Big Leo and the Dysprosium Project:

The Bigfoot Mambo at the UnWired Ball

by Joe Flower


A version of this article was originally published in Wired, vol. 2.06, November 1993

International Copyright 1993 Joe Flower All Rights Reserved
Please see our free downloading policy.


 System:                     Competition for global satellite telephone

 Phase space:                Global business                                            

 Energy source:              Money and corporate attention freed by the
                                  end of the Cold War

 System constraints  and     Rights to use spectrum; investors; 
      attractors:                 cooperation of          
                                  national phone companies

 Current state:              Stable D-soliton vortex
 Threshhold of               Not yet known                                              


Dick Tracy talked to his wrist. He had a little radio (and eventually, a tiny television) strapped to his wrist. The signal leapt, by the magic of comics, from Tracy's wrist to a satellite perched visibly over the buildings of the city like a Budweiser blimp. He could talk to anyone, from anywhere, because of that satellite. Pretty nifty.

By now we are used to people flipping open phones in restaurants, on the ferry, in theaters during love scenes -- teeny-weeny cellular phones that slip into a jacket pocket without making an unsightly bulge.

But our lantern-jawed crimefighter would have tossed today's cellular phones into a desk drawer and stuck with his wrist gizmo. Why? Today's cellular phones have ghosts and cross-talk. The spectrum is crowded. Other people can overhear-- the grand satraps of the underworld can certainly afford scanners. Sometimes when one "cell" hands you off to another you get dropped like a trapese artist with a timing problem. A cellular phone can't always call everywhere. If you want to call London or Brunei, you may have to get to one of those clunky old phones that actually has a wire attached to it.

But worse than that, a cellular phone can't easily leave town. Mine won't work at all in that big nothing on the drive to Vegas. I could be out of touch for hours. If I take the phone to another city, I have to set my phone to "roam" and pay extra. If I go to Europe, the phone won't work at all. They have a whole different standard over there.

And even worse, if someone wants to call me, they have to know what part of the world I'm in. This can be a real problem. No, I'm with Tracy -- I want a real phone, something I can toss in the pocket of my genuine Banana Republic photo-journalist's vest and take anywhere. I want my agent to dial my number and get me, whether I'm chatting with roustabouts in the oilfields of Kazakhstan or sipping kava in the Friendly Islands.

I know this problem worries you, too. I can feel your frustration, as you wander the tombs of Monte Alban without any idea what the Nikkei's doing, or sit in traffic between Heathrow and Soho with no way of dialing up Lagos.

Have faith. You have not been forgotten. The big boys are working on it. Give them another five years, and your troubles will be over. You'll be connected, always and everywhere, clear channel, error-corrected, voice- and data-capable, pagable, locatable, and encrypted -- all with one phone number, no matter where you are. Ask and ye shall receive.

Well, okay, you didn't ask, but they know you were about to.

How will they do this? Satellites, just like in Dick Tracy. Not the geo-stationary telcom satellites like Telsat that we're used to, big things 23,000 miles out. No, we're talking "Big LEOS," Low Earth Orbit Satellites, much smaller gizmos whizzing by 300 miles to 750 miles out (and even one M[edium]EOS system at 5600 miles out).

Who is going to do this? Ah. That's the $300 million to $3.4 billion question. Five companies and a major international organization are competing for this one, and others are trying to join in. No one has yet fired rocket one. Everyone is still in R&D on the system itself, the satellites, the handsets, and the ground stations. Yet the competition is already three years into a ferocious, public, many-sided corporate brawl conducted not in space, but in boardrooms, at international conferences, at lavish parties, in fax flurries and global conference calls. In six years, Motorola alone has spent $100 million on this idea, and it's still just an idea: a cellular system with "very tall towers" called satellites.

One one level, this is just another dance of the behemoths. One or more systems will get built, and in a few years you can sign up for it, if you like. Yet it's a dance that tells us a lot about who we are, what we expect, and how we deal with change at the end of the millennium.

The Product: What's the plan?

Follow the bouncing ball:

The Bigfoot of this dance is Motorola. This is the company with the gold-plated system, the one you you have most likely heard about, the one with the greatest number of satellites, the highest price tag, the biggest PR team, and the most peculiar design, castigated by a rival CEO as "very inefficient and expensive," characterized by another as part of "a strategy that fundamentally seeks a monopoly."

It's called "Iridium," after element 77 on the periodic table, because its original design called for 77 satellites (Iridium also happens to be so much more common in meteorites than in earth-bound stone that geologists use it as geological marker of extraterrestrial invasions).

Now the design calls for 66 satellites, but no one at Motorola has called for changing the name to that of element 66. "Dysprosium" doesn't have the same ring, and besides, its root meaning is "bad approach."

Iridium's 66 satellites will fly in 11 nearly-polar orbits (tilted 86 degrees) 420 miles up. Because of the near-polar orbits, Iridium will blanket the globe. If you're on an oil rig off the North Slope, and mama's at McMurdo Sound in the Antarctic, dial away.

The Iridium satellites will not only talk to handsets and ground stations. They will also talk to each other, forming a network aloft, passing on conversations, and handing them off when they have drifted out of range.

Because of these satellite-to-satellite "crosslinks," the Iridium system will be able to handle calls to other Iridium phones without reference to any ground stations at all, once the link is established. Say you're stuck in traffic in your Jag convertible on Sunset in Santa Monica. Good time to call the manager of that hot new act you've just signed. You have no idea where this person is (not to mention their honesty, drug profile, business acumen, legal name, elected gender, or actual level of control of and/or ability to motivate this hot new act), but you do know their Iridium phone number. You turn on your phone and dial. The signal goes from your handset straight to an Iridium satellite, which sends a query through the network of satellites to one that is over Iridium system headquarters. At HQ a computer certifies that you have punched in the right PIN number, and that you have paid your bill. Then it searches the database to find the most likely parts of the world to find the person you are calling, and sends the query back aloft, through the network of satellites, until one satellite gets a response from the ground. Your call is connected, and someone yells in the other end, "Who's calling me at 5 in the friggin' morning?"

"Baby! It's Duane. Your new agent. Where in the world is it 5 a.m.?"

"Oh, Duane. Tokyo. Remember? The tour?"


All of this is handset-to-satellite-to-satellite-to-handset. Or it could be, in theory. In reality, it most often won't be, for reasons that are, as we will see, extremely terrestrial.

The other systems are variations on this theme. They have fewer satellites, in less-tilted orbits, or in higher orbits (which gives each satellite a bigger "footprint"). One company, Ellipsat, plans on using elliptical orbits, which allow each satellite to spend part of its orbit going more slowly, skewing coverage to the more-populated areas (Ellipsat CEO David Castiel says, "Frankly, my business plan can do without the people on Easter Island."). TRW's Odyssey system features high-flying MEO birds that use inertial guidance systems to point themselves, focussing their beams at their selected continents. Nobody but Iridium plans to crosslink their satellites -- everybody else's birds are simple "bent-pipe" repeaters, sending the handset's signal back down to a ground station that feeds into land lines. All their handsets are "dual-mode:" they are cellular phones first, satellite phones only if they can't make a cellular link. All code their messages differently than Motorola, and all promise to do it for dramatically less money.

The Dance: What's really going on?

What's really going on is something between a minuet and a World Wrestling Federation Battle Royal. It has something in common with the Great Red Spot on Jupiter, a soliton, a vast storm of surprising stability that bubbles up out of the chaos, swallows everything in its path, and then sticks around, year after year, swirling, enormous, and nearly permanent. It began in late 1990, jump-started when tiny Ellipsat applied for an FCC license, followed rapidly by Motorola and the other competitors. They will stay locked in this vicious courtly scherzo until one or more of the systems goes up and others have given up -- unlikely, according to those who watch it most closely, before the turn of the millennium. The dancers include not only the companies involved, but the departments of State and Commerce, the FCC, various world bodies, a hundred or so national phone companies around the globe, commercial airlines, technology giants in Europe and Japan, rocketmakers in Russia, and even radio astronomers. Iridium even recruited the ambassador of Mali, plus his wife and staff, to act in a promotional video. The dance is political, bureacratic and corporate, but its realities can only be descried deep in the differing technical choices made by Motorola and its competitors. Each technical choice effects the business end, and the business decisions push the politics, and the politics mold the technology, around and around.

Everybody dance
To make a system like this happen (satellites in the sky, ground stations on the ground, listeners with phones to ear), you need several things. The biggest thing you need is a piece of the broadcast spectrum, (and things get rapidly more complicated if it's not the same piece in every country). Second, you need a license and an agreement with the phone company in every country in which you want to operate. Third, you need investors -- but if you build it they will come, if you have the spectrum and the licensing all taken care of, you will have little problem getting the investors. Fourth, and a distant fourth, is the technology itself.

The first two hurdles are very high. The spectrum that's available is small, and not empty in all countries -- and re-designing the equipment so that it won't interfere with the services already using that spectrum may make them impossibly expensive. Besides, it's nearly impossible to get every country in the world, even just the important ones, to agree to anything. Both Europe and Japan are busy developing their own cellular systems, and have hopes for their own satellite systems. Why should they sign on? And if they don't sign on, the projects are probably dead. Motorola official Mike Pellon has said flatly, "We do not see Iridium without European participation."

Motorola got caught in a wringer right up front at the 1992 WARC ("World Administrative Radio Conference"). It had used its considerable corporate weight to convince the U.S. government to ask for a piece of the spectrum around the world for Iridium and its competitors, even though a good chunk of that spectrum was already given over to radio astronomers (not a group notorious for its political clout), and the Russian GLONASS global navigation system -- a system the FAA is planning to incorporate into American civil aviation. This led to lots of head-scratching, harrumphing, work-arounds, technical footnotes, and power limitations, which did more damage to Iridium, the high-powered system, than to its rivals. In the end WARC `92 did recommend that the spectrum be made available.

But Motorola's real problem at WARC was its design. The system it proposed had nothing to do with land-line systems. It just ignored them. Technically cool. Cool for the user: one price, global coverage, don't need no stinking badges, just flip it on and talk.

Not cool for governments. Almost everywhere except the United States, the phone service is run by the government (usually as part of the postal service). In many small countries, it is one of the government's major sources of revenue, and international calls turn the highest profit. Now Motorola was suddenly proposing a system that would take all that profit out of the country and put it in Motorola's pockets.

So what would you say if you were Motorola? "Neat trick, huh?" Think again. Motorola is the largest manufacturer in the world of cellular phones and the equipment for cellular systems. These PTTs ("Post-Telephone-Telegraph" systems) are its major customers. In many parts of the world, they are its partners. It can't ignore them.

So now Motorola's Iridium phone, like those of all the other systems, has become "dual mode:" when you turn on the phone, it's just a cellular phone. Only if no cellular net is available, or if you push the satellite button, does it hook into the satellite system. So most long-distance phone calls from an Iridium phone would in fact never go through a satellite. And Motorola has its hands full assuring PTTs around the world that it never meant to steal their revenues. Never. Wouldn't dream of it.

Say this three times slowly: TDMA vs. CDMA
So how will the different systems divide up the spectrum that the `92 WARC made available?

Imagine eight people at a dinner table trying to have four conversations at once, with nobody seated next to the person they're talking to. Cacaphony. You could give each person one minute to talk while everyone else shuts up: a minute for one biker to tell the other about his new super-stroke, a minute for one father to brag to another about his five-year-old. Call this "time division."

Or you could assign each conversation a language: the bikers talk in Farsi, the fathers in Nahuatl. As long as they know the right languages, and as long as no one shouts, everyone can have their conversations at the same time with no interference. Call this "code division."

In January 1989 the U.S. cellular industry accepted TDMA ("time division multiple access") as the digital standard that would replace the current analog AMPS ("Advanced Mobile Phone System") standard. It allows a cellular operator to divide up the signal in tiny fractions of a second, allowing three times as many people to use the system at the same time.

Three months later, San Diego's Qualcomm Inc. introduced CDMA ("code division multiple access"), a new flavor of an idea that has been used in military satellites for decades. After three years of testing, the industry accepted it as a second standard, one that would increase the capacity of the system 10 to 20 times.

The big difference between the two is this: CDMA allows 4.4 trillion different codes, so different callers, even on different systems, can use the same spectrum at the same time. The time division in TDMA has to be done in one computer -- two different systems can't use the same spectrum at the same time.

CDMA is into sharing. TDMA is not. Systems that use CDMA can all co-exist in the same spectrum. For different systems to use TDMA, someone has to divvie up the available spectrum ahead of time.

All the global satellite systems are interested in sharing the available spectrum, and have decided to use CDMA -- except Motorola, the Bigfoot, which has stuck to TDMA, and demanded exclusive rights to a segment of the spectrum (in addition to 200 megahertz of spectrum for their satellites to talk to each other). Motorola's attitude has been characterized by Philip Malet, one of its lawyers, as, "Give us the spectrum and the let the others fight over whatever is left."

As long as there was only one system using TDMA, it could do all right in the same spectrum with other CDMA systems -- if it kept its voice down. But Iridium is a shouter -- TDMA, in Motorola's design, needs more power than CDMA to blast through to the inside of a cab or a building. The other systems can't sit next to a shouter.

The FCC's response was to order everyone to sit down and work it out. That didn't work. In "negotiated rule-making" sessions this spring, everyone agreed to share the available spectrum together -- except Motorola. So this fall the FCC is expected to issue its proposed rules, beginning a process that may take up to a year.

Congress has come up with one solution almost everyone hates: a bill authorizing the FCC to auction off pieces of the spectrum. As consultant Leslie Taylor put it: "Auctions would really complicate things, slow things down, and add cost -- so they create a tremendous pressure for the companies to come to some kind of agreement."

"An auction," says Nowacki, "would result in a monopoly. This is a public resource, and should be open to multiple access."

The likely outcome? They'll do it Motorola's way, giving 8 Mhz (half of the available spectrum) over to Iridium. "Iridium cannot operate without that," according to Mary Ann Elliott, president of Arrowhead Space and Telecommunications. "Motorola won't get tossed out by the FCC," says another consultant, "because they have paid off so many people by hiring the right consultants, giving the best parties, lobbying the right people. They have mowed down the opposition."

But the 8 Mhz that would be left for all the CDMA participants "is not enough," according to Ed Nowacki, vice president of federal systems for TRW. "The amount of traffic you could fit in that spectrum wouldn't make the project economically viable" for the other companies. David Wye, of Congress's Office of Technology Assessment, says that such a solution "might, from an engineering standpoint, be completely useless."

Once the FCC gives them their licenses, everybody has to re-convene before the International Frequency Rate Board, fighting off all comers to secure those frequencies on a global basis. If the U.S. competitors don't work something out, says Arrowhead's Elliott, somebody else "will likely be in place and offering service before any U.S. company has obtained a clear license."

The Dance Card
               Motorola's      Loral/      Ellipsat's     Constellation's    TRW's      
               Iridium         Qualcomm's    Ellipso          Aries         Odyssey     

 Satellites       66              48           24              48              12       

  Cost ($B)      $3.37           $1.6         $.4             $.29            $1.3      

   Orbits       Circular       Circular      Elliptical       Circular        Circular    
              86 dgr. tilt    52 dgr. tilt   + Circular                     55 dgr. tilt     

  Turn key       1998            1998         1996            1996            1998      

  Altitude        420             750        4680 X 312        550             5600      
    (NM)                                       6000                                       

   Handset       $3000           $700       $1000 (or         $1500          $500-600    
    price                                       $300                                       
    (est.)                                    cellular                                     

  Price per        $3         $.30 (plus    $.50 (plus       $.50 (plus      $.65 (plus   
   minute                      ground      ground             ground           ground      
   (est.)                     charges)     charges)           charges)         charges)     

Others are crowding onto the dance floor, elbows out, looking for room. A sixth company, San Diego-based Celsat, plans to enter the race with two geostationary satellites. Calling Communications Corporation, of West Covina, California has announced a $6.5 billion plan that calls for 840 lightweight LEO satellites and 84 orbiting spares. A number of other countries, including Russia, Singapore, Mexico, Tonga, Saudi Arabia, and Indonesia, have announced plans for systems that will compete for the spectrum given out at WARC `92. The American Mobile Satellite Corporation already has a license (in fact, it claims, the exclusive license) for "in-fill" mobile satellite service in North America, using geostationary satellites -- and expects to be operating by late next year.

The power mystery player is Inmarsat, an international organization with 66 member countries that already runs a voice and data satellite communications system for ships at sea. The terminals now cost $45,000 and are as big as file cabinets -- even the newest, briefcase-sized models cost $25,000, plus $5.50 per minute of use. Inmarsat wants to get into the handheld business, it has given the idea a name -- Project 21 -- and has commissioned studies by a number of major companies, but hasn't yet decided just how it wants to go about it. Iridium CEO Bob Kinzie earlier dismissed Project 21 as "just words." But now Motorola is lobbying the U.S. government to stop Inmarsat from competing with Iridium (the U.S., through Comsat, is a member). Inmarsat's own director of strategic planning, Patrick McDougal, admits that they'll be "fighting for the same customers." With 66 governments on its side, geostationary satellites already in place, and a rapidly growing cash flow, Inmarsat could end up with a global monopoly on satellite phones (in combination with AMSC). Many investors are sitting on their hands, waiting to see what Inmarsat will do.

"If this sounds vague and confusing, it's because it is," says David Wye at the Office of Technology Assessment. "None of these systems is operational. You can only do so many studies. Everything is up in the air."

Finally, there is one major tripwire that has everyone coughing politely and looking the other way: Section 310B of the Communications Act of 1934 categorically forbids the FCC from giving a license to any company with more than 20 percent foreign ownership. It's hard to tell -- since no one has raised the question, the official documents contain no answers -- but that would seem to rule out Iridium, with its Japanese partners; Globalstar, with its European partners; indeed, almost everybody. But I won't mention this if you won't. We wouldn't want to spoil anything, would we?

Wild Rumors
Because of its high price tag, its obsession with covering every inch of the globe equally, and its exclusivity, some people figure that Motorola originally designed Iridium for the U.S. military and intelligence services (a "three-letter" system -- DOD, DIA, CIA, NSC), just in time to run into the falling Berlin Wall. Motorola admits that some Iridium technology came from military systems, but denies that the system was designed for the military.

The first meeting of potential Iridium investors in March 1991 had a cover charge, an entrance fee to assure that no one was wasting Motorola's time and resources in idle curiosity. The cover charge was "approximately $1 million," according to industry sources quoted in Mobile Satellite News. Iridium spokesman John Windolph says, "That's ridiculous."

Get serious. Who'd pay for global phone service?
Is there really a market for this? Motorola figures Iridium needs a million customers to break even. Even shoestring Ellipsat is looking for half a million. What happens as cellular expands its geographic coverage, as it goes digital (and data-friendly)? Does the market for satellite phones disappear? How many globe-hopping CEOs and hot-shot Hollywood agents are there in the world?

"I'd sign up for it in a second," says Richard Buckberg. "I've been looking for something like this for years." Buckberg, a consulting biologist, often spends days or weeks on remote mountainsides, counting marbled murrelets or spotted owls. He often has to consult his office, or his clients. Once he gets outside of a cellular system, he falls back on the RCC, a suitcase-sized affair under the front seat of his truck. It consumes some 40 watts of power, it can't be removed from the truck, it can't send data, it is complex to operate -- he has to know the channel of each "repeater" station on the route -- and no one can call him when he's in the boonies. "A global phone that could receive calls and send data? That would be a godsend," he says.

Buckberg is far from alone. In fact, many people have jobs that take them to remote areas without good phone service, and many of these people need to be able to send data: construction engineers sending plan revisions, oil-company geologists uploading test results, surveyors asking for previous maps, adventure tour operators posting itinerary changes. Others, such as salespeople, field producers for television networks, and reporters, may be usually within reach of a regular phone, but they can be hard for the home office to track down (even paging relies on the person you are calling to find a phone and call back). These groups constitute a second market for global phone service.

A third market is the "in-fill" crowd: people who are well-off enough to pay for the service, but live in remote, thinly-populated areas that will likely never be covered by regular cellular service.

Finally, in many third-world countries, both regular and cellular phone service only cover a small fraction of the land area. If the governments of countries like Botswana, Mongolia, and Peru want to know what's going on in remote villages, they can set up land-line phone systems, with microwave and fiber links. Or they can try to teach the village leaders to operate a short-wave radio. But it would be far easier and cheaper to simply give each village leader a satellite phone -- instant infrastructure. Globalstar has promised complete pole-mounted solar-powered terminals for $2000 each that would give a local village phone system direct access to the public switched network in the faraway capital, with no need to string wire across the vast jungle/desert/mountain/swamp in between.

Add to that the many urban parts of the world (including many formerly Communist countries) where bureaucracy and antiquated systems can mean years of waiting for a phone line to be installed -- a wait that is unacceptable to fast-moving international businesses.

Counting all these groups, there probably is a very large market for satellite phones, especially if the phones (and the online time) can be kept cheap. Most systems are aiming at matching the cost of cellular, 30 to 50 cents per minute, plus any land-line long-distance charges. Motorola aims to charge $3 per minute to start with (including all charges).

And there's the rub: the Iridium design costs more. If you want your satellites to talk to each other, they have to have on-board computers to handle the complexities of networking. The computer has to have a backup, and so does its memory. They both consume power. And satellites that talk to each other not only need an extra set of antennae, they also have to be much more precisely pointed and positioned than satellites that just talk to the ground. Their positioning thrusters use more fuel. Altogether, the Iridium satellites will be bigger, heavier, more expensive, and more expensive to launch than their competitors. Arrowhead's Elliott says, "I don't know who would use it at $3 per minute, if there was an alternative."

Iridium's gruff, avuncular, white-haired grin-ready CEO, Bob Kinsie, harrumphs, "Motorola knows the market. Those other guys are just satellite builders."

$3.37 billion? Can Motorola swing it?
Not by itself. It has been looking for partners for three years now. As of this writing, it has "non-binding commitments" from such big Japanese players as Sony, Mitsubishi, Kyocera and long-distance carrier DDI, and is coyly hinting that big news will break any moment. Other companies, with cheaper systems, need less investment. Yet they are equally coy. Doug Dwyre, president of Globalstar (which counts among its partners such European firms as Aerospatiale, Alcatel NV, Alenia Spazio, and Deutche Aerospace) told Satellite News, "We don't really make a lot of noise about our investment successes, because we don't really see that it's quite necessary yet." Another CEO brags, "We have taken away major investors from Iridium," but he won't name them.

So who's winning?
Clam city. No can tell. They are all in the process of scrambling to convince various investors and national PTTs to back them, cooperate, cut deals. As one CEO put it, with his best Cheshire-cat grin, "Some are having more success than others." TRW's Nowacki acknowledges, "None of the systems have overwhelmed the world with their progress. No one has huge investors lined up at the door." Even Iridium's investors are, at this writing, still soft, check-less. Robert Paulson, director of McKinsey & Co.'s aerospace practice, says, "Everybody wants to be the fourth or fifth investor in this kind of thing. Nobody wants to be the pioneer."

There is something like a consensus among the competitors that one or more systems are likely to make it -- researchers for Connecticut's International Resource Development say there is room for "at the most, two of the `Big LEO' systems . . ." -- and that Iridium is the most likely to succeed. The next bets are Globalstar and Odyssey, or a conflation of several of the CDMA plans.

But not everyone is so sanguine. One report from the Office of Technology Assessment says that spectrum problems alone "may make operating a truly global system technically unrealistic." And even enthusiast Elliott, of Arrowhead, points out that AMSC's system, due to start service next year, has taken 13 years to get a full license, and concedes that "it is very unlikely that we will have any system operational before the year 2000 -- and any system will be hard pressed to keep up with Inmarsat."

This is a complex game on every level -- the technical difficulties of building a system, the intricate many-sided revenue-sharing contracts among scores of PTTs around the world, the legislative and bureacratic minefields of scores of countries, not to mention the diplomatic and political back-scratching needed. For instance, the technical demands of the Globalstar system call for about 125 ground stations eventually. But company officials estimate that they will need another 75 ground stations to solve diplomatic and political problems ("If you are giving our hated vile-dog enemy a ground station, we must have one, too, or you can't operate in our country, which is beloved of all the gods."). It will take enormous corporate persistence and large buckets of cash to get it done.

So the Dance of the Long Knives will go on. And on. And on.


A matter of style
Motorola, as an executive for a Pacific Rim PTT put it, "speaks with a big mouth" through lavish parties and slick presentations. Iridium's PR man sat me down in a high tech computer-controlled oval presentation room in Washington, D.C. for a series of videos. Across town, Ellipsat CEO David Castiel spent two hours with felt-tip pens and big paper explaining the whole system. Across the continent in Palo Alto, California, Globalstar's Executive Vice President Tony Navarra popped a globe onto his desk, rummaged in the drawer and said, "Okay, now imagine that this eraser is one of our satellites. This pencil sharpener is another. Here they go . . . "


In The Hall Of The Space King
(from the notebook:

Loral Space Systems satellite-testing facility, Palo Alto, California)

The sight is awesome. The feeling is awe. Is it the size? Or the strangeness?

To get here, we have gone through a security check, two electronic doors and two airlocks. We have had our coats taken away and replaced with white lab coats, our hair wrapped in coverings, our shoes vacuumed.

We are in a room with the size of a cathedral and the shape of a box, as tall as it is wide. All of its inner surfaces except the patch of floor on which we stand are covered in neat rows with thousands of narrow pyramids, each one two feet tall, six inches across at the base, blue, with a black tip. At an angle to the far wall, and at an angle to the side wall, two vast flats like drive-in movie screens stand cocked slightly toward each other, shiny, machined, rectangular, slightly curved, with each edge lined by rows of long, narrow, white triangles, like some gigantic child's cardboard cutouts of crocodile teeth. The light, falling from rows of halogen lamps in the ceiling high overhead, is even and brilliant. There are no echoes. Every voice seems a whisper.

At the center of this end of the room stands a great, complicated, wheeled pylon, holding high in the oppressively silent air a monstrous bug seemingly concocted of masses of black plastic film, tape, gold-colored mylar foil, cable, aluminum and carbon fibre. Smoke-gray dishes protrude from its sides, pointed at the screens. A mass of yard-long plastic-wrapped stalks of all sizes protrude from its face like horns or nozzles. A cone that could be a rocket nozzle -- or a cheerleader's megaphone -- protrudes from its back. Two men, in their own white coats and white hair coverings, work at a console on the floor below it, acolytes in an arcane rite.

It all seems other-worldly and, for this once, the feeling is entirely appropriate. This monster over our heads was not made for this planet. It is a satellite. Its true home is the stark darkness of orbit, of space and the solar wind.



Technology | Articles | Main Page