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

The Pulpit
The Pulpit

<< [ To the Moon, Alyce! ]   |  You Can't Get There From Here  |   [ Oh, and We Also Saved the World ] >>

Weekly Column

You Can't Get There From Here: The myth of bandwidth scarcity and can Team Cringely really make it to the Moon?

Status: [CLOSED] comments (56)
By Robert X. Cringely
bob@cringely.com

Several years ago I wrote a column describing a system I had thought up for sharing Internet hotspots that I called WhyFi. Among the readers of that column were some entrepreneurs in Spain who went on to start the hotspot sharing service called FON, which now has more than 190,000 participating hotspots. Those Spaniards have been quite generous in attributing some of their inspiration to my column. And now this week FON signed a deal with British Telecom that promises to bring tens of thousands more FON hotspots to the UK and beyond. This isn't FON's first deal with a big broadband ISP -- they already have contracts with Speakeasy and Time Warner Cable in the U.S. among others -- but it is one of the biggest and points to an important transformation taking place in the way people communicate.

These FON deals remind me of President Lincoln's original Emancipation Proclamation from 1863, which I'm sure you'll recall didn't free ALL the slaves, just those in the rebellious states of the Confederacy. BT is not the largest mobile phone company in the UK, though it IS the largest ISP, so offering the ability for customers to make free or very cheap VoIP calls using FON software running on their home routers or that of another participating BT Fusion customer works well to tie those customers to BT ISP service, where the future of telephony clearly lies. It's good for customers, sure, and I am happy for them, but the reason BT and others do these deals is because it is BAD for their larger mobile competitors.

This is exactly why T-Mobile -- a smaller U.S. mobile carrier -- sells combined GSM/VoIP phones in the U.S. and the other carriers don't. T-Mobile, with no landline customers and no consumer ISP service, can only improve its U.S. business one way, by attracting more mobile customers from other carriers. Adding VoIP capability, while it may hurt mobile revenue a little, also costs T-Mobile almost nothing to provide, so its customer-attracting capability is justified. For exactly the same reason T-Mobile is happy to accept iPhone users who prefer to not use AT&T: it costs T-Mobile nothing more in support (because the phones are effectively unsupported) and costs AT&T mobile customers.

These games are played over and over by communication service companies of all kinds, and at the heart of it all is a big lie -- that bandwidth is scarce.

Bandwidth is not scarce. America and the world are bound by more fiber that is dark than is lighted. If backbones needed to be 10 times larger than they currently are, they quickly could be. On a local basis, the cost of provisioning a 1.5-megabit, 6-megabit, or 24-megabit DSL connection is essentially the same to the ISP, meaning the "bigger" pipes are vastly more profitable and that's all.

Bandwidth scarcity isn't peculiar to the United States, it is just managed differently overseas. You can get a 100-megabit-per-second Internet connection for $12 per month in Korea, sure, but it won't access most non-Korean Internet services any faster than a U.S. DSL or cable Internet connection could. There are few data resources anywhere, in fact, that can be accessed at such line speeds because it isn't in the economic interest of the ISP to make that much bandwidth available. It could be done fairly cheaply, but then who would pay more for a faster line?

Profit is to be found not just in pleasing dissatisfied customers, but in dissatisfying them in the first place so they will then pay to be pleased.

This isn't just a retail phenomenon, either. This week IBM announced to its workers that it is selling its network services business to AT&T. Will this please customers? Probably not. Will it put them in a position where they will pay extra to be pleased? Probably.

To this point AT&T has been just one of IBM's telecommunication service providers, but it has the noteworthy distinction of being by far the most expensive. Handing over the business to AT&T will not save IBM customers any money. IBM intends to continue to provide the same communication services to its customers, though now through AT&T, and AT&T says it expects to gain $1 billion in sales. Those sales have to come from somewhere, and in this case they will come from taking business from lower-cost providers.

What the deal WILL do for IBM is get 2,200 workers off the books cheaply (that $80 million charge against earnings AT&T is taking for the deal will mainly go for picking up the underfunded pension obligations of the transferred IBM workers). It will be good for the workers, too, because they'll be going to a company with a real benefits package and a solid pension plan.

But it won't be good for customers, whose charges will only go up if an IBM sales commission will now be pasted atop overpriced services that were, in the past, often sold at a loss. AT&T is not in the business of selling at a loss. This will have the effect for IBM customers of firmly defying the trend toward lower communication service costs, which maybe was AT&T's whole intent in the first place.

Whatever the cockeyed logical basis of this transaction, it will do nothing to change the deliberate bandwidth scarcity problem that plagues businesses and consumers alike worldwide. The only way to solve THAT problem is by taking back ownership of our own last-mile connections and creating a true competitive marketplace for backbone services. It's a move that would pay for itself almost instantly, but I doubt that it will ever be allowed to happen.

And speaking of radical change, last week's column about my plan to vie for the Google Lunar X Prize produced a flurry of interest and more than 50 project volunteers, including a few true rocket scientists (though, alas, not a single woman). We'll have important announcements concerning Team Cringely two weeks from now after the Lunar X Prize committee issues its final contest guidelines and we've had a chance to map those rules against whatever cheating techniques we can think up.

I never said Team Cringely would win gracefully.

But I do think we have a chance to win and I'll use this space to address the considerable skepticism that greeted my announcement last week.

Here is a typical message from a real space expert:

"The way it looks now I'd be embarrassed out of my skull to even have my name mentioned in association with this project. I can't even begin to describe just how clueless this team looks to anyone with any real knowledge about these things. Sorry if this sounds harsh, but it really is THAT bad.

"You can barely build and launch a nanosatellite for the budget they're talking about. And they're talking of developing and building THEIR OWN F**ING LAUNCH VEHICLE as part of the project? May I remind you that no privately funded rocket has ever made it into orbit yet and the leading contender (SpaceX) has spent over $100M so far?

"The minimum budget that seems remotely realistic to achieve the prize goals is well over $15M, assuming you get experienced individuals (i.e. people who built things that actually flew in space) to work for free, somehow make it incredibly lightweight and manage to find a really sweet launch deal. But this is for an unconstrained schedule. Achieving the goal FIRST will be significantly more expensive than that. Remember that the suborbital Xprize was won by a very well funded team. Their budget was estimated to be over $35M - bigger than the prize. The next closest team was VERY far behind in their progress. Armadillo Aerospace probably could have done it for a budget lower than the prize but it would have taken them a few more years."

This is Bob again. The guy above may be right, but I look at this project as a chance to break some rules and do things a little differently, possibly at the risk of looking foolish, but also possibly making some significant progress.

If fear of embarrassment keeps this guy from thinking out of the box, Team Cringely doesn't want him, though we will still buy him a beer.

Our air launch system is well proven by both the U.S. and Russian military, both of which use it to shoot down satellites. Our rocket engine has been in development for nine years and fired over 300 times without a failure. Our f**king launch vehicle, as he puts it, is a carbon fiber tube. All the other parts come from Russia, where they have been used on similar-sized rockets for decades. We want to put 30kg on the Moon and, according to the same calculators NASA uses, it looks like we can. Rutan and Company spent a lot of money, yes, and because they did we don't have to spend as much. PLUS, there are a lot of good ideas about how to accomplish this mission that won't look anything like what this guy thinks we'll be doing.

I'm not saying it will be easy, I am saying it can be done, and is worth trying to do.

Comments from the Tribe

Status: [CLOSED] read all comments (56)

I, Cringely:

Hello! Me, Anthony here. Well, light up some of that dark fiber, for Pete's sake, and get us going at light speed I say. All comes down to profit, don't it? Maybe Bill and Melinda will decide to help out...or Paul Newman can come up with a new product line to provide bits and bytes to the digital poor? Love your stuff. I cross-posted on your piece to http://blog.innovators-network.org The Innovators Network is a non-profit dedicated to bringing technology to startups, small businesses, non-profits, venture capitalists and intellectual property experts. Please visit us and help grown our community!

Best wishes for continued success,

Anthony Kuhn
Innovators Network

Anthony Kuhn | Oct 09, 2007 | 4:41PM

I love your column, Bob, but I disagree with you this week about the bandwidth scarcity issue, mainly because available fiber is not the limiting factor for bandwidth. All that dark fiber is unusable until it is "lit up," the extensive process for which I have outlined in my responsive post (http://blog.vpisystems.com/22/bandwidth-scarcity-myth/). There is, unfortunately, a tremendous investment of time and resources necessary to transform dark fiber into bandwidth. Saying it "just" needs to be lit is like saying that the oil in the ground "just" needs to be pumped, refined, and distributed to fuel my car.

As for your perspective on carriers creating a market of dissatisfied customers, I fear it may come to pass. Most carriers this week at the World Broadband Forum in Berlin reported that they feel that the current model of them continuing to add enough bandwidth to keep good QoS for best effort traffic is "unsustainable" as a business model.

Thanks for sharing your insights every week (and keep reaching for the moon)!

Best regards,
Mark Mortensen
VPIsystems

Dr. Mark H Mortensen | Oct 10, 2007 | 3:41PM

> though, alas, not a single woman

So much for Bob's other goal of joining the 238,857 mile high club.

paulwesterberg | Oct 12, 2007 | 11:17AM