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

The Pulpit
The Pulpit

<< [ Natural Deselection ]   |  Digital Hubris  |   [ No Confidence Vote ] >>

Weekly Column

Digital Hubris: Apple's Tablet Computer Might Finally Be That Link Between Your PC and TV

Status: [CLOSED]
By Robert X. Cringely

High-tech is relentlessly optimistic and for good reason: the good times -- ALL the good times -- are caused by product transitions. New stuff costs more, has higher profit margins, and occasionally leads to changes in market leadership. A year or two later, these products will have been commoditized, the profit sucked out of them by intense competition, and it will be time to move on to the next big thing. Four years ago, the cheapest 802.11b access point you could buy cost $299. This week, I saw one advertised that with rebates brought the final cost down to zero, nothing, nada, zilch. Time to move on. So high-tech is always looking forward, never back, and taking a gamble on something new isn't perceived so much as a gamble but as a way of life.

The techniques for getting us to buy new stuff vary. In the best of cases, these new sales are driven by new functionality -- a color printer instead of black-and-white, a notebook computer instead of a desktop, a DVD instead of a VCR. At other times, the upgrade is driven by bloat as new MIPS-burning applications and operating systems make our old stuff too painfully slow. This doesn't happen by accident, folks. And into this performance abyss we throw not just new products but new TYPES of products, because industrial dynasties come from defining new market niches. Hewlett-Packard, for all its glorious history, is more than anything else a laser printer company. Cisco Systems, for all its desire to be something more, is a router company. These are niches they defined and that have led to decades of success.

And that brings us to the tablet computer, a tightly-defined product still in search of success.

Tablet computers have been around in various forms for years. Back in the early 1990s, we called it Pen Computing, and VCs lost a lot of money trying to get us to exchange our keyboard for a touchscreen and a stylus. The product success that emerged from that experiment was something both more and less than what was expected -- the Palm Pilot and later Windows CE. We didn't replace our desktops and notebooks with pen computers, but we added a new type of little computer to our lives. It was that perfect technical play -- the chance to replace a seven dollar, little black book with a $399 PDA.

A couple years ago, pen computers re-emerged as tablets with a larger form factor, supposedly expanded functionality and definitely expanded pricing. Microsoft made a special version of Windows just for tablet PCs, and most of the big hardware OEMs churned out tablet designs. But we haven't been buying them. In a U.S. market that supports sales of 50+ million PCs and notebooks per year, total tablet PC sales from all manufacturers this year will be less than 100,000 units. The screens are bigger and brighter, the applications smarter and the handwriting recognition better, but tablet computers are still looking for their killer app.

Apple Computer has been decidedly absent from the tablet game. In part, this has to do with the failure of the Newton, which will always be associated in the mind of Steve Jobs with his former friend and nemesis John Sculley. "Real computers have keyboards," Steve has said a zillion times, and he'll mean it right up to the moment he changes his mind.

That moment appears to be coming soon.

Quanta, the Taiwanese company that makes many Apple notebooks, has been apparently switching its production to the new tablets, or at least that has been reported in the Taipei press since early this year. If this is the case that Apple is introducing such a machine as early as January, how is it likely to be different from the Windows-based tablet machines that have so far failed to excite buyers? And why, in the face of such lackluster sales, has Microsoft done another rev of its tablet operating system? What is it about this product niche that makes it so attractive to vendors despite more than a decade of failure?

The simple fact is that until next year, the parts won't have been there to make tablet PCs successful. What's missing has been the killer app, and what kept a killer app from appearing was a lack of hardware support, which I believe will be over soon.

For Apple, doing a tablet really isn't much of a gamble. Macs still dominate the graphic design market despite Adobe's recent switching of allegiances to the Windows camp. The graphics market, which already absorbs a lot of Wacom tablets for drawing on Macs and PCs, can easily support Apple-sized volumes of high-end tablet computers. Give artists a big tablet screen to draw on, add wireless networking and good battery life, then throw this all on top of a powerful and easy-to-use OS, and Apple can be assured of at least breaking-even. They will become must-have gizmos in graphics departments everywhere. It's Apple's BMW strategy all over again, and virtually guarantees at least modest success.

But Steve Jobs would prefer something more than just modest success. He wants to define a new product niche or, in this case, finally give practical definition to a niche that already exists, kinda-sorta.

The tablet PC killer app for the mass market is functioning as a digital hub, a general concept both Apple and Microsoft have been pushing for a couple years. It's the idea that your computer ought to control your TV and your stereo and your VCR. The only problem has been that there isn't a good way to link these things all together, and even if we do, that digital hub isn't anywhere near your TV, at least not yet.

This little problem of the TV being in this room and the PC in that room was supposed to be handled by home networking, except that just doesn't work very well. When Bob Kahn, Vint Cerf, and God were inventing TCP/IP, they didn't put much thought into how their technology might handle HDTV streams. In practical terms, you might be able to use 802.11g to carry such a stream. But it wouldn't do so very well, and just one stream would take nearly all the available bandwidth. This would put a real hitch in your daughter's web surfing and totally annoy the guy in the next apartment, who'd like to have a little channel space for HIS video stream, too, thanks.

This is the reason why networking companies like Linksys and D-Link are pushing adapters they say can send audio and photos from your PC to your TV, but not video.

What's needed is a networking technology optimized for video transport. One has been in the works for sometime, and down at the IEEE, they call it 802.15.3, and this is where I believe Apple sees opportunity.

If your idea of having a life is following IEEE standards development, you know that 802.15.3 looks to be the first Ultra-Wide Band (UWB) networking technology used by folks who aren't spies and secret agents. You also know that the 802.15.3 committee is at something of an impasse with two technical approaches vying for approval. Most of the committee is in favor of an approach called Multiband OFDM, invented primarily by Texas Instruments and supported by Intel, Microsoft, and most of the big Japanese and European consumer electronics companies. The other approach, called DS-CDMA, comes primarily from a company called xTremeSpectrum, and is supported by all the Japanese consumer electronics companies that don't already support multiband OFDM. And since XtremeSpectrum was just bought by Motorola, DS-CDMA has just enough oomph behind it to keep Multiband OFDM from getting the 75 percent majority vote required to become a standard.

Understand that 802.15.3 is a high data rate Personal Area Network with a range of about 10 meters. This isn't a WiFi competitor. Think of it as a kind of Super Bluetooth, capable of sending video over short distances without interference and with true quality of service, which 802.11 can't provide. Conventional wisdom says a deal will shortly be worked out in the IEEE, Multiband OFDM will become the standard and we'll see products appear in late 2004 or early 2005.

But sometimes conventional wisdom is wrong, and I believe that is the case here.

For all the momentum behind Multiband OFDM, the only 802.15.3 silicon available so far is from xTremeSpectrum, and Motorola has pledged to start shipping products using those chips by January. No, they won't be 802.15.3 compliant, but they'll work and they'll be a year ahead of anything else. Think back to when there were several proposed 56 kbps modem standards and products were shipping all over that weren't V.90 compliant. So while meeting the 802.15.3 standard is nice, it isn't absolutely required to achieve commercial success.

Besides, Multiband OFDM as it is currently proposed doesn't really work.

The problem with Multiband OFDM is its very high Peak to Average Power Ratio. The FCC put great constraints on UWB to keep it from interfering with any other radio service and as it stands right now Multiband OFDM violates FCC rules for distortion and interference if operated at the power levels allowed by those rules. Absent some clever workarounds (several of which are in the works), the only way to make Multiband OFDM be truly FCC complaint is by reducing the power, called OFDM Backoff. This also cuts range to the point where your video and stereo components might be able to talk to each other in an equipment rack, but they might not be able to reach the tablet PC on your coffee table. And they sure as heck won't reach your PC in the next room.

A year from now, I am sure these technical problems will be solved, but by then Motorola will have been shipping DS-CDMA products for 12 months. And some of those products will be going in that Apple tablet computer. Watch TV in your bathroom, access your audio and video collection from anywhere in the house, control your big screen TV and route video to it from your desktop or the Internet. Take a dozen movies and your entire music collection with you on a trip. Strap the gizmo to the back of your car headrest and entertain the kids. Grab e-mail from a passing WiFi hotspot. Surf the web. Play video games. It will still cost too much, but a million early adopters won't care.

And a year later, when Multiband OFDM gets its act together, Microsoft will claim to have invented it all.

Comments from the Tribe

Status: [CLOSED] read all comments (0)