Mapping Strategy

Sometimes The Best Strategy is Patience

From a recent IBM press release [emphasis added]:

A complex physics calculation that will take [the new] Roadrunner [supercomputer] one week to complete, would have taken the 1998 machine 20 years...

Put that in the back of your mind. Now step back and ponder the following hypothetical situation:

You're a business executive (or government official) facing a big, gnarly problem that absolutely, positively must be solved in ten years. If you don't make the deadline, something bad will happen. You'll go out of business, you'll fail to win re-election, your shareholders will revolt, the Huns will invade. Something bad. You've got to solve this problem. Your career (and maybe more) depend on it.

With current technologies, processes and general know-how though, you know you can't make it in time. Sure, you can reach the goal -- in theory, eventually -- but it's going to be really really expensive. And it's certain you're going to be late. Ten years late, in this case.

The problem will get solved, but the Huns will be running the place by that point. You'll be out of a job, maybe worse: lawsuit, jail, early retirement, HBR case studies that make you look like a complete idiot and laughingstock in hindsight... that sort of thing. Your grandkids will still love you... until they get to business school and read the case. But I digress.

So, you have two choices: 

Start throwing wheelbarrows of cash and thousands of bodies at the problem. Or invest judiciously in innovation -- and, at least as importantly, an innovation culture -- and wait patiently for those to bear fruit. (There's a third option: waiting for others to innovate for you, but we'll leave that aside for now because it only applies in limited situations where you can be fairly certain someone is working on the thing you need and that you'll be able to buy whatever it is they come up with.)

In either case, (#2 or #3, that is) you may have to wait years. You won't have much to tell your shareholders in the meantime, other than "we've hired the best, given them the best resources, incented them up the wazoo and tried not to be too meddlesome -- those creative folks don't take well to corporate bureaucrats, y'know." But when something does come along, you'll be able to really impress them: "You remember that problem we thought would take twenty years to solve? We finally got started on it, and we're targeting completion... next Wednesday." Sounds like promotion material, doesn't it? And your grandkids won't be laughing at you when they get to business school.

The tough part, of course, is all the verbal bobbing and weaving for the first ten years when nothing seems to be happening. No highly visible hordes of minions toting 10-ton blocks of stone up pyramids on wooden rollers (and dying by the dozens, getting you in trouble with OSHA). Just a bunch of folks wandering around in lab coats (or, more likely, surf shorts and goatees, playing foos-ball between all-night stints at the white-board).

Sometimes waiting (and investing) is better than charging ahead, using up all of your cash on old, dumb, slow methods that you know will be a day late and a dollar short.

H/T: Irving Wladawsky-Berger

Forward Into the Past! (A Tale of How Not to Listen to Your Customers)

Jason Fry's 'Real Time' column in today's WSJ is worth checking out (subscription required).  He writes:

Earlier this month, Nielsen Soundscan released 2007 data painting yet another portrait of a music industry struggling to make the transition to a more-uncertain, less-profitable business model. For the year, overall units sold rose 14% -- and sales of digital tracks rose 45%. But not all units are created equal: Total album sales fell 15%, and that blow was softened by a 53% rise in sales of digital albums. Subtract digital-album sales, and physical-album sales declined 19%, to 450.5 million in 2007 from 555.6 million in 2006. (Dig into the numbers yourself here¹.)

Or, if you prefer, take this arresting anecdote from the Economist² (found via Nate Anderson's interesting read³, in the always-great Ars Technica): In 2006 EMI honchos invited some London teens to the label's headquarters to discuss their listening habits. By way of thanks, the kids were offered whatever they wanted from a big pile of CDs. Offered free CDs, the kids took … nothing.

[Bold added; links and italics in the original.]

And why should they take anything? The perceived price of zero isn't even the issue, as more folks fall into line and eschew P2P sharing networks in favor of obeying the law. My daughter's latest iPod is the size of a cookie (and not those super-sized ones they sell in airports, or even Oreos, but more like a thin Chips Ahoy). Continuing to sell CDs into this climate is like selling 78s when I was young.

The Hidden Complexities of 'No-Brainer' Decisions: Off-Shoring, Wage Inflation and Those Stubborn Circadian Rhythms

The front page of Tuesday's W$J featured an interesting piece on the hidden challenges of hiring workers in faraway places, especially when the goal is saving money ("Some in Silicon Valley Begin to Sour on India") Emphasis added:

Last year, Mr. Shah paid his engineers in India about half of Silicon Valley levels. By early this year, it was 75%... India's software-and-service association puts wage inflation in its industry at 10% to 15% a year. Some tech executives say it's closer to 50%. In the U.S., wage inflation in the software sector is under 3%, according to Moody's Economy.com.

Rafiq Dossani, a scholar at Stanford University's Asia-Pacific Research Center who recently studied the Indian market, found that while most Indian technology workers' wages remain low -- an average $5,000 a year for a new engineer with little experience -- the experienced engineers Silicon Valley companies covet can now cost $60,000 to $100,000 a year. "For the top-level talent, there's an equalization," he says...

Some Indian outsourcing companies are themselves looking to other countries -- mostly as a response to the globalizing nature of their business, but also to tap new labor pools amid a tight hiring situation at home. TCS recently opened a center in Mexico and is planning to move into Morocco. Wipro has two centers in China and is thinking about adding one in the Philippines...

[Wage] increases... have spurred a lot of job-hopping in India. Pervasive Software Inc. of Austin, Texas, opened a Bangalore unit in 2004 and hired 45 people. But soon its turnover was more than 25% a year, says the company's CEO, John Farr. The company kept having to invest in training workers, only to see them leave. A year ago, it shut its Bangalore unit.

Hidden outsourcing costs surfaced for other tech companies as well. To bridge the geographic and time gaps, some have found they need to hire more U.S. managers to handle their Indian teams. Kana Software in Menlo Park has one engineering manager for every 25 to 50 engineers, but it found it needed one for every five to 10 engineers it employed in the Indian city of Chennai. In December 2005, Kana decided to close its Indian operation.

Mr. Khan, Riya's California-based vice president of engineering, says he often stayed up until 4 a.m. so he could talk with the team in Bangalore. Mr. Shah, the chief executive, flew to India six times a year to make sure things were running smoothly.

Straddling time zones slowed development work. Sometimes a researcher gets stuck on a problem. "It's not really a good thing to keep bugging people at midnight every day, so that introduces some delays," says Mr. Dalal. He also felt cut off. "For us sitting here in India, it's hard to get the business aspect of the problem," he says...

Shutting down in India isn't cheap. Teneros, which closed a 30-person operation in New Delhi in late 2005, says it spent $2 million to do so. It had to get out of contracts, and it brought 12 of its Indian workers over to the U.S. on work visas, incurring immigration fees. Teneros left partly because of wage inflation and a lack of information-technology infrastructure that was slowing its work, says its CEO, Steve Lewis. Mr. Shah says Riya's shutdown costs, such as immigration charges and a broken lease, will be in six figures.

He has cleared eight desks in San Mateo for the eight Indian engineers coming over and is waiting for their paperwork to clear, hoping they'll be in Silicon Valley by the end of the year. "I thought I understood India," Mr. Shah says, "but now I know it's so much more effort to have a remote office as a start-up."

The article is instructive in its particulars. One of the things I find most fascinating are the hidden risks  some may be incurring in dealing with a partner in one  highly stable country ('A') who is turning around and outsourcing to countries B, C, D, and E, each of which is exposed to far more significant risks. (Pop quiz: what former world power used to be talked about enthusiastically in the 1990's as a major source for technical talent... but isn't much anymore? Hint#1: they have some of the highest rates of alcoholism, TB and premature death on the planet. Hint#2: they're sometimes referred to as a deflationary authoritarian kleptocracy with nukes. Hint#3: they use the Cyrillic alphabet.)

As both customer and service-provider, I've dealt with companies operating in India (as well as some headquartered there). The former experience, e.g. talking to a remote call center, is far less unique than it used to be. Few in the U.S. have not been forced at some point to try and deal politely and expeditiously with the intangible but very real challenges of accent, culture, pacing and shared interpersonal communication conventions easy to take for granted--until they're missing.

Dell's withdrawal from India is the most well-known case illustrating that even if all aspects of it cannot be fully defined in advance, customer experience eventually translates to the bottom line--for good or for ill. There's something jarring and difficult that I've never quite been able to put my finger on about sitting down at the end of a long day and dealing with some poor sleep-deprived third-shift worker in India. It's just as difficult--in a different way--to re-read my 2AM e-mail response to a colleague on the other side of the world over coffee the following morning  and wince a la Hunter S. Thompson at how poorly my brain must have been working when I wrote it. It may be true that money never sleeps, but like it not we must.

I find the article even more interesting however as a lesson in the increasing fluidity of a global economy and the fundamental uncertainties and complexities of contingent strategic decision-making on such a stage.

Painting with a deliberately broad brush for a moment, I'll assert that difficult-to-revoke choices entailing 'big-bet' consequences tended--in the past--to match the scale of uncertainties with the resources the decision maker had at his or her disposal.

A small business (or her banker) needing to choose a retail location for (say) a bakery has always been able, without much trouble, to scope out the key dimensions of site risk (e.g., sidewalk traffic, local competition, parking, utilities, rent, wage rates for counter help, etc.). By contrast, large multi-national corporations have always been able to afford (whether they chose to or not) to devote several individuals--if not an entire planning department--to analyzing a vast array of long-term economic scenarios and political risks for operating around the globe.

In today's super-connected world however, it's much easier for smaller companies to find themselves making massively risky geopolitical decisions without fully realizing they are doing so and without the ability to assess or the scale and diversity to hedge such decisions. The anecdotes in the WSJ article prove that out.

The point? It's easy to read about globalization and come to the premature conclusion that one is missing out on riches by not plunging in headfirst and surfing a trend that's pervasive. As a friend and former colleague enthusiastically put it to me recently: the future is all about Dubai, Mumbai and Shanghai! That's fine insofar as it goes. This post is not about ignoring globalization but about assessing its complexities and the myriad strategic options it presents with thorough research, appropriate reverence for what is unknowable and sufficient thought about alternatives and contingencies.

Depending on one's business, Dubai, Mumbai and Shanghai may be fine choices--provided one stays aware of (respectively) increasing political and cultural extremism, the aforementioned inflationary flies in the ointment of international wage arbitrage, and the perniciously arbitrary nature of operating in a Communist country.

All may be risks worth taking if one goes in with one's eyes open. Unfortunately some don't. Sound bytes and rumors about the grass being greener make for lousy strategic planning.

Recursive PMs: Using the Crowd for Cover

In the middle of May I speculated--reasoning from first principles only and with no inside information--about how prediction markets might help the book industry pick winners:

"For some types of books (and authors) prediction markets might be appropriate."

That sparked a note from a reader, tipping me off that:

"Confidentially, something's about to launch very soon in this area."

Sure enough, MediaPredict launched a week later, eliciting this highly positive piece in the 5/21 NYTimes: "Publisher to Let the Public Have a Vote on Book Projects".

Media Predict is soliciting book proposals from agents and the public, and posting pages of them on the site. Traders, who are given $5,000 in fantasy cash, can buy shares based on their guess about whether a particular book proposal is likely to get a deal, or whether Touchstone Books, an imprint of Simon & Schuster, will select it as a finalist in a contest called Project Publish. If either happens within a four-month period, the value of the shares go to $100 apiece; if not, the share price falls to zero.

The devil as they say, is in the details. Traders are voting not on how popular a work could become with the book-buying public but with the binary, intermediated and likely recursive question of whether the editors at Simon and Schuster will select it.

In other words, S&S editors are going to be breathing their own exhaust--now with more butt-covering camouflage. (Which reminds me a little of this.) It's the kind of conversation I've seen some couples get into:

Man (off-handedly and with a smile): What would you like for dinner, honey?
Woman: Oh, whatever you'd prefer is just fine with me!
Man (thinking burgers and beers with the guys): I'd prefer what pleases you, dear.
Woman: Well I don't know. What were you thinking?
Man: I was thinking I'd like to know what you're thinking.

And 'round and 'round and 'round it goes. That's an imperfect and terribly stereotyped analogy, and no, it's not my happy marriage. Well, OK sometimes maybe it is. But let's move on, shall we?   :)

The more abstract version of that analysis is this: The system S&S has set up with MediaPredict has no outside, arbitrating reference point--such as how many copies a particular book would actually sell.

Hey, now there's an idea: ask the market directly about its spending enthusiasm for a given title. (Print-on-demand technology, not to mention digital rights management, render completely moot the objection that only a few select books can be produced. The hurdle revolves around marketing and a legacy business model of annointing a few winners.)

Unfortunately, S&S's Touchstone Books does not appear to take that important dis-intermediating step. I doubt it's MediaPredict's fault. I suspect (without any proof except my experience with how Fortune 500 executives think and in particular how publishing executives think) that the use of prediction markets may have been de-clawed in this case to avoid dis-enfranchising S&S editors. Think about it:

The PM predicts Bestseller status for a book the editors decide to pass up. It gets picked up by another publisher, fulfilling the PM's expectations and making bundles of money for an S&S competitor. At that point, S&S editors not only have egg on their face, but the S&S board (and shareholders) are emboldened to ask: Tell us why exactly we need you if all you're going to do is to second-guess and stand in the way of this smart little prediction market thingy we've set up over here to perform the same function?

Justin Wolfers (respected PM guru at Wharton) notes the same basic flaw, though you'd have to read to the end of the article to find his critique.

[Wolfers] said that if Simon & Schuster relies on the traders' judgments to select a book, it could skew the bets themselves.

"If they say we find it really persuasive that everyone bet on book A, they're just looking at a book that everyone bet that everyone else bet that everyone else thinks is the best book," Mr. Wolfers said. "So you don't end up with the wisdom of crowds, but the infinite reflection of crowds looking at crowds."

This is at least the second time I've noticed a major media outlet burying what should have been the lead on a prediction markets story--and with it, Wolfers' critique. If I were him, I'd be a little peeved. Last fall, the Washington Post carried a piece ("The Top Pickers vs. The Pack") that completely mashed two irreconcilable ideas, as I noted in "WaPo Misguided on 'Experts' vs. The Swarm":

The article takes the academically and empirically well-proven notion of collective intelligence (aka the wisdom of crowds), and presumptively twists it into the more comforting and familiar (if statistically invalid) notion that "father-knows-best"--only we haven't done a good enough job of finding the right "father" up until now.

I don't know why the media's treatment of prediction markets is missing the mark, but they need to get smart faster. PMs are not going away. Not taking seriously the critiques of Wolfers and others isn't helping. The NYTimes digs itself further into a hole by taking a flawed comparison at face value:

Media Predict is modeled after other so-called prediction markets like the Hollywood Stock Exchange, which allows traders to bet on the four-week North American box office receipts of movies, or the Iowa Electronic Markets, which allow people to bet on election results...

The glaring difference: the HSX and the IEM ask traders to vote on highly distributed end-result questions. As I've observed many times (e.g., here, here, here and here) prediction markets--while very cool indeed, and applicable to many more things than those to which they are being applied today--are just plain lousy at calling things like Papal elections or the nomination of judges where information is closely held by a select group. I suspect it's even worse when that group has sponsored the market and each is trying to predict the actions and preferences of the other.

If the same principle were to be applied to American Idol, voters would weigh in with their opinions not on which singer(s) they liked best, but on which one(s) they thought Simon Cowell was going to like. That's not entirely without its entertainment value, but it adds little to what Mr. Cowell was going to say anyway. Simon (& Schuster) could take a lesson from what the other Simon (Cowell) lacks: humility. The crowd is often right. If they're paying, they're even more right.

As for MediaPredict itself, I'm reserving judgment. At this stage of market development, more new players trying more approaches, covering more questions and attracting more corporate attention is a good thing--and entirely expected. As the MP folks write on their blog: "Love us or hate us, we’re here…" That may be a little extreme. How about: I love that you're here.

Fallacies of Telecom Network Resilience: Lessons From the Taiwan Earthquake

Picking up from my post on the same topic earlier this week, more is becoming clear about Internet and telecoms network failures to and within Asia as a result of Tuesday's major earthquake(s) in and around Taiwan.

• Boats are in place and the fix is already underway, contrary to earlier reports indicating that the cables wouldn't even be touched until after the turn of the new year.


• Laws designed to keep fishing vessels from dragging undersea cables carry fines pitifully out of proportion to the potential damage (roughly a million-to-one). It's not clear why, in a nation (China) not shy about capital-punishment, someone has not lost their head over previous cable cuts (e.g., April 6, 2004).


• The capacity of land-lines from Europe to Asia is tiny in comparison with those from the U.S. to Asia and the U.S. to Europe. This global asymmetry  raises a host of issues ranging from economic development to politics to outsourcing to culture.


• Satellites are expensive and have only limited capacity. The illusion that they offer backup to undersea cables is just that--little better than thinking that carrier pigeons will fill the gap.

With that and other emerging news, it's also becoming clear that lessons learned after 9-11 haven't 'stuck'.

Within the IT-savvy portion of the New York financial community at least (and, I would have imagined, well beyond it) a great deal was learned, publicly documented and put into practice to address the fragility of such networks. The cost was not small, however the there was unanimous agreement that the alternative was worse. The direct and indirect costs of closed markets (four days in the case of 9-11) were understood to be vastly larger--something that many in Asia are rediscovering to their collective chagrin.

The lessons of 9-11 (from a business resilience perspective anyway) are not particular to New York, to the U.S.,  or even to a cause. I'll rephrase and repeat that last bit, because without justification, thinking about business resilience is too often segmented by what went bump in the night rather than what happened as a result of it.

The business resilience lessons of 9-11 apply as well to natural as to man-made disasters; as well to widespread as to 'point' impacts; and as well to protracted as to short-duration events.

The lessons are immutable--essential principles for building and maintaining resilient networks and organizations. Two biggies are:

Understand layer one. Don't assume that a carrier's logical architecture (much less it's marketing architecture) showing a ring, double ring, or even mesh network has anything to do with physical reality. Both 9-11 and the Taiwan earthquake illustrate how economic pressures and crowd dynamics drive networks towards similar if not identical paths-of-least-resistance. (I suspect that carrier assurances in articles like this are not really about physical routes.) This is no more the fault of a particular company or person than is Interstate 80 through the California Sierras, the Golden Gate Bridge across the mouth of San Francisco Bay or the majority of the U.S. financial community on one tiny island (Manhattan). Those were simply the most logical  and least costly places to put those things. So it is under the ocean... 

Get a cross-carrier view. Don't assume that because you have signed up with three carriers that your data (or voice) traffic is moving via three distinct networks, much less three physically dispersed paths. Inter-carrier 'grooming' agreements can spring up without any legal requirement for end customers to be notified.  This is what happened on 9-11. In a similar vein, don't assume that the Internet itself is resilient because it was designed to withstand a nuclear attack. See point one: physical routes matter.

I'll address other laws of business resilience in subsequent posts.

Counterfeiting An Entire Company

Anyone who's ever walked down the street in Manhattan for more than five minutes knows about brand counterfeiting in watches, handbags, CDs and the like. Now here's something much bigger and more insidious that large organizations in any consumer-touching business may need to put on their list of difficult-to-fix strategic threats they never imagined they'd have to think about:

After two years and thousands of hours of investigation in conjunction with law enforcement agencies in China, Taiwan and Japan, [NEC] said it had uncovered something far more ambitious than clandestine workshops turning out inferior copies of NEC products.

The pirates were faking the entire company.

Evidence seized in raids on 18 factories and warehouses in China and Taiwan over the past year showed that the counterfeiters had set up what amounted to a parallel NEC brand with links to a network of more than 50 electronics factories in China, Hong Kong and Taiwan.

In the name of NEC, the pirates copied NEC products and went as far as developing their own range of consumer electronic products - everything from home entertainment centres to MP3 players.

They also co-ordinated manufacturing and distribution, collecting all the proceeds. The Japanese company even received complaints about products which were of generally good quality that they did not make or provide with warranties. [emphasis added]

Vertical vs. Horizontal Thinking

Zenpundit Mark comments on a post at the Eide Neurolearning Blog, on vertical vs. horizontal thinking. Both are well worth reading for the distinctions they draw between deep expertise and the holistic, often intuitive, cross-disciplinary thinking that comes naturally to strategists and those with "the vision thing":

Complexity in thinking is, strictly speaking, different from the type of reasoning derived from having a high level of expertise in a given field, often referred to as vertical thinking. 

Be sure to click through to Mark's full post at his new Newsvine site.

UPDATE: Readers can find more thoughtful commentary on this concept over at Steve DeAngelis' Enterprise Resilience Management blog:

If I understand the Eides correctly, some complex thinkers create a mini-Medici Effect in their minds... We normally think of creating a Medici Effect by bringing vertical thinkers together in a group to generate innovative solutions to a particular problem or to generate entirely new thinking. I would be willing to bet that individuals who successfully facilitate those gatherings are, more often than not, horizontal thinkers. [link and emphasis added]

That works for us! Even better is this comment by DeAngelis:

Horizontal thinkers see and understand horizontal scenarios better than other people. What does this have to do with resiliency? A lot... Most businesses have intentionally set up vertical organizations. By that I mean they have separate siloed departments that deal with specific, specialized functions. As I have argued repeatedly, however, today's paradigm requires horizontal sharing of information across department boundaries...

By no means does that denigrate the importance vertical thinking (which furthers knowledge), vertical organizations or industries (which promote efficiency), or vertical action (which responds to vertical scenarios), but it does underscore the fact that within and between organizations more (not less) horizontal interaction needs to take place. Every organization needs a horizontal thinker who can make sense of these relationships and help establish a proper balance between them. Why? Because, if the Eides are correct, horizontal thinking may be a capability that is either learned early in childhood or is simply innate. [emphasis added]

The ERM blog has occasionally put its founder's business (Enterra Solutions) overmuch in the spotlight for our taste. (The line between a corporate website and a blog is a fine, subjective and constantly evolving one.) With this post however, DeAngelis seems to be breaking out and finding his blogging 'voice'. His passion for the underlying concepts and deep knowledge of the unmet needs shine through without the 'pitch' getting in the way quite as much.

In any case, given that Steve is a dedicated and highly successful serial entrepreneur, we'll take the pitch as passion. He's building it and they are coming. Again.

I share with Steve a strong conviction that out there in the larger world, understanding of resilience (much less resilience management) is still very much a concept in its infancy, even as the term has come into widespread use since 9-11. Much remains to be done before organizations truly "get it" up and down the 'stack' - from infrastructure, through people, all the way up to long-term strategy (e.g., portfolio and innovation). That's another post altogether...

Exchanges (Finally) Go Global

It's unfortunate that the WSJ chose to run this fact-packed, big-picture story (subscription required) on the Saturday of Memorial Day weekend. Being the strategy geeks that we are however, we were all over it. The story charts the recent sea-change from human-driven, city-centric, product-specific, cooperatively owned stock exchanges to global electronic powerhouses trading in a wide range of securities and derivatives.

[U.S.] exchanges find themselves motivated by a new phenomenon: Big overseas companies no longer see the U.S. as a place where they must list their shares to raise capital. Tougher U.S. regulation of companies and the ease with which capital can flow around the world have made it attractive for companies to list closer to home. Next week, the Bank of China is preparing a nearly $10 billion offering that will take place on the Hong Kong Stock Exchange... In 2000, nine of every 10 dollars raised by non-U.S. companies outside their domestic markets was through U.S. exchanges... By last year, only one in 10 such dollars was raised in New York. [emphasis added]

As the WSJ notes, that's a truly radical change. Some of it is no doubt driven by exchange rates, but only some. Another major factor has been increased regulatory oversight in the U.S., (e.g., Sarbanes-Oxley), providing a sobering lesson in the unintended consequences of well-meaning legislation in a fluid, free-market global economy.

Massive, permanent structural shifts like this are what we refer to when we talk about "discontinuous change" and the need for creative scenario thinking in order to survive it. Linear forecasting and trend-spotting does little good. One day the world looks much like it always has and business-as-usual works just fine... and then one day it doesn't. Credit the NYSE's post-Grasso leadership with seeing and responding to these changes as best they can.

In early 2004, when John Thain arrived to take the helm, he quickly discovered that the NYSE had serious competitive issues. Specializing in stocks, it was less electronic than its peers. Visits to Chicago and Europe showed how the New York exchange had fallen behind innovative rivals such as the CME, Euronext and Deutsche Börse. Mr. Thain invited one NYSE staffer to his office a few months after he arrived and told him that the exchange had to go public, build a derivatives business and go overseas... [emphasis added]

Better late than never. The other thing they've been needing to do, which the article also notes, has been to marginalize and ultimately eliminate a byzantine human-dependent floor-trading system that by preserving antiquated pre-computer trading methods only lines the pockets of seat-holders and specialists. Other Mapping Strategy posts on NYSE doings can be found here, here, here, here, here, and especially here ("Competition and Terrorism in the Securities Industry").

One Billion Internet Users

A significant milestone has been passed. Ironically, the website of the firm that released the survey is down - presumably due to a traffic spike from some of those one billion users.

For perspective the top five sports audiences of all time are:

1. 2002 FIFA World Cup Final (1.3 billion viewers)
2. Euro 2004 Soccer Final (153 million viewers)
3. 2004 Olympic Games: opening ceremony (127 million viewers)
4. 2004 Olympic Games: closing ceremony (96 million viewers)
5. 2004 Super Bowl (95 million viewers)

More perspective: The Wall Street Journal's 'Numbers Guy' column (Carl Bialik) estimates that there are only 1-2 billion working television sets in the world (yes the spread is really that wide) and 2 billion radios. Furthermore...

...the highest rated U.S. television show of all time -- the final episode of "M*A*S*H" on Feb. 28, 1983 -- was seen by "only" 105.5 million people, according to Nielsen Media Research.

A Beatles enthusiast begs to differ. And lest we get carried away with such competing claims, all based on different, unstated assumptions, 'Numbers Guy' notes further that:

...numbers [in the multi-billions] for the Olympics and the Oscars [and Princess Diana's funeral] are... estimates of potential audience, not actual viewership. Ditto for the oft-repeated claim that one billion people watch the Super Bowl each year. So why do these TV audience figures become gospel? Blame a mix of the media's hunger for numbers (especially very large ones), organizers' tendencies to hype events with the biggest audience figures possible, and a dearth of better data. [emphasis added]

The point(s)? 1) Only twelve years after the invention of the world wide web, the Internet has just begun to have its full impact on society, commerce and the world. 2) It's frighteningly easy to get numbers into popular circulation (far easier than words, I've found). Once there, they often take on a life of their own when no credible authority exists to say otherwise.

Perspective on Predictions

Buried deep in a wide-ranging WaPo article on U.S. statistics (derived from "the 28.5-pound, $825, five-volume Historical Statistics of the United States, Millennial Edition... [so-called] because it's the first major update of this monumental work since the Bicentennial Edition, 31 years ago.") is this gem:

...too much faith in numbers has consequences. In 1980, IBM predicted that the total market for personal computers in the following 10 years would be something like that year's 246,000. Not 250,000, but a nicely precise, confidence-inspiring 246,000. The actual outcome was almost exactly 100 times greater. If the leaders of IBM had been more optimistic, would they have ever dreamed of letting their operating system be supplied by some snot-nosed kid named Gates?

How different the world might have been if IBM had not been in thrall to such over-precise, linear assumptions. All we know for sure is that the future will be... different. Then there is this more philosophical observation on sense-making and prediction:

Counting things may have started as an act of faith, giving us control over the universe. If you can count the days, you can determine how much food and firewood you'll need until spring returns, giving you control over nature, or so the builders of Aztec temples and Stonehenge believed.But the world doesn't really look like numbers, although it does obey Einstein's laws.

One of the editors of the Historical Statistics book notes that:

"The world is a buzzingly complicated thing. To make sense of it, we abstract. We don't tell you everything we see and smell and hear. We pick out what we think is important. Think of it as a painting or a poem. It may be meaningful, beautiful or evoking emotion. But it is incomplete. That's why Magritte, in his painting of an apple, wrote below it, in French, 'This is not an apple.' Social scientists would say the same thing. These are abstractions with a purpose, to say something insightful or provocative about concerns people have today." [emphasis added]

The article continues with more implications for scenario thinking:

That's why thumbing through the Millennial Edition is like cleaning out your mother's attic and finding a big old box of snapshots. You compulsively make connections. Was that Grandfather Alphonse? What a bizarre moustache. No wonder cousin Fred looks the way he does. This photo of a woman in a tennis outfit. Was that Aunt Agnes before she became a nun? Wow. Great legs. The photos are pretty random and not much of a narrative until they are made into an album, probably with the help of an older relative. You can piece this album together however you wish. But until then, you don't have a story of who you are and how you got that way. All you've got is a box of historical snapshots. [emphasis added]