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
Pulpit Comments
September 14, 2007 -- The Power of Six
Status: [CLOSED]

Interesting ideas...what about security and e/m devices in the 700mhz frequency?

a.n. | Sep 14, 2007 | 1:17PM

How can this be, every Apple product is perfect, isn't it?

SJGMoney | Sep 14, 2007 | 1:20PM

It seemed to me that the greatest impediment to Google actually spending the big bucks to win the auction (they could clearly afford it) is that the mobile phone and data businesses aren’t as profitable as Google’s own search and advertising businesses, which means making such a move would hurt Google’s earnings and be a drag on the price of its shares.

But Google can't double their existing business at the same profitability, so in order to grow they need to branch out into new areas. This might be their best option for that.

David B | Sep 14, 2007 | 1:21PM

What about video? Why not provide an all inclusive communications service with data, voice and video? Charge half the price of all wireless phone, match the average price of all ISP's and half the price of all cable and satellite TV companies. Throw in movie distribution to put Blockbuster out of business too. The number of packets should be about 10 to the power of 100, 10^100, a googol.

D. B. | Sep 14, 2007 | 1:30PM

To quote Spock from the Star Trek episode: "Logical, flawlessly logical."

AT&T is too big and too committed to their existing practices to do some different with wireless. Verizon has decided to scare away innovation with law suits.

Lets assume Google wins the auction. At some point Google will want to start building a 700mhz wireless network. They could probably use lots of cell phone towers. Sprint has lots of cell phone towers. What if Google partnered with a forward thinking Cellular telco?

AT&T and Verizon's greatest risk is not Google, it is not working WITH Google. The firms that reject change and the future will be the one's most hurt.

Spock | Sep 14, 2007 | 1:33PM

I, for one, welcome our new Google overlords.

Interesting column.

Pug Matt | Sep 14, 2007 | 1:55PM

The one thing I truly hate about the cell phone business in the US is that consumers cannot use whatever phone they want with whatever features they decide they need, without being hindered by a contract. If Google set up a network that had nice customer support, a reasonable monthly fee, reasonable prices for data, no red tape, they would KILL AT&T, Verizon & Sprint. Sadly, no existing phones use the 700 MHz spectrum, so Nokia, Sony, Motorola and Apple would need to have phones that used that spectrum.

Wally Glenn | Sep 14, 2007 | 1:56PM

You forgot to factor in the Apple side of the deal. Apple has been studying the content distribution service (iChat, iTMS, iTMS(WiFi)) and they have some hardware (iPhone, iPod Touch, and most importantly iTV and Mac Mini). Put a 700 MgHz card in the Mini and the iTV and Google doesn't even have to provide hardware. They only need to subsidize the price to push out the Mini or iTV. Add your black box functions and in short order, you have your network with all the content you could wish.

Maybe we could have a new cable/satellite option?

Matt | Sep 14, 2007 | 2:06PM

A couple of things -

First, Google and the telco's will bid for the spectrum. It's an auction, and FCC rules only change the profitability of the investment and therefore the individual limits for the bidders. That was, of course, Google's entire purpose of trying to add constraints to the award - to reduce the value of the spectrum to the telcos and reduce the ultimate amount paid.

Second, I do think that Apple will partner with Google for the spectrum. I think they could care less about their relationship with ATT. We have to ask why Jobs is so eager to get the iPhone installed base up that he's willing to drop the price considerably despite the fact that sales seem to be right in line with their estimates. I think there's a critical mass in installed base that needs to be reached to make a move for the spectrum profitable, and he wants to get there as fast as he can. The form this partnership might take, however, doesn't have to be a joint bid. There could be an agreement between the two for use of the spectrum should Google succeed in the bid. But I do think Apple is involved.

mika | Sep 14, 2007 | 2:08PM


I think it may be a more astute move for Google to simply purchase a major cell provider rather than try to partner with them. That way you get the providers owned tower infrastructure, plus you get access to the colocation agreements the provider has w/ other providers.

But as for Cringely's point about Apple not being able to do two things at once, I don't buy that. I think it'd make a MASSIVE amount of sense for Google and Apple to invest in a jointly owned company whose sole purpose in life would be to provide an unfettered, cheap, fast, and omnipresent conduit to the net. Voip iPhones for all!

gb506 | Sep 14, 2007 | 2:08PM

Please be true. Please. Please. Please.

Jeeaye | Sep 14, 2007 | 2:10PM

This is a genius idea. Actually, it's two genius ideas: The integrated mesh wireless blah blah future utopia, which will never happen. And a plausible business case that fits with the biases of the players involved.

And the thin end of the wedge is Google's legion of nerd followers, many of whom hate the telcos with a passion. I switched to VOIP even at a higher cost just so I could keep from dealing with or giving money to SBC. I'd immediately drop up to $500 on something like this, and the rise of Fon suggests I'm not the only one.

William | Sep 14, 2007 | 2:11PM

700 mhz is not very good spectrum for providing broadband accesses. It travels far and through walls very well but the bandwidth through put is so low that it's usefulness is more for voice than anything else. An ISP using 700 mhz? no way.

dannyboy | Sep 14, 2007 | 2:12PM

"Is there anyone on this ship, who even remotely, looks like Satan?" -- Kirk

* "I am not aware of anyone who fits that description, Captain" -- Spock
* "No, Mr. Spock, I didn't think you would be" -- Kirk (The Apple)

it is happening (mereka)(fon et al) and speculation is nice- my bet is 5 years, but really it is already here just not cheap and truly ubitquitous. hopefully the gov't will see the light and make (or allow) something to happen with prescient governance - but look at telco mess- so don't count on it in the short term. it is inevitable though, just who can seize the day?

opensourceservers | Sep 14, 2007 | 2:18PM

Mesh networks require serious saturation to work properly. Are you really proposing that Google is going to have a Google Cube every 200 feet or so (in a city)? If we assume a perfect grid in a place like manhattan, covering one square mile of city would require six or seven hundred google cubes. Unless, by owning the spectrum, they were able to boost the power waaay up... but then are we proposing that Google is going to invent its own Wimax-like standard?

Christopher Mims | Sep 14, 2007 | 2:21PM

One interesting observation is that a large portion of the cell phone towers are no longer owned by individual carriers. Many are now owned by "tower operators" and serve multiple operators. Many local governments have REQUIRED multi-carrier towers to limit the number of towers. It would be really interesting to know how the contracts work - i.e. such installations may required to allow additional carriers!

Interesting read!

Bob Randall | Sep 14, 2007 | 2:27PM

I'm a big fan of Google on their own merits *and* I'm particularly in favor of anything that pokes a stick in the eye of the wireless/telco/cable oligarchies.

It's interesting that Sprint was mentioned. Their ultimate success in their own WiMax venture will cannibalize their existing Wireless business. If they bet the farm on WiMax - it could be a win-win for them to sell towers (and whatever other assets) to Google.

TimB | Sep 14, 2007 | 2:37PM

Looks like Verizon must have read your article, Bob - suing the FCC over this today.

D | Sep 14, 2007 | 2:45PM

Like Fred Hoyle's "Industrial Corporation of Eire" (ICE), Google is showing us what happens when bizarre and unprecedented amounts of intelligence control the vision at a massively wealthy company. I'm deriving immense pleasure from watching the dynosaurs lumbering panic as Google eats their eggs, their lunch and their futures. Now, if the dynosaurs had lawyers, would they be extinct now anyway?

Robert Kane | Sep 14, 2007 | 2:46PM

Bob, 700MHZ is a P.I.G. pig! At it's best, 2Mbits of bandwidth. I'm employed by an older dialup networking company who over the last two years has been building-out muni Wi-Fi networks... I've managed to learn a few things from our RF guys. 700MHz, just like 900MHz and all other sub-1GHz frequencies are at best 1 to 2Mbits. Hence the reason for all the metro Wi-Fi hype in 2.4GHz through 5.8GHz which provide for as much as 54Mbits. The only advantage to 700MHz is it's ability to punch through trees better then the higher frequencies, which in the case of 5.8GHz, one large tree will trash the signal. I'm sure Google knows this fact so I'd have to think their use of 700MHz will be for something much simpler but certainly not for backhaul.

dial up internet man | Sep 14, 2007 | 2:51PM

Sound svery whiz-bang, but coverage would be extremely patchy, there would be little to no planning against interference and 700MHz has pittiful bandwidth compared to the 2-3GHz spectrum used for cellular 3G. Google would have to come up with their own air interface and development of entirely new infrastructure and handset radio technology and hardware.

These are interesting thoughts, and a lot could be done with this spectrum by an innovative company with no legacy investment, but it would have to be done within the limitations of the spectrum. You cannae change the laws of Physics!

Simon Hibbs | Sep 14, 2007 | 2:55PM


I enjoyed today's sermon both for your insights into what the heck Google is up to, and for reminding me with the "Real Soon Now" phrase, that I have not thought of Jerry Pournelle in a long time!

Thanks for both.

Rick Hamrick | Sep 14, 2007 | 3:01PM

To those who believe 700 MHz is somehow going to produce slower throughput rates solely based upon being below 1 gigahertz are mistaken. Television stations are pumping 19 megabits through a 6.5 MHz channel *below* 700 MHz.

zentec | Sep 14, 2007 | 3:08PM

The first thing that came to mind, when reading of Apple's interest in the auction, was a protocol - 700 MHz GSM, or 700 MHz WiFi. Something that could allow Apple to tap one universal US network, without having to add another radio to the iPhone.

Or possibly WiMax, Intel's WAN wireless technology.

I think we'll see Apple subsidizing Google's bid for a bigger warchest, with an agreement that Apple get access to that spectrum for its consumer products.

Graham | Sep 14, 2007 | 3:09PM

I guess I am old, and remember what the difference between TV frequencies(54 to 88 megahertz for channels 2 through 6 and 174 to 220 megahertz for channels 7 through 13) and cell phone frequencies(824 to 849 megahertz)!
Now let's see, doesn't that mean a lot less towers, will penetrate through almost anything and can deliver almost any amount of data!
I guess Google really must not know what they are doing???

Michael Ganey | Sep 14, 2007 | 3:13PM

I am not much of a Google fan, but if they did something like this I would be a dedicated customer.

Allen C | Sep 14, 2007 | 3:26PM

Google cube? Isn't the new form factor for that the google phone (and future iphones)?

dj | Sep 14, 2007 | 3:46PM

JOOC, what is the title supposed to mean? The power of six WHAT? There isn't any six of anything mentioned anywhere.

Rick | Sep 14, 2007 | 4:01PM

I like your writeup, but I don't think there will ever be a Google Cube. There's already an Apple TV.

Brian Bennett | Sep 14, 2007 | 4:03PM

This is all a headfake. Google is actually off secretly partnering with electricity distribution companies for pushing comm over power lines. We'll wake up tomorrow and be able to throw all our DSL and cable modems out the window. Hmm, so how do you get a digital signal to jump a transformer? Ah, of course - headfake #2.

Dave Cline | Sep 14, 2007 | 4:13PM

If only.

George M. | Sep 14, 2007 | 4:18PM

I still think the real use of this initially would be a set of phones and small tablets. Imagine the google-phone would not only let you search for eateries nearby, but also have coupons for said places. Americans are willing to put up with a lot for a couple of bucks off( like rebates.) And then google can get paid by the location for delivering customers.

Google get cash per customer and the restaurant has clients.

Add the possibility of reservations, ordering from the phone and google get a cut. And if you are a small business you would get your interface device( tablet) from google and you would be glad to use it as it is bringing customers into your location.

Piotr | Sep 14, 2007 | 4:21PM

When will they change the name of the Google Cube to Airport Ultra?

Doug Kelley | Sep 14, 2007 | 4:27PM

You forgot about Apple, Cisco (iPhone deal) and Disney.

Lots of good alignments in these organizations working together against the entrenched monopolies of cruft purveyers.

And "zentec" nails it on the bandwidth. Learn some physics folks !

Secure Care | Sep 14, 2007 | 4:28PM

"It seemed to me that the greatest impediment to Google actually spending the big bucks to win the auction (they could clearly afford it) is that the mobile phone and data businesses aren’t as profitable as Google’s own search and advertising businesses, which means making such a move would hurt Google’s earnings and be a drag on the price of its shares."

I don't think that's accurate; Google wouldn't be looking to make any money at all from the wireless business. I think they're looking at it like this:

1. They make money from advertising-fueled software services.

2. The current wireles/handset industry is failing to make it easy for consumers to use Google's services.

3. Google, by getting into this business, can provide a streamlined wireless path between the consumer and their more lucrative advertising business.

They stand to make a lot of money from it, just not directly from selling wireless handsets and data plans.

Eric | Sep 14, 2007 | 4:56PM

Spelling mistake in paragraph 8, limitsto.


David Cameron | Sep 14, 2007 | 5:13PM

I own a WISP and we use some Ubuquiti 900MHz gear (700MHz should work identical). We get 25 to 30 MB/s of real-world throughput on 900MHz (and it's mesh to boot). To the guy stating 2Mb/s is max throughput on 700MHz, I say that is simply not true. Ubuquiti radios emulate 802.11a/g.

Kenny Bain | Sep 14, 2007 | 5:43PM

Mr. Cringley,

It is interesting that you mention this concept. I would like for your readers and yourself to also view this interesting speech by 'Van Jacobson' at Google Speaker Series. I have included the URL above, but if it does not work, be sure to just search with terms 'google speaker series van jacobson'

Devang Shah | Sep 14, 2007 | 5:46PM

Mr. Cringley,

It is interesting that you mention this concept. I would like for your readers and yourself to also view this interesting speech by 'Van Jacobson' at Google Speaker Series. I have included the URL above, but if it does not work, be sure to just search with terms 'google speaker series van jacobson'

Devang | Sep 14, 2007 | 5:48PM

I can see it. One thing about Google, consumers do not buy from Google. Businesses buy adwords. Consumers use everything for free.
Actually selling to consumers, and lots of them is a big cultural change. Maybe Apple can help.

Because of this and the real technical issues of an infrastructure and consumer product role out, I have to disagree with the time line. If Google wins the auction and has the bandwidth, it will be about 2 years before the service is available in major metro markets. It will take longer to get out to the sticks where you and I live.

They have some pieces, sure. But until they put cash down on the spectrum, this is not a real project. Very few Googlites are working on this when the risk of loosing the spectrum auction is large and real.

If this is the grand, but not evil, plan for world domination, and they loose the auction, what is the backup? These guys are too smart to not have a backup.

Doug Withau | Sep 14, 2007 | 6:23PM

Uhm, you keep making predictions about what people will do, Bob.

I keep waiting for them to come true.

Mind you, I don't TRACK your predictions, so I'm not saying none of them do. I just don't have a feel what whether you are or are not "a pretty good guesser."

How about posting a list of your last X predictions - that have resolved themselves - and establish which ones were right and which wrong.

And I don't mean your New Years predictions, I mean your column predictions.

I still don't see the Google container data centers all over the place, I don't see the Google Wi-Fi anywhere but Mountain View, I don't see a lot of things that sounded good a few months ago, but have no trade press anywhere other than your column.

Doesn't make me confident that Google - or Apple, or anybody else you've predicted on - is going to become the Borg any time soon. Not in comparison with Microsoft, anyway.

Richard Steven Hack | Sep 14, 2007 | 7:14PM

do you really think micro$oft would let Google win that auction? hell no.

dbro | Sep 14, 2007 | 7:37PM

I read the somewhat negative reviews of the iPod Classic also, but I'm not upset at all about the third-party dock incompatibility. Apple made some fundamental changes to the video capabilities -- a good thing.

From "Previous video-enabled iPods offered composite video output from a video iPod Dock or via a cable that plugged into the headphone jack. The new Classic, Nano, and the Touch all offer both composite video and now higher quality component video for use with a HDTV or modern standard definition TV. Component video delivers a higher quality picture."

From "Apple claims that video out works only with accessories that support the new iPods’ 'enhanced video capabilities.'... these 'enhanced video capabilities' hint that these iPods will eventually play video encoded at greater resolutions and bitrates and that’s why the circuitry had to change (and old accessories had to bite the dust in terms of compatibility)."

Mark B | Sep 14, 2007 | 10:01PM

think about what the effects widespread virtually free (filter out p2p) bandwidth would have...on education, the economy, and just culture in general. Commoditization of the internet is in societies' best interest but it has to be managed in a strict sense of neutrality. Maybe like the olpc project some concensus will form that we must act and get some pilot projects going and build excitement for a positive constructive project. It is so much bigger than most anyone can imagine i think it is really hard to grasp both the breadth and depth of it and also its awesome potential. I say go for something even if it fails.. the time is now. think different and dream big.

nxrw33 | Sep 14, 2007 | 10:06PM

1.)Microsoft would NOT be a white knight for the mobile carriers - the government oversight is nearing the end, and there's talk of extending it, as Microsoft's behavior doesn't seem to have changed enough for some of the overseers. Why complicate things with a gentlemen's agreement?

2.)If Google bids and win the contract and implements the 700MHZ immediately, don't be surprised if at Christmas, Steve Jobs roasts AT&T like chestnuts over an open fire and publicly announces a new mobile alliance with Google - cheap or free service. I can imagine the other mobile carriers suing because they can't compete with free - the judge(s) will say tough, bring your case to court when Google owns 99% of all mobile business and you can prove that Google is anti-competitive and the service they provide is lousy.

Kevin Kunreuther | Sep 14, 2007 | 11:39PM

Could Google kick-start the build-out of this mesh network by making a deal with some retail business that has a lot of locations? Maybe Starbucks would be interested in hosting a device that could bridge WIFI with the Google 700 MHz network...McDonald's, Wendy's, KFC, Burger with chains could get tens of thousands of mesh devices in place very rapidly. It's the Disney advertising model.

Does anybody here live near a Starbucks or McDonald's?

P.S. Rick, a cube has 6 faces ('The Power of Six').

stephen | Sep 14, 2007 | 11:55PM

1Mbit or higher Speeds: 700MHz, 900MHz, 2.4GHz or any of the licensed spectrum, it all boils down to how much power the FCC will allow to be piped over those radio antennas to achieve high bandwidth speeds worthy of providing back haul or WiFi-like access point services. Yes -- I'll take the 'heat' from my prior post. I was somehow assuming 700MHz would be unlicensed spectrum, thus regulated by the FCC to a max 400Mw of power. I'm not one for knowing what power level those giant 200 foot (or taller) TV station towers put out but certainly anything above the dismal 400Mw power restriction on commercial WiFi access points would allow for high data flow. However, I still have a problem with 700MHz, at any power above 400Mw, with towers spaced a few miles apart (similar to cellular services), providing for anything above 1Mbit bandwidth. It certainly will provide for the higher bandwidth speeds but the equipment required on the customer side will also need some significant power to reach back to the tower location[s]. Broadcast TV was easy, it's a one way street. Ramp up the power on the broadcast tower, have people mount a fairly odd looking antenna on their roof which just so happens to be significantly larger then any notebook computer or PDA wireless antenna, plug in an AC powered TV set, and receive some stations. Now switch sides, and try sending a high speed signal back to that TV broadcast tower. Good luck and if you make it, you'll get the big article in popular science. -- I'll conclude with Maybe, just maybe if the winner of 700MHz sticks to a muni-wifi type arrangement, an access point every 1,000ft or so, the speeds/services will be acceptable but certainly not if trying a cellular model of broadcasting every 3 to 5 miles in a 'cell' pattern. The end result for any speed above 1Mbit, will require a box (radio,antenna) plugged into everybody's house. People are not going to be able to roam around with their notebook or PDA using a 700MHz card expecting anything above 1Mbit. Interestingly enough, Craig McCall's Clearwire company has been building out their high-speed data services in several cities in the U.S. using licensed spectrum, using boxes that are required at each location, and their posted download rate is 2Mbit. Now if don't mind me adding a second 'to conclude' I'd have to think that if you take that box, in this case I'm talking about a 700MHz box communicating back to the tower, and make it a Wi-Fi access point (or bridge) and rebroadcast 700MHz or 2.4GHz...., wow -- that could be a very 'viral' mesh network build-out. An access point stuffed into various houses and businesses across entire cities and towns. The network would keep getting stronger as more and more people purchased services requiring they too plug in a 'box' inside their location. 700MHz on high power reaching back to the tower (back haul) and 2.4GHz access point at 400Mw in everybody's house reaching out 600ft to 1000ft, but using mesh technologies so that the SSID's become seamless and not a sloppy mess. Look, I wrote enough for now but if this would be the proposed solution, a viral mesh build-out, but using a cellular tower type back haul arrangement, this would be something to watch out for.

dial up internet man | Sep 15, 2007 | 12:03AM

To back up the other WISP poster, i am a former WISP employee and the determining characteristic of wireless bandwidth is how wide the "channel" is. 802.11a/b/g all use a 20MHz channel by default. Using the frequency techniques of 802.11a/g, you can push ~54Mbits aggregate (up and down) from one client to another in a 20MHz channel. This is a bit efficiency of ~2bits/hz. It doesn't matter if this 20MHz is 700-720MHz or 5.300-5.320GHz or 5.800-5.820GHZ. If you use the underlying technology of 802.11a/b/g, thats how it works.

802.11n uses a technique called Multiple-in/Multiple-out to improve this efficiency, but uses the same general channel size. All this can be looked up on Wikipedia if one cares to.

N M | Sep 15, 2007 | 12:13AM

the datamining potential of all the users out there on this new wireless network may be great enough in their own right to justify a free system. wimax could be a contender as well, building vertical markets is good but there should be spinoffs and unintended good consequences as well as things like terrorist activity and botnets. as the clash is wont to say,"the future is unwritten".

root | Sep 15, 2007 | 12:29AM

Sprint and Cable companies partnership is the real dark horse in the auction:

Let's keep speculating about VZW and GOOG's plan for the band. That way we can all be surprised when Sprint/Comcast/TW get the spectrum.

Eric | Sep 15, 2007 | 9:26AM

I'll have to hunt it up somewhere on my computer.

Essentially cell phones talk to each other locally but also link to internet for longer distances.

No cell phone company required.

cannuck | Sep 15, 2007 | 10:10AM

Google/Apple should buy the spectrum and run this service platform: on the resulting network. Slam dunk.

Rickard Olsson | Sep 15, 2007 | 2:21PM

It would be nice.
Sometimes I equate capitalism with dictatorship.
As consumers we are over-whelmingly restricted with our options, as well as over-paying for said options.

5andman | Sep 15, 2007 | 6:15PM

The flaw of a plan like this is that if Google did become the super-dominant ISP/phone/everything provider, they'd be ripe to get broken up by the government. Except, oh yea, the government is completely dysfunctional and will never stop a merger or break a company up. I forget, we'reliving in corporate anarchy.

Bog | Sep 15, 2007 | 8:11PM

Why not? Just kinda changing monopolies aren't we? I'm a little tired of how mobile operators have stymied design and development aren't you. We are behind every other country due to our trying to control everything in today's architecture. And for Job's stirring things up. It's about time someone did!

David Keller | Sep 15, 2007 | 9:57PM

The Google Cube is already out, it's called AppleTV, or AirPort Extreme, or iPhone, or iPod touch, or any Mac since 1999. They all run OS X, all have Wi-Fi, and with the exception of the Mac they are all $299 or less. And all are purchased and set up by consumers. Apple has already acted as Google's dream consumer hardware arm. In addition, the default search in all of Apple's products is Google, the Maps in iPhone is Google, and YouTube is on every Apple product with a display.

I don't see Google and Apple competing at all, each knows better. Apple's not going to offer Web search and Google's not going to try to sell its own iPod. They each have much better things to do with their own successful world-class businesses. If one of them buys 700 MHz (or they buy it together) it will be good for both, it will be a link between Google's servers and Apple's clients.

I don't think Apple cares about what AT&T thinks at all. The reason Apple doesn't care about 3G is they are going from 2.5G all the way up to Wi-Fi. You're supposed to curse out EDGE when you're using an iPhone. You're supposed to hunger for nationwide Wi-Fi.

Fred Hamranhansenhansen | Sep 16, 2007 | 12:40AM

One current problem in local internet service is there isn't much competition. The local oligopolies aren't interested in competing with each other; they're busy recovering their infrastructure costs (and churning out substantial profits). This is one of the flaws in fields where there's a large fixed infrastructure cost.

You've also forgotten about fon, which is google-funded. That, togther with their local wi-fi experience, means that they've been learning about how to provide large amounts of bandwidth to large numbers of users by distributing POPs for free.

What does that add up to? A company that can, if it wants to, distribute large numbers of POPs nationwide, for free. Those POPs are free to subscribers, and pay spots to non-subscribers, just like fon. With the large backhaul, google becomes the de-facto nationwide ISP.

Why would they do this? Becuase they can!

Actually, they do this because they need to jump-start ubiquitous, cheap, fast internet. That makes all their properties that much more valuable. Why pay for cable when you can watch the shows you want on youtube - for free? Pick any show - any actor - any director - and it's available on-demand, for free, available across google's backbone and user-installed POPs. They could provide the whole quadruple play: cable, phone, tv, wireless. And why not?

mannyv | Sep 16, 2007 | 1:13AM

"tiny Linux or Linux-likeserver, and a few gigs of flash RAM memory cache... WiFi access points and 700 MHz mesh backhaul devices...local caching, video preloading, and truly local DNS service"

Joseph Coates | Sep 16, 2007 | 1:47AM

This is probably a stupid question, but why wouldn't Microsoft bid in the auction for the 700 Mhz bandwidth?

knowsnothing | Sep 16, 2007 | 1:49AM

Let's say that Apple introduces an iPod Touch with a microphone that works with WiFi. Not much different than they have now. And, Google provides a way to go from WiFi to the 700Mhz band. Would that be breaking the contract Apple has with ATT?

What if the iPod Touch could connect to the Internet over that 700Mhz band? Would that violate ATT's agreement?

And finally, what if Google had a webapp that allowed the iPod Touch to do phone calls via the 700Mhz band?

Google and Apple have a little business partnership going on. Safari uses Google as its default search engine. Apple has Google apps on the iPhone. And Google doesn't come out with a gPhone, a gPod, or a Google PC. I bet if Apple promises to produce the hardware that will use this 700Mhz spectrum, Google will have no qualms buying a chunk of it.

And where is Microsoft in all of this? Certainly, they also have the money to bid, they don't want Google getting a foothold in this spectrum, and they also want to make sure that the "mobile phone cartel" doesn't get a monopoly on this spectrum. You think they're thinking of doing a bit too?

David W | Sep 16, 2007 | 1:58AM

Don"t forget the iPhone deal with Oracle. That is a third party that would fit nicely into this deal - networking. Oracle has had influence on the Apple board in the past too with Larry Ellison resigning. There would be money in the Oracle bank as well. How better to change the world than to give the phone companies a run for their money. What made Oracle back down on the law suite over the name?

Ads, Hardware and Networking. We are the consumers!

Airrow | Sep 16, 2007 | 4:50AM

If Google were to buy TerraNet or some similar company for their peer-to-peer phone systen they could get a 700MHz phone network up and running with far less infrastructure build-out.

All the conclusions about Google's existing distributed data centres and fibre network remain the same, of course, but the phones could be working in people's hands much sooner and the hit against the 'traditional' providers much bigger.

SimonBrooke | Sep 16, 2007 | 5:32AM

Several very astute points here Mr. Cringely. Verizon and other companies will try to buy 700 mhz spectrum if the price is right but with no intention to use it and a hugely expensive deployment with no seemless integration economically available, they only thing a telco can do with the spectrum is suppress it. The last dual entry was Cellular vs PCS spectrum.

You also note the lack of benefit for Google is straight forward. They are are an enterprise network operator with skills and money to be a carrier network operator. Yet, suppressing their profit margins with the network operator profits is not attractive. Only companies interested in infrastructure are interested in high volume low margin business models. Googles speculative investors cannot handle a nongrowth period yet. Their board understands low volume transactions and high margins, not high volume low margins.

There are very very few companies who can consider building our carrier networks with intention to build content (High volume network with low low margins for users, built in such a manner to offer content with high margins, inernally owned).

If any company buys in strongly to 700 mhz, they will get a bargin if the manufactures respond with crystals and network equipment. Otherwise it will be as costly as other spectrum without licenses.


Kelly | Sep 16, 2007 | 12:21PM

Interesting take Mr Cringley. Is the Google Cube really an Apple TV mashed with Airport Extreme + and a few bits of gubbins? Sounds more likely after reading your article!

Darryl Collins | Sep 16, 2007 | 1:50PM

WISP guy here again. Great article (again), Bob! Interesting comments as well. I wanted to comment, once more, by throwing out some ideas I had after thinking about this for a couple of days. First of all, to the dial-up guy: whomever gets the 700MHz spectrum will be able to use much higher amounts of power than what we get in unlicensed spectrum. I would assume that the 700MHz licensed spectrum would get an order of magnitude higher (in power) than what the FCC would allow for unlicensed spectrum. At that rate you should get penetration similar to what was seen with UHF TV.

Back to the article: if this is Google's strategy, it is brilliant! What they are doing is not actually making another pipe into the internet, but basically making a network that exists as an independent "extension" of the internet. First of all, they must have a plan for all of that dark fiber they bought up a few years ago and using it this way allows Google to handle the enormous amounts of traffic that voice/video/data will need INDEPENDENT OF ANY ISP. Therefore, the first layer in this new network is made by lighting up the dark fiber network.

The second layer is wireless, but probably not 700MHz. If I were them, I would install full-duplex, carrier-grade microwave links to all of my "cell sites". These links should be able to handle 100Mb/s (or more) to each cell site, with only one or two hops to Google's fiber network. Ideally, I would partner with a national carrier (such as Sprint) and utilize their existing network of towers.

The third layer in this new network is 700MHz. High-powered radios with adaptive array antennas are used to communicate with client devices (Google Cube). For urban users, simply connect a small antenna to your device and you can easily go online. For rural users, an external (directional) antenna may be required and the cube will have a program that lets you see the signal strength as you rotate your outdoor antenna (possibly a small beam or dish).

The cube should also have a WiFi radio (802.11n) inside. Once it is powered up, the cube becomes a WiFi portal that allows any WiFi enabled device to access Google's network.

Now, let's go back to the Sprint partnership. We let Sprint handle the billing side of things and split the revenues with them. Therefore, in addition to the revenue brought in with enhanced search offerings (location based), another revenue stream is produced by access revenues.

What is the end result of all of this? First of all, this nueters network nuetrality. AT&T can no longer threaten to charge or block Google (or anyone) because the Google network now exists nationwide and probably will deliver more throughput than AT&T does to end users. Second, the broadband providers will be forced to come up with new and innovative ideas to deliver (even more) throughput to end users or will face the possibility of losing all of their subscribers to the Google network. And finally, the most important thing is that the consumer comes out on top!

Kenny Bain | Sep 16, 2007 | 6:23PM

Wow. Sounds right-on, Robert. Excellent analysis.

John | Sep 17, 2007 | 3:49AM

Having worked for a major telco recently (product architecture) I was impressed by just how scared they are of municipal wireless, and how it affects their entire suite of product offerings.

This affaire le Google is the first true challenge in the history a communications infrastructure long past it's business plan use-by date.

Getting Google to run the new infrastructure you've described promises a shift far more primal than the evolution of POTS to mobile, and dispite the size of the newly rising monster I think it has far more potential to flex with new technology as it arrives than the sagging wire telcos can ever offer again -- The Bell Labs days are gone, gone gone. They're not being regulated out of existence as they claim (I live in Oz, go figure) they're just old & busted.

I haven't seen so many stunned faces since I worked at a company that made dedicated word processing machines just as the PC came out.

Wait for the sound of the thud when the titan falls, it won't be long now.

Nefarious Wheel | Sep 17, 2007 | 5:48AM

When is "Triumph of the Nerds" part two coming out?

Jim | Sep 17, 2007 | 11:47AM

Interesting... I guess will have to wait and see if Bob is a good prognosticator or not. Here in NJ about 7 miles from the original Bell Labs, Verizon's FIOS is just a theory. Despite all their advertising there's no sign of it coming anytime soon. It would be great for this Google scenario to happen and not only be a cool set-up in itself, but light a real fire for high speed connectivity competition.

Steve Yakoban | Sep 17, 2007 | 11:57AM

Google can and must bid on the 700 mHz. auction...why?:mobility,mobility,mobility. The iPhone can easily be made into a 700 mHz wimax "mobile phone,data,video" device with all of the cool apps that Xgen consumers want...Schmidt and Jobs see that there's a new elephant in the room and it aint Ma Bell. Google can do a deal with American Tower Corp or Crown Castle and have instant coverage of 90% of the US...hello broadband, bye bye 4g. With copper prices so high now the telco's can scap all of there outside plant before they go the way of the Dodo bird.

Stu Browne | Sep 17, 2007 | 12:52PM

Think WiMax when you think of Google's bid. The cost for building out WiMax is dramatically lower than the cost of building out a traditional cell network. WiMax is all IP and Google cna be counted on to support VoIP on such devies and spectrum. All IP is what they know how to manage and have built their backbone for. They have Google Messenger and can easily add SIP based VoIP. WiMax will work with a future generation of phones, laptops and other devices. We can expect to see devices in 2009 and this is the earliest that Google would have spectrum.

The economics of WiMax are attractive and are truly disruptive wrt the incumbent carriers. The new spectrum coupled with WiMax create an opportunity to disrupt the market and Google knows this.

A user | Sep 17, 2007 | 1:14PM

Google, Apple, and Microsoft would all gain from having a ubiquitous and guaranteed neutral network. That would be a formidable trinity. I think that's actually predicted in the bible somewhere...

Matt Clary | Sep 17, 2007 | 2:12PM

Google and Apple makes more sense everyday, and I don't think for a minute Apple (Jobs)really cares what AT&T thinks since it will be competing with it for its customers local broadband wired services anyway when they get the 700Mhz.
If Google elects to wholesale the C Block spectrum they will find many Service Providers/Independnat Telco (ILEC) and maybe a CellCo ready to work with them to build a local 700Mhz network and lead with Google/Apple Content/Aps.
gPhone: I also think Google will provide an OS, Softphone and Search/Ads package for any of the vendors to incorporate into their new Dual Mode handhelds. If they do deliver hardware they might deliver a $100 Handset (Throw away if you will) for the low end of the market.
Verizon Wireless will not survive long term without the enhanced bandwidth provided by the 700Mhz (22Mhz) for their lame Narrowband (EV-DO)data network, so they will need to win this spectrum.

Titanic battle looms-do I hear $ 20 Billion Plus??


Jim | Sep 17, 2007 | 4:24PM

All that's needed is "Execution" and as the Bard says, herein lies the rub.

Google have incredible expertise on the server side - they have no expertise on the client side. Apple the masters of hardware built an incredible iPod - and then added some RF capability. They are still learning and they've had 20 years with operating systems and hardware.

Talk is cheap - what counts is execution - and just look at all the things they have going on, and ONLY ONE of them actually makes money.

Right now they're hunting for a second string to the fiddle.

Peter Cranstone | Sep 17, 2007 | 5:37PM

Google Cube. It sounds like a creepy episode of Star Trek with Steve Jobs as Locutus. ATnT will be assimilated. And we can watch it all on our Classic iPod.

Scott | Sep 17, 2007 | 7:59PM

New Google Ad: "Resistance is futile. Your life with cellphones is over. All your 700mhz spectrum are belong to us".

MattyD | Sep 17, 2007 | 9:23PM

Google/Apple/??? wins 700Mhz.
1. The Google and Apple team are missing only one major piece of the Worldwide Broadband Data/Content Distribution Network: The Last Mile.
2.Google does not need to develop a hardware gPhone, Apple is far better at that, and already has 2-3 devices for Mobile and in the Home Networks. Plus Google has seat on Apples board.
3. Google has the Data Centers (Fiber linked) and are preparing to offer a higher level (replacement)Internet Service to any Service Provider willing to work with them-Share their customer base. Fiber link to Google data Centers and with services distributed over both the Service Providers wired (FTTH/DSL etc) and the new the Wireless 700Mhz Net-allowing them to compete with the CellCos narrowband data nets.
4. Apple and Google both own and control (agreements inplace)major Content/Apps which can all be moved to and stored on Googles Data Centers, dramtically enhancing quality of delivery.
5. All this team needs is a Network builder to partner with them.
6. 700Mhz Spectrum, the power levels the FCC will allow for this licensed spectrum, and the 22Mhz available under Block C will allow a true 4G level network to be deployed nationwide. Unfortunately the 12 Mhz available in the remaining lower Block (A & B) are not sufficient for a real MultiMedia distribution network, unless they are combined with the Lower Block C frequencies (already sold).


Jacomo | Sep 18, 2007 | 11:17AM

Sounds reasonable and quite interesting. But the big question is, will they do it? It might also be just exactly what the US needs with respect to (a) the phone providers - cell and land, (b) ISPs, and (c) wireless comms. The US is severely behind in all three as the companies behind them do not want to spend the money to do the right upgrades (e.g. fiber to the home, true high-speed Internet, etc.) and this might just force them to do what needs to be done in order to compete. A new player is likely needed, and one that takes on all three would be really neat to see.

TemporalBeing | Sep 18, 2007 | 12:35PM

Google Cube == up-powered 700MHz Sprint Airave?


David Nessl | Sep 18, 2007 | 2:18PM

I own 4 ipods.. 2 old one two new ones...

I am using the old one because the new ones have to many bugs...

Not really don't buy one until at least 6 weeks from now...

1. cover flow is useless in 80 and 160 slow if you put too much music and photos.

2. photo is useless also since the icon shown are not the picture shown and the if you scroll pictures you get the black screen of death (freezes).


4.It (IPOD) vibrates in your hand when you look for something (HARD DISK SPINNING) crappy feeling feels like a vibrator...

5.sync erros by the 100'ss, photo icon click is not photo shown (1400 photos).

6. nobody tested the ipod with a lot of song and photos and video once is over 40 Gb it slows down to a crawl ...

7. video out does not work.. All othe ipod stuff you have became useless: Sh** my cool LCD eyeglases can not show video , I can not turn video ON...

shame apple looks like you are Microsoft 2....

AR | Sep 18, 2007 | 3:35PM

AR: Apple now requires that you buy a dock based adapter to get video out and it's hardware keyed so they can gouge you (Think TPM) or charge less to people who don't want that feature (think licensing fees that Apple pays). Seems like some of both to me.

Pecos Bill | Sep 18, 2007 | 6:41PM

AR: Apple now requires that you buy a dock based adapter to get video out and it's hardware keyed so they can gouge you (Think TPM) or charge less to people who don't want that feature (think licensing fees that Apple pays). Seems like some of both to me. I presume your LCD glasses can plug into the composite dock connector (possibly needing an adapter).

Pecos Bill | Sep 18, 2007 | 6:42PM

Having been directly involved in telco and the wi-mesh worlds I can only concur with your thoughts.

If you look back in your archives for LocustWorld you'll see that you became very excited even then... about the prospects of the wifi based mesh networks.

LocustWorld has taken things much much farther since then, having added, squid caching, meshTV, sip proxy, and many other rather forward thinking and extraordinary features. They also had the foresight to conceive of a media agnostic application of their mesh nodes, thus negating whether wifi, wi-max, fso, or fiber were the transmission medium chosen.

I previously believed that the community networks would swarm and eventually win out in the race to deliver broadband to the people, not so much anymore.

I 100% agree with your prophetic statements, but I think that in order for the real promise to be delivered upon, it will take more than just the 700mhz band as a technology catalyst.

If it is to truly happen, then we will need to see rapid development and deployment of frequency portability (SDR) radios and capabilities for custom waveform is deemed a necessary compliment to frequency portability.

This alone is the technology to incubate, for it will shatter the 'licensing of physics' as we know fcc licensing to be today.

Show me that google or any concerted REAL entity (I am aware of the GNUradio project) is making great strides in these areas, and then I will become once again giddy over the societal changes that ubiquitous broadband can bring forth to humanity.

(Societal Changes? yes.. reduction of fossil fuel consumption, reduction of green house gasses, vastly improving our nation's education systems by merging poorly performing school districts into successful programs via distance learning., etc)

I so wish the world all the luck in this endeavor, but in the end, I fear that will only serve to create a separation of networks (legacy vs modern/proprietary/open)... who's in charge of dns? hint hint

But that also brings me to my excitement of your prediction, and that is... ONLY google would have the capacity to "cache" the legacy network for portability of the content back into its newly deployed "consumer created" network.


1$ £1ƒ3 $0 Ð34r, 0r p34(3 $0 $w337, 4$ 70 b3 pµr(h4$3Ð 47 7h3 pr1(3 0ƒ (h41n$ 4nÐ $£4v3r¥? ƒ0rb1Ð 17, 4£m19h7¥ 90Ð! 1 |{n0w n07 wh47 (0µr$3 07h3r$ m4¥ 74|{3; bµ7 4$ ƒ0r m3, 91v3 m3 £1b3r7¥ 0r 91v3 m3 Ð347h! - Patrick Henry March 23rd, 1775

Mark Williams | Sep 19, 2007 | 5:59PM

Apple is still trying to dismiss the lawsuit? I can only venture that attorneys are still cheaper than buying them out, and I'm sure is looking for a big payday, either by settlement, buyout or winning this and future lawsuits. Unlike SCO in their lawsuits against IBM and potential Linux users, they actually have a case against not only Apple but every company or technology that has a media player in this market.

Kevin Kunreuther | Sep 19, 2007 | 8:57PM

Eric Schmidt said, "Wimax is coming" at Apple's keynote introduction of the iPhone. Journalists haven't paid a lot of attention to that comment (just google it), no matter that in the same breath he coined the term "applegoog" to describe how closely Google and Apple would collaborate--to "merge without merging," as he put it.

So your wifi prognostication seems way off to me. Wimax is superior in every relevant way (especialy QoS) and would make the best use of 700MHz spectrum. Besides, Google already has a partnership with Sprint, which has a partnership with Clearwire--the two Wimax telcos.

Why do you think Google's stock price would go down if they bought the spectrum and announced intentions to compete with telcos? I think it would skyrocket, enough to pay for the whole deal.

michael | Sep 20, 2007 | 1:37AM

the FCC won't let it happen because AT&T might lose market share and profit... and AT&T provides infp to the NSA.... so how will the gov't get their intel now... it won't be "complete"

mh | Sep 20, 2007 | 10:17AM

I'm with Rick. Weirdest Cringely title ever. The power of six what? I don't think a six-sided cube is an acceptable explanation of the title.

Come on, Bob, you got some `splainin' to do here.

Jeff | Sep 20, 2007 | 1:13PM

I'm with Rick. Weirdest Cringely title ever. The power of six what? I don't think a six-sided cube is an acceptable explanation of the title.

Come on, Bob, you got some `splainin' to do here.

Jeff | Sep 20, 2007 | 1:14PM

The thing I dread most about Verizon (or any other traditional ISP) winning the spectrum auction, is that they won't have to pay for the build out the infrastructure like you suggest. Like the build out of almost all previous infrastructures in the US, Verizon will successfully get a new "tax" passed that they will apply to all of their customers (whether wireless customers or not). So we will pay (again) for them to build out a new infrastructure (poorly, since there is no risk in the venture) whether we as customers plan to use it or not. All Verizon will have to do is front the money for the build out, and even that money might be put up by Wall Street given that they'll surely know of the coming new "tax".

Russ | Sep 20, 2007 | 3:54PM

"Cringely fiction" is about as useful as Japanese Godzilla stories. Why does this idiot make things up, then talk like we should wait for it to happen?

Despite this idiot's many many Google conspiracy articles, all Google does it sell ads, and mostly give alway a few weakly-featured network applications. That's it.

I'm still waiting for the hard drive that puts all the other drive makers out of business. LMAO...

Mkkby | Sep 20, 2007 | 7:47PM

Would you add into the mix Google's interest in FON, maybe as a kind of experiment?

Venkman | Sep 21, 2007 | 10:25AM

A shake-up in the industry has been needed for some time now.

Neil Anderson | Sep 23, 2007 | 9:04AM


Please call up any wireless expert and ask them HOW MUCH data you can actually push over 700Mhz. You'll be surprised, it isn't a whole lot.

Irving | Sep 25, 2007 | 3:19PM