum use.
The biggest danger to this vision is coming, the mergers of local and Internet backhaul outfits to be known as Verizon and AT&T.
If those companies are allowed to consolidate and control Internet backhaul and sell it through an eye-dropper, as they now sell broadband through an eye-dropper, then they can halt the American wireless revolution in its tracks.
My friend Glenn Fleishman points out that a city concession dramatically cuts the cost of a WiFi cloud. The "winning" company has the right to use city streets, utility poles, and city owned land. They have a big leg-up in the market. (That's Glenn, left, with his son Ben, now 14 months according to his blog. My own son John is now 14 YEARS old, and I was feeling nostalgic.)
But are these benefits truly necessary?
Would the lack of a concession double a company's costs? Triple them? I don't know. But I believe, given the recent advances in Wi-Max technology, they're not determinative.
Should Earthlink prove "successful" in Philadelphia, in other words, there is nothing to prvent Verizon from creating its own, competing cloud. There is nothing to prevent a competitive fiber company, such as Level 3, with installing a collection of WiMax base stations in any city where it does business and offering WiFi backhaul to whoever wants it.
A mesh, in which all devices on a network are connected to all other devices, finally has a hockey-stick chart. InStat's new report has last year's $33.5 million in sales growing to $974.3 million in 2009, a classic hockey stick formation.
InStat credits military needs with developing the technology, but there are many advantages to installing a mesh as opposed to a single hotspot:
A mesh can cover a large area.
A mesh can make certain the coverage area is completely covered.
Mesh can connect to many different access technologies, not just WiFi but UWB and WiMax on the wireless end, or WiMax and a fiber pipe on the backhaul end.
The best way to kill a promising technology is to argue about it in standards bodies.
That's why UltraWideBand hasn't come to market yet. The technology works, but there are two ways to implement it.
One path is offered by the Intel-led WiMedia Alliance, supporting something based on OFDM, a technology already-implemented in newer WiFi standards.
The UWB Forum is concentrating on gaining members in China and Japan, supporting something called Direct Sequence-UWB. (That's the UWB Forum logo to the left.)
The technical merits of these two solutions is beyond my expertise. The plain fact is that they continue to argue, and now that we're ready to roll out product users will have to choose which side to support.
Most will simply choose to support neither. And by the time one side muscles the other out of the marketplace, there may not be much of a market worth competing for.
The movement of network boundaries ties together all the trends of the present time.
By the network boundary I mean the point where your client, which you control, ends and a network which is beyond your control begins.
Crossing the network boundary requires more than a cost-benefit analysis. It also requires a trust-benefit analysis. You have to trust the network, and the network owner, before you make the jump. (The illustration of the word Trust is from Professor Myoung Lee of the University of Missouri.)
So trust is a vital asset to any company seeking to lure people across the boundary. This is why Google's credibility is so vital, and why CEO Eric Schmidt has to go, because he doesn't understand that and his actions threaten Google's credibility.
The frontier in computing today is the placing of personal data and applications on the other side of the network boundary. GMail represents both data and applications. That's what makes it an important product.
But there are many other appications that could be handled on the other side of the network boundary. All the things we consider desktop applications could be handled on the other side of that boundary. Trust,. or the lack of it, is what keeps those assets on our side of the boundary.
We have known for years there are many benefits in placing our data and applications on the network side of the boundary. Our clients can become simpler, for one thing. Our costs can be reduced, for another thing. Our stuff is more accessible, especially if we build access to it into all our clients.
But there are risks to doing this, trust risks. Government could get into our stuff if it's on the other side of hte boundary. So could private actors -- bosses, competitors, hackers. And then there's the question of how fast and reliable the network connection is, which now separates us from our stuff and our applications.
This is why the U.S. technology lead is threatened by politics today. Our lack of trust in the government keeps us from moving our stuff and our appilications across. And the government's asinine policy on networks -- private unregulated duopolies of cable and phone giants -- means the cost benefits of moving these things across is lower for Americans than for people in other countries, in Asia and Europe.
The speed of networks determines our technical ability to cross the network boundary.
Every decade of computing technology can be summarized fairly simply. (That's an Apple ad to the right.)
The 1950s were the decade of the computer.
The 1960s were the decade of the mini-computer.
The 1970s were the decade of the PC.
The 1980s were the decade of the network.
The 1990s were the decade of the Internet.
The 2000s are the decade of wireless.
It's now clear that wireless technology defines this decade. Mobile phones are opening up Africa as never before. WiFi is making networking truly ubiquitous.
Walk or drive down any street, practically anywhere in the world, and you will find people obsessed by the use of wireless. Behaviors that in previous decades were shocking -- walking around chatting animatedly to the air for instance -- are now commonplace.
What's amazing, as we pass the halfway point, is how far this evolution has to go, and how easy it is to see where it can go:
WiMax to link islands of WiFi, and to make true broadband mobile.
Interlinks between cellular and WiFi networks.
Devices that truly take advantage of wireless broadband.
Applications that work automatically, with wireless as a platform.
Intel holds the telecommunications balance of power in its hand.
Here's how The Register puts it, with its usual hyperbole:
Intel is throwing its financial, technical and lobbying weight behind the rising tide of municipally run broadband wireless networks, seeing these as a way to stimulate uptake of Wi-Fi and WiMAX and so sell more of its chips and increase its influence over the communications world.
The question of Wi-Fi and real estate is about to come to a head, at Boston's Logan Airport. (Picture from MIT.)
Declan McCullagh reports that the Airport is trying to close Continental Air's free WiFi service, based in its Frequent Flyer lounge, in favor of a paid service on which it gets a 20% cut of revenue.
Continental has appealed to the FCC under the 1996 Telecommunications Act. Massport, which runs the airport, is making bogus arguments about security (its paid service uses the same spectrum as Continental so if one goes under its argument, both go).
If this thing goes to trial it will be a very important case. Here's why.
I'm a big fan of both Marc Canter (right) and Joi Ito . (NOTE: The picture, by Dan Farber of News.Com (and ZDNet fame), was taken off Marc's blog.)
They're both brilliant. They're both A-list bloggers. They're both rich. I've known both for about two decades.
But I think Marc has a vital Clue Joi has missed, about one of the most important trends of our time, the rise of the open source business process.
Here's why I think that.
Joi has put a lot of money into SixApart, which runs Movable Type, which powers this blog. It's good stuff. But it's being left behind because it is, at heart, proprietary. It doesn't interconnect with other software. It isn't modular, scalable, and it can only be improved by the SixApart team.
In other words, it doesn't take advantage of the open source business process, and thus there are whole new worlds it hasn't been able to scale into. It's not a Community Network Service (like Drupal), and it's not a social networking system (like MySpace).
Marc, on the other hand, has just released GoingOn. It's a new engine for digital communities, like MySpace. He launched with Tony Perkins, who will use the system as the new heart of his AlwaysOn network (no relation to my wireless network application idea of the same title).
Marc calls GoingOn an Identity Hub, something to which other identity systems can connect. (It's interoperable with Sxip Networks, for instance.)
But Marc also understands that his stuff can't be the be-all and end-all. Let him explain it:
Blogging is filled with comings-and-goings. Mostly comings these days.
But some goings as well. Some, in fact, are quite sad.
Put this on the sad list. TheFeature is no more. I found out about it a few minutes ago, and confirmed it at the site.
TheFeature was among the best blogs I've seen on the mobile Internet. Their "columnists" ("bloggers") were good writers, with good sources and real insight.
You can see the reactions of some of those columnists here. Russell Buckley, who clued me in on all this, has also announced his own blog, Mobhappy, and one of TheFeature's best, executive editor Carlo Longino (above), is moving over there.
It's nice when "real" (paid) market analysts agree with one of your premises. Especially when it's a key premise to you, as Always On is to me. (This is advertised as an Always On Server, from Virtual Access.)
M2M stands for Machine to Machine (ironically this sits right below an item about how poor most tech nicknames are) but we're talking about the same thing, intelligent sensors linked to wireless networks. Programming the sensors to deliver some result, then automating delivery of the result in some way (sending an alarm, telling the user, etc.) is what I mean by an Always-On application.
As I have said here many times the tools are already at hand, and cheap. We're talking here about RFID chips, WiFi and cellular networks, along with standards like Zigbee that let these things run for years on a single battery charge.
There are problems with every application space, however:
The Supreme Court has decided that cable networks, created under government franchises, under monopoly conditions, are entirely the property of their corporate owners who don't have to wholesale. (That's the BrandX rocket ship -- they lost the case. What follows is directed to them as much as anyone else.)
Some ISPs bemoaned this bitterly. In the near term it means most of us have two choices for broadband service, the local Bell and the local Cable Head-End, both known for poor service, high prices, and loaded with equipment it will take decades to write off.
I spent last week in Texas, dependent on free WiFi hotspots, and I learned a powerful lesson.
There is no such thing as free WiFi.
When "free" WiFi is provided by a bar, coffee shop or restaurant, there is a quid pro quo. You're going to eat. You're going to drink. And when you're no longer eating and/or drinking (and ordering) you're going to get nasty looks until you leave.
There is a cost to a shop's WiFi that goes beyond the cost of the set-up. That is the cost of the real estate, the cost of the table, and the cost to a shop's ambience when a bunch of hosers come in and spend all day staring at laptops.
The only unfortunate thing here is that the writers of the release on this interesting report call the area eHealth.
My problem is not with their intent. A rose by any other name and all that. My problem is that the term eHealth is stifling, limiting. It minimizes what is actually happening, and isolates wireless network applications to one small field.
One reason I (unreasonably) went off on Jamais Cascio is because I'm sickened at how the press generally treats Always On solutions. They only see the threats to civil liberties and tend to demean the potential user base.
After Jamais (rightfully) went after me I began looking for an article illustrating this point. It didn't take long to find one. (And the picture at right is from that very story.)
Here it is. It's a piece by Thomas Ricker of EnGadget on what are some really nifty Always On applications in the medical field.
He gets it all down, the fear of "Big Brother watching you" and the outright contempt for the infants, parents and older folks who might need this stuff.
Given all the deaths from SIDS I would think parents would love a mattress that could warn you before your child dies. Given the ravages caregivers face with Alzheimers (not to mention patients), a network of motion sensors telling you when you really need to help grandma (and when you don't) sounds like a very, very good thing indeed.
Did you even read the article? It sure seems like you didn't. I know it was long, but if you're going to attack it, do so for good reasons (like the term) and not for things I didn't say.
It's not about the government.
It's not about Big Brother.
It's not about big business or big science being evil.
It's not about the developments I discuss being unmitigated problems or uniformly undesirable.
To quote from the very beginning of the article:
"This won't simply be a world of a single, governmental Big Brother watching over your shoulder, nor will it be a world of a handful of corporate siblings training their ever-vigilant security cameras and tags on you. Such monitoring may well exist, probably will, in fact, but it will be overwhelmed by the millions of cameras and recorders in the hands of millions of Little Brothers and Little Sisters. We will carry with us the tools of our own transparency, and many, perhaps most, will do so willingly, even happily."
...and a few paragraphs later...
"But in the world of the participatory panopticon, this constant surveillance is done by the citizens themselves, and is done by choice. It's not imposed on us by a malevolent bureaucracy or faceless corporations. The participatory panopticon will be the emergent result of myriad independent rational decisions, a bottom-up version of the constantly watched society."
(Emphasis added, as you seem to have missed it the first time.)
Jamais Cascio at WorldChanging (the picture is by Howard Greenstein, and published on the blog item) has decided to dump heavily on the whole idea of the World of Always On by giving it the ear-splitting name of the Participatory Panopticon. (The speech was made a month ago, but it hit my RSS feed this morning.)
It's wildly over-the-top and based on the idea that any data you collect on yourself must mean Big Brother is Watching You. There's even the obligatory Abu Ghraib glory shot. (The guy on the box, not Lynndie England's smile.)
There are always dangers in technology. People die in traffic accidents. Most mass entertainment is bad for your mental health. If you play video games too much your thumbs will hurt.
But to condemn new technology because of the potential dangers if it's misused is Luddism at its worst.
The fact is that, where Always On technologies are concerned, we're all Big Brother. That is, we're monitoring ourselves, monitoring and controlling our own environments, finding our own stuff.
The key to making this work is, simply, Privacy Law. Data I create belongs to me. Data about me belongs to me. Data about your stuff, in the store, belongs to you, until I buy it and take it out of the store at which point the RFID chip, and its data, belong to me.
My heart rate and blood pressure, designed to keep me from dieing in that next heart attack? It's mine. But I can share it with my doctor.
The list of what's in my house? Mine, too. But I can share it with my alarm company. And if someone breaks in I want them sharing it with the police. An inventory of what's missing is a good thing.
My free free weekly e-mail newsletter, A-Clue.com, has become very wide-ranging since its launch in 1997 as a discussion of e-commerce.
One of my continuing themes is the World of Always On, with wireless networking as a platform, running applications that use data from your daily life.
But before we get there we all have to become network managers. In today's issue I consider that question.
Enjoy.
I'm a network manager. (MG-Soft of Slovenia makes products for network managers. That's their mascot, Mr. Monet, at left.)
It's not that I want to be. I'm a homeowner. My kids have PCs. My wife and I have PCs. Some years ago a friend ran wires among the rooms so everyone could share my DSL line.
There are now millions of us network managers. Recently I sat on my porch, opened my laptop, and learned that three of my five immediate neighbors now have WiFi networking in their homes. The signals were faint, but my copy of Windows found them all as soon as I booted-up. And the nearest of the three was totally unsecured. If I had larceny in my heart I could have entered my neighbor's network, used their bandwidth, even prowled around in their PCs looking for porn, passwords or blackmail material. (Fortunately for them, I'm a very nice person.)
The other two neighbors had nets which, like mine, are protected by long identifiers, input once, which validate valid PCs. One even had encryption on their system (very nice). The neighbors on the unprotected net insisted later they had the same system I do, but I suspect they haven't taken time to activate the security features.
The point is that wireless networks make many of us network managers, and Always On applications will make most of us network managers. We're not qualified for the work. We may never be qualified. Those who do become qualified become that way as I did recently, in extremis.
Despite his ponytail and his sometimes counter-cultural language, despite being what I like to call a Truly Handsome Man (it's a brighter term for bald, people) Ted Waitt was always a follower, not a leader. (The picture is from a 2002 profile in the Sioux Falls, South Dakota Argus-Leader.)
Waitt was Gimbel's to Michael Dell's Macy's. He wanted to be Pepsi to Dell's Coke.
But computing lacks the stability of the retailing or the soda business. So when Waitt announced his resignation today (at 42 it wouldn't sound right to call it a retirement) it wasn't big news.
Waitt and Gateway did well in the 1990s, following Dell into mass customization. He made his big mistake when he tried to out-think Dell, opening a chain of retail stores that caused $2.4 billion in losses, according to The New York Times.
But I personally think the mistake was more basic than that.
I didn't blog much yesterday because I was researching the state of play in Always On. (The illustration is from Georgia Tech.)
I had a book proposal before Wiley rejected out of hand. But when I then suggested to step back and do a book on RFID for the home, I got real interest. Just make it a hands-on book, I was told.
Thus, the research.
As regular readers here know well there are many Always On application spaces, that is, functions fit for wireless networking applications.
Medical monitoring
Home Automation
Entertainment
Inventory
Absent this understanding that a unified platform already exists so that all these applications can be created together, what is the state of play specifically regarding Radio Frequency Identification? (Or, if you prefer, spychips, although since I'm talking about home applications you're spying on yourself.)
Verizon pulled a bait-and-switch on New York phone booths. It installed 802.11 equipment based on the promise of free WiFi service on adjoining streets, then pulled them all back into its paid network.
The best way to understand the future is to look into how chips are changing.
Two transitions are transforming Moore's Law. The original article, in 1964, described only the density of circuits on silicon substrate.
The rule implied that chips could get better-and-better, faster-and-faster. Doubling bigger numbers means bigger incremental changes in the same time. Over the years chemists and electrical engineers learned to apply this exponential improvement concept to fiber cables, to magnetic storage, to optical storage, even to radios, so that 802.11n radios will transmit data at over 100 Mbps -- twice what earlier 802.11g models could deliver, but still 50 Mbps more.
The transitions have to do with what we mean by better.
Intel plans on mass producing WiMax chips and going into rapid deployment, offering end-user speeds far in excess of what U.S. phone outfits provide with DSL.
The problem is that's the speed limit for most backhauls. Go to most WiFi hotspots, or most home networks, and DSL is the backhaul platform. We're talking 1.5 Mbps, max.
There are two types of chips key to the Always On world.
These are sensor chips and RFID chips.
Both contain tiny radios. The two can also be combined.
A sensor chip, as its name implies, tests specific conditions, and is reporting back with data on those conditions. A motion sensor is an example. A heart monitor is an example.
An RFID chip merely identifies the item it’s on. The chips that will go onto passports will be RFID chips, and RFID identification is at the heart of efforts by retailers by Wal-Mart, as well as service providers like Grantex.
I’ve also written, recently, about applications that combine RFID and sensor ships. Bulldog Technologies is rolling out a line of these chips that not only identify containers in transit, but monitor their condition and shippers know the contents are safe.
Always On applications will use all these types of chips as clients on WiFi or cellular networks, with applications located on gateways that run at low power, with battery back-up, and have constant connections to the Internet.
The coming issue of Business Week features a short story on the Internet of Things, or Machine to Machine (M2) applications, which this blog calls Always On.
The story focuses on cheap cellular radios and industrial applications.
The story misses the opportunity and the market.
It's a good example of the Intel failure noted below because if no one is going to tell the story a reporter can't write it.
Cellular can enhance an Always On application, making it mobile and ubiquitous. If you have a heart monitor in your shirt you don't want to die just because you walked outside the reach of your Local Area Network.
But these are enhancements. And the industrial market is just the tip of the Always On iceberg.
The big money, as I've said, is based on the wireless broadband platform.
It's true that wireless broadband isn't seen as a platform now. It's seen as an end-point. It's seen as a way for you to link your PC to broadband resources. It is seen as an extension of an existing IP protocol. And a lot of people are waiting for IPv6 to tag every device with a unique number before getting excited over linking such devices.
This is very misguided. You can build true PC functionality into something that runs on rechargeable batteries for just a few hundred dollars. Instead of placing the processing of applications on a desktop PC that's turned off, or a laptop that might be taken away, this puts processing for these new applications on the network itself.
Last month Intel's mobility chief Sean Maloney was in the hunt to head H-P, a job that eventually went to Mark Hurd of NCR. (Watch out. Dana is about to criticize a fellow Truly Handsome Man.)
But how well is Maloney doing his current job?
Intel's role in the development of Always On is crucial, and its strategy today seems muddled. It's not just its support for two different WiMax standards, and its delay in delivering fixed backhaul silicon while it prepares truly mobile solutions.
I'm more concerned with Maloney's failure to articulate a near-and-medium-term wireless platform story, one that tells vendors what they should sell today that will be useful tomorrow.
Intel seems more interested in desktops and today's applications than it is in the wireless networking platform and tomorrow's applications.
Lenin named his small movement the Bolsheviks, a word meaning majority. He called his majority opponents Mensheviks, a word meaning minority.
The point is that if one side is large and undisciplined while the other side is smaller but tightly disciplined, the smaller group can win a political struggle.
That seems to be the case with municipal wifi. It's an undeniable good everyone wants. It's relatively cheap to install and maintain. It should be a no-brainer.
But it's losing to telephone monopolies because of lax discipline.
Long story short I thought it would help if I described what might be a better plan for citywide WiFi. Apologies to those of you who have read this before.
The short answer is WiMax. The long version follows the break.
Corruption is a cost of doing business. Someone's already going to jail over graft on the Airport WiFi system. You think no one's going to take a dime here they don't deserve?
This is paid access. This is not going to benefit the poor people who supported this project. Philly is actually building a WiFi cloud that ISPs and others will re-sell.
Verizon will sabotage it. As we saw with CLECs in the late 1990s there are lots of ways an incumbent carrier can sabotage a competitor, simply by stalling cooperation. Verizon has every incentive to see this fail, and they're going to make sure it does.
Those are the obvious problems. But wait, there's more:
The war against 802.11 hotspots, which I predicted last week, has already begun.
I don't expect free access to survive it.
The fact is that a hotspot without registration allows hackers to insert viruses undetected, allows criminals to hack into databases undetected, and allows spammers to spam undetected.
The New York Times had a feature this weekend , picked up by the Financial Express, alleging half the crooks caught in a recent sweep dubbed Operation Firewall were using public hotspots.
A recent piece from the Medill News Service (my j-school alma mater), picked up by PC Advisor, suggested that people should never conduct personal business through a hotspot, for fear it is actually an "evil twin" set up by a hacker to grab passwords from the unwary. An IBM spokesman also detailed this scam for Newsfactor.
This summer will be the peak of the Voice Over IP (VOIP) boom. (The illustration, by the way, is from Poland. No, he doesn't look Polish.)
It's an easy prediction because Philips announced at CTIA a reference design for "converged handsets," with 802.11 and GSM or GPRS cellular in the same package.
We've seen the success of Vonage and Skype. We've seen the growth of 802.11 "hot spots" in hotels, airports, and on campuses. We've now seen the cellular industry adopt to VOIP. It's happy days.
Berresford is the FCC's senior antitrust lawyer and a professor at the right's favorite school, George Mason. He has power and the connections to turn his statements into policy.
So when he came out with a paper today about spectrum policy, it was bound to be read avidly.
In his paper Berresford favorably compares the law of land property to that of spectrum. He notes how property rights and spectrum rights are limited under the law, often in the same ways, and states that "efficiency" should be the watchword in spectrum policy.
We should know what we're in for when, in his first paragraph, he mischaracterizes the debate:
Debate rages about whether the allocation and management of the radio frequency spectrum should be mostly a political process, treating it as “The People’s Airwaves,” or mostly market-driven, treating it as private property.
That's not the debate. The debate boils down to science and markets. What treatment of spectrum best serves the market, that of a government-owned monopoly or a carefully-managed resource?
We haven't just "discovered" how to use vast new areas of spectrum in the last 20 years. We've learned a lot about how such spectrum can be re-used, again-and-again.
Thus the argument of property vs. commons isn't a left-right argument (as Berresford supposes in his introduction). It's an argument over science and efficiency.
And the plain fact is that the spectrum which is most efficiently used in this country, which makes the most money per hertz, by far, is the unlicensed spectrum.
Berresford ignores both the science and market forces behind this fact.
It's easy for someone to criticize Wind River's strategy as an attempt to maintain proprietary control in a world of open source, but the fact is there are opportunities here for the Always On world that need to be explained, and then seized.
Fact is Wind River's VxWorks is the leading RTOS out there. RTOS stands for Real Time Operating System, folks. An RTOS is used to make a device, not a system. You find RTOS's in things like your stereo, and your TV remote. What the device can do is strictly defined, and strictly limited. Your interaction with the device is also defined and limited.
An RTOS is not a robust, scalable, modular operating system like, say, Linux. And over the last few years, Wind River has been creeping into your world. VxWorks is used in most of your common WiFi gateways. This limits what they can do. They become "point" solutions. You can't run applications directly off a gateway, only off one of the PCs it's attached to.
This is true for hardware, for software, and for services. Future hardware designs must make it easy to connect, hands-free. Software must have intuitive user interfaces, as simple as speech. Services need to be spur-of-the-moment.
A lot of the mobile services I see today violate these principles big-time. They're based on Web interfaces, and thus have a limited time horizon. The key is to get inside the phone, so you're bought as soon as the customer thinks of buying.
Should WiFi cover every inch of ground or should it be concentrated where people congregate.
Today we have two designs in the news, one meeting each need.
From Steve Stroh comes the idea of smartBridges, a directional antenna providing enormous bandwidth, and backhaul, within a small defined area.
"There are many, many service providers that have very profitably deployed such a hybrid infrastructure - use Wi-Fi where it makes sense - where it can be highly localized and you can take advantage of higher power, more sensitive receiver, and directional antennas on an outdoor Access Point."
Sounds cool, but there's still a rival out there, Direct Sequance-Ultrawideband, pushed by the UWB Forum. The latter group has demonstrated things like home networks, while the former has pushed a Firewire replacement over a distance of 2 meters. (The illustration to the left is from Intel.)
So this is more than just a technical argument. The WiMedia folks see the technology as a Bluetooth replacement. The UWB Forum is aiming at the heart of local networking.
As it prepares for its developer forum this week, Intel faces an audience of bankers who have not lost faith in it, but who don't understand what it means by "platform."
Many different types of solutions go into creating an Always On world.
I’ve talked here often of medical applications for Always On, where you wear a monitor (or have it implanted) that connects to the network and can alert you (or others) to dangerous changes in your physical condition, thus saving your life.
I have also talked of inventory applications for Always On, in which RFID tags or bar codes give you a ready inventory of your stuff. This lets you, for instance, find your keys, or check the fridge to see what you need for tonight’s dinner.
But the low-hanging fruit lies in automation applications. CABA (it stands for Continental Automated Buildings Association) is one of the trade groups involved here. They work mainly with landlords who want to save money on utilities, provide security, and keep track of what’s happening in lots of space so as to minimize labor costs.
That's right, gang. The old joke from The Graduate is here again, aiming to drive silicon into the ground.
Nanomarkets, a market research outfit with a beat that looks like tons of fun from here (call me) has a $2,000 report out with a hockey stick chart for plastic semiconductors, estimating the market at $5.8 billion in 2009 and $23.5 billion three years after that.
Plastic electronics -- chips built on conductive polymers and flexible substrates, will be cheaper, take less power, and (obviously) be more flexible than silicon circuits. This makes them perfect for, say, mobile phones.
It will also bring a bunch of new suppliers to the electronics market, names like Dow Chemical, DuPont, Kodak, and Xerox, along with the usual suspects.
While SmartHome appears to be a general purpose gateway, the focus of Secured Digital is obviously security. The featured peripherals are cameras and biometric security devices. In these niches you need monitoring, fast response to alarms, and thus sales channels.
A PDF file describing the system shows it exhibiting most of the features of what I would call a residential Always On system. The central controller is seen controlling media, phones, water and heating systems.
Since the collapse of Lernout & Hauspie, voice has been diminished as a computer interface.
But it makes sense. It's hands-free. It requires training, meaning it brings some security with it by default. I continue to believe in it.
So does IBM.
Igor Jablatov is the man behind IBM's voice strategy. He's based in Charlotte, and has a blog, which mainly prints and links to stories and news release relating to VoiceXML. (Jablatov now heads the VoiceXML Forum.)
The Voice Extensible Markup Language brings voice into the Web standards area, and it's important for that reason. But what's more important is the extension of voice into specific vertical markets. IBM has started with things like cars and consumer electronics, and next plans a move into CRM.
These aren't the markets I would have chosen, but for now voice needs to choose markets based on their money making potential, nothing else. And I trust that IBM has done that kind of analysis here.
From Medgadget comes word that Always On was a theme of the Demo conference in Scottsdale, Arizona, last week, even if they didn't use the name.
One such exhibitor was Lusora, a San Francisco outfit that claimed to be in the security business, but also introduced a medical device.
It's all quite wonderful, but there is one big problem.
Standards.
Lusora's medical gadget uses Zigbee, and its hub, on the surface, looks proprietary, even though it's based on industry standards like WiFi and TCP/IP.
I could be wrong. I hope so. I've contacted their PR folks to see if they can be helpful. And I'm certain they can be.
Which sci-fi author did the best job of predicting what the 21st century would look like from the comfort of the 20th?
It wasn't Arthur C. Clarke. I still don't have my zero gravity toilet. It wasn't Isaac Asimov. Honda's Asimo is no Robbie. Allen Steele? No beamjacks in my world. Ray Bradbury? Larry Niven? Steven Barnes? Jerry Pournelle?
Wrong, wrong, and (sorry Jerry) wrong again. (But there are many centuries to go before your visions come up, so keep writing.)
It's William Gibson (right).
We live today in Gibson's Neuromancer. Cyberspace is everywhere, but so too are viruses. IBM notes they're appearing everywhere -- in our phones, in our cars -- and the people behind them are increasingly of very evil intent.
It's nice when someone in the "major media" gets the Always On vision, no matter how they get there.
The vision is simple. It's a wireless Internet platform. You get there by combining robust scalable PC applications with Internet connectivity and WiFi.
By Always-On, of course, he doesn't mean what we mean here. He's talking about background programs that run whenever the computer is being used, such as security programs.
Coffee compares these programs to your staff. They're going around your PC doing their thing, and that's useful. But if employees aren't prepared to be interrupted by a higher-priority task, if they won't take direction in other words, you fire them.
The significance of WiFi-cellular roaming lies in Always On applications.
Think about it. Cellular channels are relatively low in bandwidth, WiFi channels are high in bandwidth.
Now, you're wearing an application, like a heart monitor. When you're at home, or in your office, this thing can be generating, and immediately disgorging, tons and tons of data, detailed stuff that may be fun for your doctor to analyze later.
Same thing, really. In discussing this with Motorola spokesman Paul Alfieri recently, once thing was crystal clear.
They get it.
A few quick corrections to our most recent post on the MS1000. You can build applications that run in flash on the device, that don't depend on the PC. The device is wide-open to applications through its OSGI-based Linux software core.
This is big stuff, a real "killer app." I lost my best teacher ever, Dick Schwarzlose, to a heart attack last year, an attack that could have at least been treated had his doctor known it was coming.
Hundreds of thousands of men and women die each year from sudden heart attacks which are not detected, but most of these people were known to be at risk based on factors like their blood pressure and cholesterol levels. (Heck, I'm at risk for those reasons.)
If this solution can be productized and delivered with, say, the client monitor and communications hidden inside an Under Armour shirt (wicks away the sweat and looks wicked cool), then many lives can be saved, and many middle-aged men can look marvelous at the same time.
Along with all their other implications, the mass adoption of mobile phones represents the first step in the single-chip era.
If you look inside the guts of your phone you are unlikely to find a big honking circuit board. (The circuit board illustration is from Sciencetechnologyresources.com.) Instead you will find one, two or three single chips performing major functions in an integrated way.
This is happening across-the-board in technology. We've gone from circuit boards in the 1980s to modules in the 1990s, to single chips. Just as early IBM PC add-in board producers created "multi-function cards" to assure a price worthy of retail distribution 20 years ago, so chip makers today put multiple functions on many chips, creating entire systems no bigger than a finger-nail.
The idea is that you sync the phone to your home using a verison of the old Palm cradle, then control home automation applications remotely using the phone.
For the last year I've been harping here on the subject of Always On.
The idea is that you have a wireless network based on a scalable, robust operating system that can power real, extensible applications for home automation, security, medical monitoring, home inventory, and more.
As I wrote I often came back to Motorola and its CEO, Ed Zander. They would be the perfect outfit to do this, I wrote.
Little did I know (until now) but they did. A year ago.
It's called the MS1000.
The product was introduced at last year's CES, and re-introduced at various vertical market shows during the year. It's based on Linux, responds to OSGi standards, and creates an 802.11g network on which applications can then be built.
At this year's CES show, Motorola is pushing a home security solution based on the device, with 10 new peripherals like cameras and motion sensors that can be easily set-up with the network in place, along with a service offering called ShellGenie.
Previously the company bought Premise, which has been involved in IP-based home control since 1999, and pushed a version of the same thing called the Media Station for moving entertainment around the home.
What should Motorola do now? Well, the platform is pretty dependent on having a home PC. The MS1000 could use space for slots so needed programs could be added as program modules. They need to look at medical and home inventory markets, not just entertainment and security.
But they've made an excellent start. And from here on out everyone else is playing catch-up.
I first wrote about Zigbee about seven months ago, with some enthusiasm, so you're forgiven if you've forgotten. But Zigbee is a low bitrate, very low power standard for powering what I call Always-on sensors.
A Czech outfit called Bladox has developed a system called Turbo SIM that basically lets you turn any obsolete mobile phone into an Always-On sensor.
For example. You can turn your dead mobile into a car alarm. No sensors, no wires. Just plug this card in place of the dead phone's SIM, in the back of the phone. If the alarm detects tilting or other movement without being disabled, it doesn't just wake the neighbors -- it sends an SMS message to any phone number you designate.
The BBC has a piece today showing how the World of Always On could be invisible, worn instead of held.
We've already seen undershirts embedded with medical sensors. But Ian Pearson predicts we're going to move, over the next 10 years, to a world of devices imprinted on the skin.
If you're a kid it's pretty exciting. But I've seen this parade before, many times. I can enjoy your pleasure, but that doesn't mean I'm not going to join "Santa" later at our favorite bar for a few pops and expect he'll make me pick up the tab.
The big truth is that it's now pretty trivial to put a TV tuner into a mobile phone. Yes, the days of Dick Tracy's wristwatch TV are really here. (Image from USA Today.)
Of course you remember what Tracy used the technology for? To see his boss while he was talking to him. I did the same thing at the 1964 World's Fair, where AT&T had a videophone demonstration. It was no big deal then, and it's no big deal now.
Entrepreneurs like Blake Krikorian of Sling Media, who is profiled in the Business Week story above, think you'll use the capability to watch their streams. Most people watch brief snippets of TV anyway, not whole shows, he says. No time. So offer them such streams for the moments where they're standing in line at the Airport, or waiting for a meeting to start, and they will pay through the nose for them.
Philadelphians are celebrating an agreement with Verizon which, they say, allows them to offer a citywide Wi-Fi network despite a law, signed (shamefully) by Governor Ed Rendell yesterday, aimed at stopping the municipal WiFi movement.
HB 30 prohibits the state or any municipality (or any municipally owned or operated entity) from providing any sort of telecom or broadband service for any kind of remuneration. The bill grandfathers any existing systems, tho, so no one will get cut off.
While searching for stories on Open Source (for our new ZDNet blog) I came upon a conference in New Delhi where 200 engineers from around Asia shared experiences on speech synthesis and recognition.
It was the speech of Dayanidhi Maran (right), India's IT Minister, that first attracted my attention. He wanted applications in all India's languages, not just the "majors" like Hindi, Tamil, and English. And he wanted this to be open source.
But it was the conference itself I found most fascinating.
The physics folks at UT-Dallas have won nearly $1 million to study the idea of making every radio part of a giant "mesh network." (That's the UT-Dallas seal over there, with its motto of discipline, civilization, and absolute rule by a self-appointed elite.)
The idea is that mobile phones and laptops could act as relay points for other users' transmissions, creating what the physicists call "cooperative wireless networks" but which might best be termed a "giant mesh."
The idea, which is valid, is that through blogging ordinary communication becomes content. I know this is true because my own newsletter, a-clue.com, has been losing readers ever since I started blogging here. It's not just that readers prefer getting my thoughts through the blog instead of e-mail. It's that the one-week lag between my writing and your reading is eliminated by blogging. You're not just an audience here, you're practically reading over my shoulder as I type.
My vision of Always-On has always been based on the idea that applications would live on the wireless network in your home. A Wi-Fi set-up has both the bandwidth and computing power needed to handle several such applications.
But early medical applications move with you. And thus they ride on the mobile network.
The Mobile-Technology Weblog (the picture is taken from that blog) has an example today from Korea. (It's a small world. The author lives in Munich, he's represented for ads by a British firm, and here I am blogging about it from Atlanta, Georgia.)
The hook is "here's the fat police," but the story is that there's a Samsung SPH-E3330 mobile that can measure your body fat level.
If you're to take Always-On applications into the world with you, they have to be fashionable. They have to look smart. It would be very nice if they were machine washable.
With all the hoop-de-doo over mobile phones being bad for you, it makes sense that a wireless headset, connected to your phone or (maybe even) your iPod, would make sense.
I'm sure it was written before the Red Sox played St. Louis last night, because otherwise the author's fingers would have been shaking too much for the story to be as clear as it is.
But Robert Weisman, in his zeal to give props to a local vendor, falls down on his understanding of what Ember is doing and how Always-On applications must reach the market.
Ember, as regular readers of this blog know, uses Zigbee. Zigbee is a very low-power, relatively low-bandwidth technology. Zigbee, known the IEEE geeks as 802.15.4. may or not prove relevant as Always-On develops. For now its only proven applications are in factories, where it can be used to control things like oil refineries.
In other words, it's very expensive, and takes a very long time, to develop a Zigbee application, although the results can be very powerful for your bottom line.
Unfortunately that means Zigbee is not the "Internet of Things" Weisman touts it as. It could become that, but first it must clear some big challenges:
The deal, announced at the CTIA's San Francisco conference, is that Clearwire will use Intel WiMax gear and, in exchange, Intel will invest in Clearwire. Essentially Clearwire gets the gear for stock.
I mean, really. An underpowered PC in your Hummer, with Bluetooth and a GPS receiver? Peter Wengert of Microsoft told News.Com about applications like finding cheap gas, because the GPS receiver would have your location.
But if you just look under the hood here, you're going to find something really cool.
The shorts monitor your heart and warn emergency services of a change in the heart rhythm. It can make this connection through a built-in GPS system with mobile phone. And for you ladies there's the Electric Bra (pictured).
This is similar to the MyHeart project, which is putting the sensors in shirts.
It's the kind of medical application of Always-On I've been looking to see for some time. But I've got a few problems with it:
David Peskowitz over at The Feature wrote yesterday about how Always-On medical applications might be implanted in your skin.
He, however, was talking about a specific product. UbiMon, from UbiCare, a project of the Imperial College in London, is precisely the system I've described before, measuring heart function so that a "critical event" can be detected before it happens and kept from becoming a fatal event. (Image is from UbiCare.)
Usually you think of a physical location, a building (or a piece of one), in which you find merchandise and a cash register.
But is that the only store? Isn't Amazon.Com a store? Amazon consists of warehouses and databases and delivery vehicles. That model is actually 150 years old, dating from the Sears or Montgomery Ward "Wish Books" of the 19th century. (This here store is actually just a cute piece of bric-a-brac, available here from Aspencountry.com, a Web site.)
But is that the only store?
No. Now a store can be anywhere thanks to Always-On technologies (and Verifone, which has enabled its use for Point of Sale transactions).
With Microsoft cutting back, with Intel dropping product lines and with entrepreneurs stuffed by a fearful capital market, a journalist is hard-pressed to find a brave company taking big risks in the face of real market opposition.
IBM is putting $250 million into a new division that will push RFID and other Always-On technologies. The division's name is the Sensor and Actuator Solutions. The division will become part of IBM's Pervasive Computing business.
Now when anyone tells you Always-On is a myth, you send them to Armonk.
Carnegie-Mellon and Cal-Berkeley are going to spend $1 million in the government's money over the next three years trying to create a general speech recognition chip for the market.
When they succeed, and I have no doubt they will succeed, it will be a true revolution.
IBM has decided to make some of its key speech technology open source. (That's an old Kurzweil AI poster found at Ethicon, a Johnson & Johnson company that could be important in what follows after the jump.)
This is great news, and they're doing it for all the right reasons. The following quote, from a New York Times story on the decision, could have been written by Linus Torvalds himself:
"We're trying to spur the industry around open standards to get more and more speech application development," said Steven A. Mills, the senior vice president in charge of I.B.M.'s software business. "Our code contribution is about getting that ecosystem going. If that happens, we think it will bring more business opportunities to I.B.M."
Broadcom absolutely rocks in the Wi-Fi chip market. It is constantly ahead of the curve. It has great relationships with OEMs and product marketers. TI and Intel look good, but no one plays the inside game as well as Broadcom, trust me.
And if Broadcom is catching a cold, then everyone else has pneumonia.
Over at C|Net, Michael Kanellos has one of the stupidest leads I have ever read.
"When it comes to radio frequency identification tags for humans, the people have spoken.
They hate it."
Well of course they do. When you sell it with scare, when you don't push the benefits, when you fail to advocate for the necessary legal doctrine (the data you create belongs to you), when every other word you write is about "big brother watching you," then of course everyone who writes into the newspaper is going to come off as a dyed-in-the-wool card-carrying Luddite.
Each time a member of my family needs any type of medical care I'm frustrated. (The image is from the good people at Tympanitis.com, fighting the good fight against otitis media, and is used here completely out of context.)
Mainly I'm frustrated by paperwork and waste. I'm frustrated with filling out forms, and dealing with bureaucratic clerks.
I'm frustrated because it's all so expensive, and all so unnecessary. There are technologies available today that will cure this problem, solutions that can be implemented now.
If big insurers gave clients smart cards, and insisted that all members of their networks take smart card readers, it would be a big start. Entire medical histories, and biometric identification, could be mounted on the cards cheaply. Read the card and the doctor's network automatically knows what it needs to do, about the patient and the billing.
This means you can have a device that supports multiple networks, and the chip will tune itself to whatever is available.
This is one step toward Always-On, but others are needed. We need agile radio chips that not only tune to the right frequency, but deliver just the right power to reach another chip in the network, and no more. We need agile radios whose antennas can point to other agile radios and not spread radio waves in every direction. When we have that we can "mold" local wireless networks to just the coverage area they need to have, no more.
While there are lots of high frequency bands in which WiMax could live, the inescapable fact is that the lower your frequency the farther your waves can travel. That's why AM stations can be heard across the country (when conditions are right) while FM stations have trouble being heard across town.
Intel executive vice president Sean Maloney (above, from the Intel site) is lobbying China, the UK and the U.S. to open up space in the 700 MHz band, frequencies UHF TV stations will be abandoning as they move to digital broadcasting, for unlicensed use as WiMax transmission bands.
A decade ago, with my alma mater still basking in the glow of our great "Buckyball" discovery, I challenged our President with an ethical question. (This Buckyball was caged at the-scientist.com back in 1997.)
He dismissed it, and I felt the whole room cool toward me. (A few more glasses of wine solved that.)
Well, the debate is finally upon us. How else do you explain such contradictory headlines for the same report:
RFID is a vital Always-On technology. With RFID on your stuff, you can have complete control of your personal inventory through a wireless network.
But RFID markets don't work that way. RFID applications will come at us through business. The initial demand will come from big merchants and governments seeking tighter control of their huge inventories, not from you and I seeking some control over our small inventories.
And, to me, that's fine. That's how computing markets work. You go from big applications to small, from business to consumer. So what is this debate really about?
The New York Times is currently featuring a story about sensor networks in the wild.
What they've got, in fact, are a bunch of Always-On testbeds.
For instance. A sensor network checks the condition of grapes and the soil they're growing in, letting the grower know when to water, and when to harvest to get the best wine.
That is a great Always-On application. How many of those can you sell, maybe 1,000?
But what if, instead of building a big sensor network you built a little one, along with a simple drip irrigation system.
Know how many of those you can sell? You can sell millions. Start with golf courses, move on to schools and office parks, and watch sales zoom with water prices. Just follow Moore's Law and within two years you've got a consumer product, the whole set-up available at a do it yourself shop for maybe $300.
You ever leave a pot of coffee on the counter too long? I have. After several days it starts tasting really funky, and these nasty white organisms start partying on the top of it. (BREW logo from Vayusphere.)
Well, that's what seems to be happening to Qualcomm's BREW development environment, in which Verizon is demanding all applications on its cellular network be written. There's no circulation in a proprietary environment . If the creator doesn't apply regular heat (and risk that nasty, metallic taste) things are going to get funky fast.
The history of computing is defined by entrepreneurs.
First came the Tom Watsons, Senior (1874 - 1956) (pictured to the right) and Junior (1914 - 1993) (below left, both from the Watson Labs' official site).
But Watson Jr. retired in 1971 and IBM became lazy. They defined themselves by their sales relationships with the Fortune 500, which used Big Iron. Under John Akers IBM became vulnerable to the first "kid with a Clue" who could see that the "platform story" IBM depended upon was tired, not wired.
That kid, of course, was Bill Gates. He saw that the PC would be the new platform, and that its foundation wasn't in the hardware, but in the operating system. By first selling IBM on MS-DOS, then outmanuevering IBM on Microsoft Windows, William H. Gates III became the true heir to the Watsons' legacy.
Yet computing's evolution has continued to accelerate. Microsoft had little play in local networking, which defined the 1980s, and struggled to get on top of the Internet, which defined the 1990s. (The Gates caricature is from a fine Italian artist. Visit his site.)
Now history is about to throw another threat Microsoft's way. Ironically, when he should be at his most comfortable, Bill Gates is now at his most vulnerable.
One difference between my youth and maturity is that, in my youth, the future was seen as positive while today we look upon it with fear. (This might be why the ratings for Star Trek shows have fallen so far.) (The image is from a Star Trek fan site.)
I don't want to be a Pollyanna here. There are dangerous implications to Always On applications. Unless the data we create is acknowledged to be ours by the government, and unless that ownership is respected, it could become an Orwellian nightmare.
But instead of emphasizing this simple solution, a new IEEE magazine on the technology goes on-and-on about danger, danger, danger.
All the accessible stories in this series -- the news, the analysis, even the fiction, state unequivocally that privacy is impossible, privacy won't exist, privacy is a hopeless goal.
A few years ago, the fuel cell market was aimed at the market's high-end, providing back-up power to electric utilities and phone companies, especially in places where anti-pollution laws might prohibit other types of generators. (Illustration from DCViews.Com.)
Now it's moving quickly into the replacement of batteries in laptop-sized and smaller devices. Fuel cells last longer between charges than batteries, and they can be recharged with new fuel rather than new batteries, fuel that might be available where batteries are not.
Now Toshiba has entered the technology side of this market. That is the most promising point of the story, not the specifics of what they're offering. (The device announced this week is reportedly much smaller than the one shown in the picture.)
The fact is this is Toshiba, this is a big company that doesn't do things halfway. They see opportunity here. So should you.
The early days of any technology are nerve racking. (That's the logo of Zigbee chip-maker Airbee Wireless.)
It's a dance of small companies looking for contracts that can become lifelines to survival, larger companies making big claims they may be unable to meet, and the nagging fear on the part of everyone they're about to become obsolete.
In the case of wireless sensor networks, I call this the Zigbee Dance.
There will be many Always-On application spaces. There will medical applications, there will be home inventory applications, there will be home automation applications.
Once you see the wireless network as your platform, rather than a specific hard-wired device like the PC, all sorts of things become possible.
A few weeks ago I wrote glowingly of Zigbee and I still like it. Sensors that last a year and pass data on 802.11 frequencies? What's not to like.
But this is an open standard, and most of the firms working on it haven't delivered product. Those that seem closest to delivery, like Ember, seem focused on the industrial automation market.
All of which means there remains opportunity for a proprietary standard, if you deliver product supporting it, and if you know your niche.
That's where Smarthome is going with a new standard it calls Insteon.
Here's a nice Always-On application, found (once again) by Roland Piquepaille. British scientists have embedded a sensor network inside a Norwegian glacier so they can learn how it evolves.
The sensors are quite small (this one, from Piquepaille's site, is magnified). Installation is as easy as tossing them about. Such sensors configure their own network. (Here's the experiment's official site.)
Today's science is going to be tomorrow's application, once an engineer and a salesman get together in an airport, write a business plan, and then take a meeting with a venture capitalist.
I have wondered for months why motes, sensor networks, Zigbee, and nanotechnology don't draw more press coverage. Some of these markets are worth billions, right now. And they are going to revolutionize the world. Always-On will come.
Every industry will follow the money. Its events will be held where its markets are, or where its developers are. And the markets always have an opportunity to lure developers to them, through those events.
Thus, next month represents a great opportunity for Detroit in the history of Always-On. (This lovely photo of downtown Detroit, a view few people today associate with that great city, is from American-Products Publishing, which is based in Oregon.)
The Zigbee Alliance meeting in Seattle recently showed just how far this technology is from the computing mainstream.
A very small company called Ember was basically able to "take over."
Ember raised itself to "promoter" status within the alliance, which now has 70 members. Ember sponsored the Seattle meeting. Ember delivered the main address.
Within a few years there will be more "things" (intelligent devices) on the Internet than people.
This is not a bad thing. In fact it is a very good thing. (The illustration is from a Forbes article on RFID, reprinted at Mindfully.Org.)
Here's one example. Philips is developing a Zigbee-based garment that can monitor your physical condition, diagnose problems, and alert you or a doctor if something goes wrong.
That's the best conclusion to be drawn from a visit to Ember Corp., a leader in the field here. (The photo to the left is from Ember's Web site, specifically its page on industrial automation.)
There's always talk of politicians or pundits contradicting themselves. But what most people don't know is that companies do this all the time.
Take the case of Philips, and the technology called Zigbee.
Zigbee is a standard for low bit-rate, very low-power transmissions that could happen in the 802.11 frequency space. Such "sensor networks" would be a key component in the World of Always-On. (The illustration of the Zigbee Alliance logo is from Figure8Wireless, a Zigbee proponent.)
Philips is an original member of the Zigbee Alliance, which gets together in Seattle later this month.
What's their stand? Turns out it depends on who you talk to.
Having finished my work on residential gateways, I am more convinced than ever there is a huge opportunity here for someone. (Illustration from Parc.com.)
It won't be the phone company. Consumers don't trust the telcos and there is no law requiring that the ISP serving a gateway must be the one providing your local access service.
It may not be a router company either. There are big problems with present wireless LANs. Their reach is spotty, and boosting their signal only means the neighbors can get on your network.
The big opportunity lies in the middle. The big unmet need is for security.
Members of the Zigbee Alliance will be in Seattle May 17-21 for their annual meeting (with a press event May 19 at the Hotel Monaco), where they hope to address limitations in the standard with a new effort. (The picture, by the way, is the cover from a 2001 market research report on Zigbee by Mareca Hatler and Michael Ritter.)
As Atmel's Zigbee page makes clear the present version of the standard, dubbed 802.15.4 by the IEEE, calls for devices that can shoot data from 1-75 meters and run for 100-1,000 days.
It sounds like a lot, but with last year's chip technology, it wasn't good enough to cause a big market splash. Philips appears to have backed away from Zigbee entirely, moving participation from its semiconductor group to its lighting group.
The answer, it seems, is a new, improved standard, 802.15.4a.
What made the Internet work was the fact that it was a royalty-free, open standard. You could build on it, but there was no admission price for using it, so everyone did. (Illustration from the Daily Telegraph.)
But since the emergence of the Internet, in the mid-1990s, we've had the copyright wars and the rush to patent everything in sight. This trend is going to slow western companies going forward, and give the future (in my view) to Asian companies that innovate first and call the lawyers only after the market is won.
Zigbee may be the most important global standard effort of our time. (The picture is from Motorola, a major supporter of Zigbee technology.)
That's because it sits at the "sweet spot" of Always-On, where radios and cheap chips can create revolutions.
As a PDF overview of the standard makes clear, Zigbee is designed as a low power, low bandwidth, low range radio standard. Doesn't sound like much, until you realize what this enables.
It means you can have single-chip radio computers that transmit data when necessary, and run for years. The single-chip could be a medical monitor, or it could be an enviromental monitor. Combine this with the price-performance breakthroughs we're seeing in biochips and we have the Always-On revolution.
Linksys is a very important name in the World of Always-On.
Linksys, now a unit of Cisco, makes home routers, and it dominates the retail channel for wireless networking.
Since acquiring the company 10 months ago Cisco has mainly left it alone, just sending one of its executives down to Irvine to teach Cisco's ways to Linksys and learn Linksys' ways for Cisco. Founder Victor Tsao remains in charge.
Talk to many Linksys competitors, like Siemens or 2Wire or Netopia, and you're going to hear a lot about how things are done. You're going to hear a lot of details about wireless networking technologies, 802.11 this-and-that, about antennas and radios, all the kinds of stuff analysts like me want to hear about.
A residential gateway is probably going to define how you get your Internet service in coming years.
Why buy a modem, a router, a switch, and a Wi-Fi set-up when can get them all at once, probably free?
This makes gateways important. Since the Wi-Fi set-up is in there too, they're also going to define your Local Area Network.
And the LAN is where your Always-On applications will live.
So, yeah, gateways matter. It's a market worth studying.
Whose gear do I like? Everyone's. And no one's. There's a lot of value here, but there's going to be a lot more. I don't want to be wedded to something that's going to be obsolete before my service contract runs out.
I want to date my gateway, not marry it. At least for now.
A gateway has an Internet connection on one side, and some sort of LAN connection on the other. (Usually it includes a wireless LAN.) It's a modem, it's a router, it's a switch. It's pretty cool. (The picture is from Johnkdavis.net.)
But what is it in terms of the market? How will you get it?
Intel announced what many considered a "blow-out" quarter, with sales up 20% and net income nearly doubling.
Imagine what they could do without one hand tied behind their back. (The image is from a 1995 paper on Internet payment systems by Michael Pierce of Trinity College, in Ireland.)
In Intel's case, the hand behind its back is communications. Intel dominates basic computing, although its lead in servers is shaky enough that it needs promotions. But in the chips that run cell phones, or routers, or any device based on communications, Intel is an also-ran.
What gives this such resonance (that's a joke son, resonance) as a buzzword is its deep, true meaning.
We're talking here about taking the human intelligence of sharing spectrum and putting that into radios as machine intelligence.
A cognitive radio can sense its environment and location and then alter its power, frequency, modulation and other parameters so as to dynamically reuse
available spectrum.
Let's see what that means in products you use every day:
The trouble with Intel is it thinks it's in the chip business. (Make this image your wallpaper here.)
It's not.
Intel is in the platform business. Intel succeeds when its chips become an open platform for innovation. Microsoft did this for Intel in the 1980s and 1990s. As a result Intel dominates the PC space.
But in the wireless world Intel has lost the plot.
Bell Canada is now offering its ISP customers wireless gateways from Siemens, along with back-end software called Tango that diagnoses trouble and updates the firmware. (Image of the gateway from ZDNet Germany.)
About 20 years ago, Dave Waks was one of the founders of Prodigy, the NAPLPS-based online service that tried to give the online world graphics a decade before the Web was spun.
For those wondering what happened to this pioneer, have no worries. Dave is married a very nice former-AT&T executive, Sandy Teger, and now they work together, writing and pontificating about home gateways.
Since I just happen to be thinking a lot about gateways right now, I got a chance to talk with them.
Few people really noticed it, but Microsoft yesterday made a formal announcement for a new version of Windows CE, designed for "embedded" applications. (To learn more about using Windows CE features, consider buying this book.)
In the World of Always-On, however, this is big stuff.
It's big stuff because Always-On applications will be essentially hardware. They will run in the background. Their output will be in the form of real-time warnings, or actions. Most clients, and many servers, in this world will need and use only the embedded version of an operating system.
One of the differences between this world of journalism and the real world of technology is the concept of the deadline. (Image from Civilwaralbum.com.)
In journalism a deadline is just what it implies. If you don't get the work done by the deadline, it's dead. Maybe your career is dead.
The term originated aboug 120 miles from where I live, at Andersonville, outside Americus. Back during what I prefer to call "The Recent Unpleasantness" and most call the Civil War, Andersonville was the Confederacy's most notorious prison camp. It became a mini-Auschwitz, with soldiers starving because they could neither be exchanged nor fed.
The guards erected a line of fence in front of the stockade, and shot to kill any prisoner who crossed the line. Hence the term, deadline.
Opera, which just went public in Norway, has announced a "talking browser," based in IBM technology.
This is important stuff, and not just for my blind mother.
What's being provided here is an interface, between the Internet and voice, which can be embedded in other devices, not just PCs.
Opera and IBM are talking up the value of putting this into cell phones, which is cool. But what about putting this into Always-On networks? Now you can simply write interface routines in HTML, and interact with those routines from wherever you are in the house.
In covering the CTIA show in Atlanta this week, I sometimes found myself boiling with rage, and not knowing the reason.
I have learned to listen to this rage, and to keep it inside until I find the reason.
I found the reason while talking to Andrew Bud. Bud is an Englishman running Mblox, an Atlanta-based company (they're a few miles southeast of the Atlanta Airport) bringing Premium SMS billing to the U.S. market.
What Premium SMS does is bring indirect sales to the cellular data market. That's a key entrepreneurial idea. It's an idea being fiercely resisted by most U.S. carriers, which is why the market for cellular data is so explosive in Europe and Asia, why so many new entrepreneurs are coming up in that world, and why the U.S. is behind in the biggest opportunity of this decade.
FCC chairman Michael Powell sounded unusually intelligent during his CTIA keynote. Asked by new CTIA chairman Steve Largent about Voice Over IP, he riffed toward wisdom.
“It’s short-sighted to see Voice Over IP as another way to do something,” he said. “When you see Voice Over IP as an Internet appliation you see both the potential and the challenge. The applications change when you move from selling a service to software.”
In publishing this piece I think I learned where Powell came up with this idea. It might have been here , visiting the Wherify booth at a recent trade show.
At CTIA Powell said that, in fitting voice into a 3G data infrastructure, the wireless guys are thinking about this more than the wireline guys, but they’re still building a network utility, not applications. And complex applications demand a more complex sale than you can do on a TV spot.
The cellular world is slowly being dragged, sometimes kicking-and-screaming, into the computing mainstream. Always-On applications are playing a big part.
I saw that at the small booth of Wherify. This company makes strap-on GPS devices. The sales material in their booth was filled with kids on skateboards whose mommies worry about them. But when I sat down to talk, CEO Tim Neher admitted that the message didn’t match the market.
The company has been deluged with requests from the families of Alzheimer’s patients, he said. In response the second version of the company’s product strips away the fashion statements.
It’s a tiny box, the size of the chewing gum samples you get from cheap houses at Halloween. He noted that one woman sewed a false pocket onto her father’s trousers, and talked movingly of a patient who was traced across the LA bus system from Hollywood to Santa Monica.
I suggested glueing a pin to the thing, and mentioned a company that has turned walkie-talkies into Star Trek-style communicators, worn on the front of the shirt. That sounded like a good idea to him.
Dean's Law holds that, at some point, even the up-take on a wildly popular product will slow. A bandwagon rolls and grows, it loses its intimacy and immediacy, and suddenly there's resistance to it. Getting past that resistance requires, not more noise, but more intimacy.
Under Dean's Law, the S-curve of mass-market acceptance suddenly has a break in it, somewhere near the middle, and instead of sweeping on ahead it slows. That's because the market isn't that wide an ocean. It is more like a swimming pool. You can start a set of waves, but those waves cascade against the far wall, they come back and cancel out momentum.
In Iowa, Howard Dean saw just what Pat Robertson and Steve Forbes had seen before him. In order to "get over the hump," you have to scale down in the foreground, actually seem smaller, while building your real infrastructure below the Web.
How does this apply to broadband? Click and find out.
An Always-On home has both a wireless LAN on the inside and fast Internet connectivity on the outside.
So it would seem, on the surface, that a residential gateway (like the 2Wire unit to the left) would be the best route to Always-On. A gateway combines three important elements -- a modem, a router, and a firewall. And its cost can be subsidized by the ISP.
But it turns out there are two big problems with this approach, to wit:
People don't trust their ISPs.
ISPs don't design gateways with the local network in mind.
Ronald Schmeltzer of ZapThink has a piece out calling for the creation of "rich clients" to serve web services.
It's an interesting point. Where you put required software is important. But he's thinking of it in terms of software living either on a client PC or a Web server PC.
I think that's a limited view.
The most important place for applications to live in the future will be firmware. That is, they will be burned into silicon, shipped as boards, that either act as discrete devices or are added to them.
It wasn't by accident that a Netopia gateway is featured on the item (below) about cell phone business models.
Netopia has done more to be ready for this than any other company.
They have methodically worked with phone companies to sell modems, then routers, so they know how deals get done.
They have worked assiduously on the technical hurdles, both from the phone side (making modems work with central offices) and on the network side (getting rid of dead spots in wireless networks.)
They announced a decent product that seems to meet everyone's goals in December at WiFi Planet, and they've been working hard to create "up-sales" that carriers can use to lock-in customers.
In other words, they have done everything that we expect to see before this cell phone model can be applied to telephony.
CTIA is coming to Atlanta in a few weeks, and we haven't partied here since 1999. (I found this CTIA logo on a Taiwanese page selling trips to the show.)
The parties are going to be big and glorious. The liquor will flow. I expect to eat big shrimp in cocktail sauce. I have little doubt that the tchotscke (little gift) pickings will be lush. There will be toys, candy, maybe even t-shirts.
That's because the cell phone industry has figured out a business model that works. They give consumers something valuable, the cell phone, then charge them to use it.
Of course, it's not that simple behind the scenes. Even a "giveaway" phone costs money. What they're doing is charging you roughly $10/month for a $200 (retail) item in exchange for a 2-3 year service contract.
But the secret of the model goes deeper than that. The companies are balancing the sophistication and cost of the phone against the "billable events" it might reasonably generate. And they're giving consumers the power to make a higher-value choice, offering a range of models that do more than their giveaway model.
This has everything to do with the World of Always-On.
The distrust between privacy advocates and big business is creating a solution where there's no problem, as RSA begins offering technology to disable RFID tags. (The picture is from a California medical group offering patients information on medicine.)
As The Register reports, the Blocker Tag is a cover applied to a passive RFID tag at a register that keeps the tag on an item from signaling to a radio. RSA demonstrated this on drugs. The prescription bottle tag couldn't be read after the blocker was applied.
Kleinrock, now on the UCLA faculty (the picture is from his home page there) proclaims himself "father of the Internet technology." He pioneered packet switching while at MIT, and a decade later his host at UCLA became the first node on the Internet.
The great thing about being a great company is you can make mistakes and get away with them.
Microsoft makes lots of mistakes. Think "Bob." (The illustration to the left is from a Toastytech tribute to Bob, by Nathan Lineback.) IBM has made mistakes, the most notable being losing control of PC operating systems.
Intel has also made mistakes, especially in the area of communications. And, at least in public, it's compounding that mistake this week.
Its home networking concepts are smart, but they're being packaged stupidly.
While most makers of access points are in a standards race -- 802.11a, then g, then a+g -- a little outfit called Firetide is beating them to the punch.
The reason, in a word, is that Firetide is not selling access points, but mesh networks.
For customers, the difference between a mesh and a set of access points is like night-and-day. With an access point, coverage goes in a circle from that point, and you need a site survey to know how many points you need, and where you need them. With a mesh, on the other hand, you just keep adding nodes until you get coverage, and the nodes configure themselves.
Paul Allen's latest start-up, FlipstartPC (from which the image at right is taken) has delivered its first product, which The Register mistakenly refers to as an "Always-On" PC.
I say the reference is a mistake although I know they will disagree. I'm defining Always-On as one thing, they're defining it (in this case) as simply something that doesn't really turn-off -- it just goes to sleep when you close the lid.
Be that as it may, when I first read the news I wanted one. Ever since I defined "Dana's Iron Law of Laptops" -- an ounce on the desk is a pound in my hand -- I've been looking for a full-featured (that means it has a keyboard) PC that "gets it."
Dana Blankenhorn has been a business journalist for over 25 years and has covered the online world professionally since 1985. He founded the "Interactive Age Daily" for CMP Media, and has written for the Chicago Tribune, Advertising Age, and dozens of other publications over the years.
Video is NOT the future of the Web. (This picture, by the way, comes from 
When Paul Otellini was named the 
The Always On medical market 
America's biggest tech companies are focused today on the problem of creating, not technologies, but platforms.
The International Telecommunications Union has released a full report on 
There are cell phones, there are WiFi phones, there are cordless phones, and there are VOIP phones. But never the twain shall meet.
The folks at 
There's a lot of hyperbole there. (Patrick Henry, right, was nothing if not hyperbolic.)
Deals like the 
The
The best way to kill a promising technology is to argue about it in standards bodies.
The movement of network boundaries ties together all the trends of the present time.
Every decade of computing technology can be summarized fairly simply. (That's an 
Intel holds the telecommunications balance of power in its hand.
The question of Wi-Fi and real estate is about to come to a head, at Boston's Logan Airport. (Picture from 
I'm a big fan of both 
Blogging is filled with comings-and-goings. Mostly comings these days.
It's nice when "real" (paid) market analysts agree with one of your premises. Especially when it's a key premise to you, as Always On is to me. (This is advertised as an Always On Server, 
The Supreme Court has decided that cable networks, created under government franchises, under monopoly conditions, are entirely the property of their corporate owners who 
I spent last week in Texas, dependent on free WiFi hotspots, and I learned a powerful lesson.
Despite what 
One reason I (
Sometimes a negative response comes in here that's so well thought out and cogent that I just need to share it.
Why hasn't the World of Always On arrived?
Despite his ponytail and his sometimes counter-cultural language, despite being what I like to call a Truly Handsome Man (it's a brighter term for bald, people) Ted Waitt was always a follower, not a leader. (The picture is from 
I didn't blog much yesterday because I was researching the state of play in Always On. (The illustration is from 
The political battle over WiFi shapes up as a classic match between private interests and the commons.
The best way to understand the future is to look into how chips are changing.
There are two types of chips key to the Always On world.
Critics of my Always On rants (and I know they get tiresome) see the expense of radio-equipped sensor chips as a stumbling block.
The coming issue of Business Week features 
Last month Intel's mobility chief 
Lenin named his small movement the Bolsheviks, a word meaning majority. He called his majority opponents Mensheviks, a word meaning minority.
I am a big supporter of free WiFi. But 
The war against 802.11 hotspots, which I 
This summer will be the peak of the Voice Over IP (VOIP) boom. (The illustration, by the way, is 
When
I don't always agree with Nicolas Negroponte (right), but he made a point in Korea recently that really makes sense. 
Should WiFi cover every inch of ground or should it be concentrated where people congregate.
Failure to define a single standard for Ultrawideband is killing the technology. So say the experts.
As it prepares for 
Many different types of solutions go into creating an Always On world.
That's right, gang. The old joke from 
Last month I wrote about Motorola and its 
Since the collapse of 
From 
Which sci-fi author did the best job of predicting what the 21st century would look like from the comfort of the 20th?
It's nice when someone in the "major media" gets the Always On vision, no matter how they get there.
Over at eWeek, Peter Coffee offers a security analysis called 
Motorola calls it 
Along with all their other implications, the mass adoption of mobile phones represents the first step in the single-chip era.
Motorola has launched a very Clued-in strategy to push Always-On applications.
For the last year I've been harping here on the subject of Always On.
The Zigbee standard (also known as 802.15.4) has 
The bandwagon on behalf of "Mobile TV" 
Philadelphians are celebrating 
While searching for stories on Open Source (for our new 
The physics folks at UT-Dallas have won nearly $1 million to study the idea of making every radio part of a giant "mesh network." (That's the 
My vision of Always-On has always been based on the idea that applications would live on the wireless network in your home. A Wi-Fi set-up has both the bandwidth and computing power needed to handle several such applications.
One point often missed in the rush to Voice Over IP is how it leaves us all at the mercy of software companies playing games with standards.
If you're to take Always-On applications into the world with you, they have to be fashionable. They have to look smart. It would be very nice if they were machine washable.
The Boston Globe this morning is 
At first glance Microsoft's latest "big idea," 
David Peskowitz over at The Feature wrote yesterday about how Always-On medical applications might be 
What's a store?
With Microsoft cutting back, with 
Sometimes I get ahead of myself.
IBM has decided to make some of its key speech technology 
The 802.11 market is stalling.
Over at C|Net, Michael Kanellos has one of the stupidest leads I have ever read.
Each time a member of my family needs any type of medical care I'm frustrated. (The image is from the good people at 
A very important political story snuck by us last week. I blame John Kerry for it.
A decade ago, with my 
I have been following 
The New York Times is 
The history of computing is defined by entrepreneurs.
Yet computing's evolution has continued to accelerate. Microsoft had little play in local networking, which defined the 1980s, and struggled to get on top of the Internet, which defined the 1990s. (The Gates caricature is from a fine Italian artist. 
One difference between my youth and maturity is that, in my youth, the future was seen as positive while today we look upon it with fear. (This might be why the ratings for 
A few years ago, the fuel cell market was aimed at the market's high-end, providing back-up power to electric utilities and phone companies, especially in places where anti-pollution laws might prohibit other types of generators. (Illustration from 
In my new column at 
There will be many Always-On application spaces. There will medical applications, there will be home inventory applications, there will be home automation applications.
Samsung and Sony are getting 
Here's a nice Always-On application, found (once again) by 
I have wondered for months why motes, sensor networks, Zigbee, and nanotechnology don't draw more press coverage. Some of these markets are worth billions, right now. And they are going to revolutionize the world. Always-On will come.
Every industry will follow the money. Its events will be held where its markets are, or where its developers are. And the markets always have an opportunity to lure developers to them, through those events.
The 
It's inevitable.
Zigbee's present is in industry, not medicine.
There's always talk of politicians or pundits contradicting themselves. But what most people don't know is that companies do this all the time.
Having finished my work on residential gateways, I am more convinced than ever there is a huge opportunity here for someone. (Illustration from 
The cheap answer is 
What made the Internet work was the fact that it was a royalty-free, open standard. You could build on it, but there was no admission price for using it, so everyone did. (Illustration from the 
I've looked over coverage of President Bush's broadband plans, and they're "the old switcheroo." (Image from 
Linksys is a very important name in the World of Always-On.
A gateway has an Internet connection on one side, and some sort of LAN connection on the other. (Usually it includes a wireless LAN.) It's a modem, it's a router, it's a switch. It's pretty cool. (The picture is from 
Intel announced what many considered 
It's Cognitive Radio. 
The trouble with Intel is it thinks it's in the chip business. (Make this image your 
Bell Canada is now offering its ISP customers 
About 20 years ago, Dave Waks was one of the founders of Prodigy, the NAPLPS-based online service that tried to give the online world graphics a decade before the Web was spun.
For the last month I've been looking at residential gateways.
Few people really noticed it, but Microsoft yesterday made a formal announcement for 
One of the differences between this world of journalism and the real world of technology is the concept of the deadline. (Image from 
Opera, which just went public in Norway, has announced 
The cellular world is slowly being dragged, sometimes kicking-and-screaming, into the computing mainstream. Always-On applications are playing a big part.
Dean's Law is something I just made up. 
An Always-On home has both a wireless LAN on the inside and fast Internet connectivity on the outside. 
It wasn't by accident that a Netopia gateway is featured on the item (below) about cell phone business models.
CTIA is coming to Atlanta in a few weeks, and we haven't partied here since 1999. (I found this CTIA logo on 
The distrust between privacy advocates and big business is creating a solution where there's no problem, as RSA begins offering technology 
There are days when, talking to people about The World of Always-On, I sound a little crazy even to myself.
My recent note on 
The great thing about being a great company is you can make mistakes and get away with them.
While most makers of access points are in a standards race -- 802.11a, then g, then a+g -- a little outfit called 
Paul Allen's latest start-up, 
