Centrino: Cometa. Rosedale: Clearwire. And so on: May I be forgiven for invoking the late and largely unlamented Cometa as Intel Capital stuffs $600m (of a $900m total investment) into Craig McCaw's already bulging pockets? Absolutely, I hear you cry.
In July 2002, we first heard about "Project Rainbow," which was launched as Cometa that December; the company said they'd build 20,000 hotspots over two years, though the number was later revised down. The initiative was lauded as a combination of Intel, IBM, AT&T, and two venture capital firms. In reality, as I have tried to remind people ever since, it was Intel Capital (funding), IBM's service division (installation), and AT&T's broadband division (service). None of the companies had anything on the line when Cometa failed.
Intel hopped on Wi-Fi after the train had left the station. By the time they launched Centrino and spent tens of millions on a Centro-verified hotspot branding program, 802.11b was already clearly on the way out. Centrino systems wouldn't get 802.11g for another year, and Wi-Fi Protected Access (WPA) security lagged longer with upgrades from Intel that conflicted with manufacturers' Windows XP editions.
Centrino was a huge success from a branding perspective, and Centrino laptops had substantially better battery life while not offering great performance.
Cometa ultimately went belly up in May 2004 after securing just a few hundred hotspots, with a focus in Seattle, and losing the McDonald's bid to Wayport. The Barnes & Noble contract that they had signed just before shutting down was taken up by SBC--now called AT&T. Poetic justice of a sort.
The company suffered from setting unmeetable goals. Their founder, Larry Brilliant, now the head of Google.org, made statements that went far beyond what was achievable in a maturing market, and then left to focus on his hot-zone--as in infectious disease--specialty, leaving the company holding that that bag. Intel had also lost interest in Wi-Fi. It had become an also-ran to WiMax. Wi-Fi had no strategic advantage to the firm given that it was a de rigeur part of every modern laptop.
And that's why Clearwire, my dear friends, is no Cometa.
First, Intel has staked the future of its telecommunications effort on WiMax. They've pushed money into dozens of companies. They've headed standards bodies, built chips, and subsidized a lot of early network development. They've pissed off cellular operators by creating buzz around technology that clearly competes and complements cell networks, revealing some of the paucity of user-oriented thinking in the cell industry. Intel has burned bridges and sown salts in other fields. If they can't make WiMax bloom, heads will roll from the top down.
Second, Intel "owns" WiMax in a way that they didn't with Wi-Fi. Intel didn't make its own Wi-Fi chips and wasn't a leader in the market, which was already four years into development when Centrino shipped. While there are dozens of companies, including some enormous ones who compete with Intel on many fronts involved in with WiMax, Intel has consistently taken a position far ahead of the market and somewhat ahead of the industry. They're not leaving this standard to other people.
Third, Clearwire has a spectrum portfolio. No one can own Wi-Fi, but you can buy pieces of WiMax by having exclusive spectrum rights for regions of the country. Clearwire surely needs more licenses, and I'm sure has a strategy about that, but they can reach 90m people now, they said in their now-withdrawn IPO filing a few weeks ago.
Fourth, the technology isn't unproven or oversold by Intel or Clearwire. There is a lot of hype around WiMax, but I can't find statements in which Intel or Clearwire oversells how the technology will deliver. They're leaving that to other firms. If anything, the two companies are underselling WiMax before they can deliver the goods.
Fifth, Intel Capital has invested something like $1b in WiMax if you include this $600m in Clearwire investment. That's something like at least 50 times more than their Cometa investment.
Sixth, there's a clear market out there for fixed and mobile wireless broadband, which is why there are hundreds of wireless ISPs and billions spent on cell data. It's also why the metro-scale wireless market is so hot, and why Wi-Fi--despite it being not the ideal technology for the purpose--will be blanketing hundreds of cities by the end of 2007.
Cometa had nothing unique over other hotspot operators except a staff split among at least three different headquarters that was much too large and a lack of signed contracts by the time they shut down. Hotspots were on the way to becoming a standard commodity that could be built and resold by a host of firms even as Cometa launched.
Clearwire has a particular set of advantages. They have Intel's attention. They have a chunk of spectrum licenses. They have a technology that works. And they have an audience that wants truly large-area mobile service from anyone but the incumbents--even if that's just for competitive sake to drive prices down.
Clearwire will execute, there's no doubt, and the question will be whether they can acquire enough customers at the right price to turn a profit. Cometa once talked about creating hotspots to such a large extent that they would cover every area in which a business traveler might find themselves. Clearwire has the potential to meet that target and avoid Cometa's fate.
Those who forget the past are doomed to repeat it.
Remeber IBM and the Micro Channel Architecture that was IBM's way to "take back the PC"? Or how about 802.12 aka AnyLan VG HP's attempt to dominate networking and go past Ethernet.
Well, WiMax is Intel's attempt to take control of wireless and it will be as successful as all the others. Particuarly as they are specifically recapitulating the attempt to best ethernet, in this case, wireless ethernet.
Sure WiMax will have some success, kind of like ATM. But it will just peter out at some point and have a little niche (a niche that will be good for some companies, like Alvarian, but no where the growth (or ROI) that Intel expects.
Licensed enduser data wireless has no room to grow in the US. Unlicensed wireless (aka 802.11 / WiFi wireless ethernet) will out evolve WiMax and fill the niches before it ever has a chance to become economically competitive. Wired / fiber broadband will decimate the higher end of the business leaving a small niche in the US for WiMax. (IMHO)
[Editor's note: I should have mentioned that Intel's strategy is worldwide, so the licensed issue may have more resonance in the U.S. than elsewhere. There is a lot of skepticism about mobile WiMax particularly in the U.S. before today's announcement, but a fair amount of interest outside the U.S.--gf]
Glenn, my default position is to agree with you -- you generally do a great job of taking the proverbial grain of salt with any wi-fi story in the news. But here I have to beg to differ. Clearwire may be no Cometa, but I think they don't have a prayer.
* This is a desperation play by Intel. They've staked so much on Wi-Max that they are literally paying people to buy the equipment.
* Wi-Max spectrum in the US kind of blows. As is, I don't see how it's a real competitor with traditional broadband to the home (except in rural areas), and as a mobility solution I haven't seen anything to indicate that it will beat HSDPA or EVDO here in the US.
Outside of the US, we have competitors in Asia (WiBro). There's some shot at Europe, but even if you could get a Clearwire Wi-Max success in the US, would it translate to there?
* Clearwire has been doing pretty shabbily so far, with some largish initial investments (something on the order of 150-300 million). What are they going to do with this money that will net them paying customers?
* I can't imagine that with this much invested that Intel won't be trying to direct things over at Clearwire. Is that going to make Clearwire more agile and smarter? Or it more likely to turn them into something big and dumb?
* How many metropolitan areas could you give wi-fi to for a billion dollars?
A lot of this doesn't add up to me. I certainly think it would be cool if Clearwire were a smashing success and in 2-3 years most people in the US have cheap, wireless broadband, or mobile data rates that aren't an arm and a leg. Unfortunately, I just don't see the Clearwire/Intel/Motorola mashup as being the people to do this.
[Editor's note: Let me clear that I am not predicting that Clearwire will be a huge success (nor a huge failure). Rather, that they have an entirely different set of circumstances that could allow them success or failure than Cometa. Cometa was essentially doomed from the start, and most of us in the industry figured Cometa had some secret agreements and plans that would give it an advantage. And we kept waiting for those plans to be revealed. When they finally lost out on the McDonald's bid, the game was over, and I realized my gut instinct in 2002 was correct.
Clearwire has a much better set of starting parameters, and I don't bet against Craig McCaw. The U.S. spectrum issues are pretty ugly, but Clearwire has a portfolio and I expect some plans we don't know about (unless I am once again giving a company too much credence) to expand that portfolio.
Remember that Clearwire and Sprint Nextel have been engaged in horse trading for months, exchanging geographic licenses to the benefit of each to have better coverage. --gf]