Wednesday, December 23, 2009

Early Indications December 2009: Yet Another Predictions Issue

In this last week of 2009, it's scary to think that it was a full ten years ago that the IT profession was holding its collective breath as midnight January 1, 2000 approached. Apart from spooking us with memories of how fast the decade sped by, the Y2K issue stands as a cautionary tale for any technology prediction.

Duly chastened, I remain intrepid, with 24 questions for the coming decade. The alphabetical mnemonic I last used in 2005 cues up a question for each letter, minus the usual suspects.

Having brilliantly migrated from computers to MPs players to mobile data devices, what will Apple do for its next adjacent market? Tablet rumors surface almost weekly, Apple TV has yet to fulfill its promise, and such areas as health (iDoc?) are huge in potential. In any case, it's difficult to see Apple hitting its revenue growth numbers without an addition to the product portfolio at some point.

In case you missed it, DARPA conducted a brilliant experiment last month. Ten red weather balloons were tethered in plain sight at various locations around the country, and teams competed to supply the latitude and longitude of each one using social networking technologies. MIT won in nine hours, an amazing accomplishment considering a) the continental U.S. presents a surface area of over 3,000,000 square miles and b) teams worked to spoof each other. In the end, MIT's clever compensation model to attract the widest interest group of observers helped secure the win. In light of that experiment and its many findings, the B is for business models, specifically for the plethora of social media tools that are exploding in popularity. Not to put too fine a point on the matter, but this past year Facebook alone grew at a pace of 770,000 new users -- a day.

Here are some surprising numbers: Brazil's GDP per capita income, in constant dollars, has risen 52% in the last ten years. Singapore's is up 79%. Chile has gone up 59%. These numbers, chosen at random, illustrate the emergence of a global middle class (the C). Such groups are historically important, typically signaling political stability, economic growth, and increased presence in international trade markets. Who else will join these countries, Korea, and other fast-growing economies? How will the world change with these new entries into the economic and cultural mainstream? (For contrast, U.S. GDP per capita in 2005 dollars rose only 17% between 1998 and 2008, and class-related tensions could be big news going forward, whether in regard to labor unions, health care, the 2012 election, or unemployment.)

The D question relates to design. As the documentary of the same name makes clear, the "modern" presence of the Helvetica typeface (or its Ariel cousin-once-removed) is now more than 50 years old, yet it remains ubiquitous. Apple has of course capitalized on great design, and the slowdown in consumer spending in the U.S. in particular may be an indication that people are buying from a less disposable mindset. If people buy less, they may follow a generally European pattern and buy better designed items. For all of these reasons and others, the time is ripe for a design renaissance on par with streamlined toasters, or the neo-Bauhaus movement that poured so much concrete in the 1960s.

Drugs also merit mention. Marijuana is simultaneously a) being legalized under medical provisions in 13 states and counting, b) being decriminalized in some states, and c) contributing to political destabilization on both sides of the Mexican border. As states battle increasing social welfare and other costs in a time of declining revenues, taxing pot holds at least some appeal. In addition, mandatory sentencing laws are crowding prisons and generating hardened gang members at a staggering expense that many states simply may not be able to afford: $24,000 per year per inmate, not counting potential foregone wages and other indirect expenses. If, as The Onion memorably put it, "Drugs Win Drug War," what alternative strategies might be pursued instead?

As more of the world's citizens want automobiles, and electric lighting, and central heating, and meat in their diets (see C), the demand and competition for energy sources will intensify. That energy, usually provided by burning something, will in turn play into the global climate debate. Whether in oil prices, coal emissions debates, or nuclear power lobbying efforts, competition for energy will have geopolitical consequences, potentially including more armed ones.

Football (world football, not the U.S. version) will be huge news in 2010 as the World Cup is contested in South Africa. Apart from the intense fan interest in both powerhouses and upstarts, the role of mobile and new media will bear watching. Far more people own cell phones than own televisions, so the deluge of texts, Tweets, and web-hosted highlight clips could be a global coming-out party for social media, just as the 1958 NFL championship game or the JFK assassination were for television.

It's difficult to think of a G bigger than Google. The question before us relates to the company's many efforts to expand its presence (and eventually its revenue base) beyond the lucrative search franchise. Will the Android mobile, or the location-based ad service, or the office applications, or some new innovation break through to profitability? How will copyright-holders react to potentially universal access to their work?

H is for housing. The economic impact of the shelter industry is of course considerable, and everybody is watching home prices for both personal and analytical reasons. Beyond sales figures, however, some larger forces are coming into play. Demographically, the baby boomers now entering retirement (or an approximation thereof) want and need different things from real estate, and it will be a while until a later wave has enough children, income, and interest to buy up the empty-nesters' housing stock. In addition, as U.S. income stagnates, the average house size will likely retreat from its high point of circa 2005.

Identity is increasingly something people actively manage. What's your relationship status? How are you feeling today? What do you think about sports, politics, other people, your possessions? At the same time, lightweight and incredibly powerful tools lower the barriers to association. Whatever one's interests, whether obscure, weird, or outright criminal, finding like-minded individuals is now possible in ways that were simply inconceivable in physical space. As more people grow up breathing the oxygen of online, all-the-time social broadcasting, what will be the unintended consequences, the business opportunities, and the backlash?

Building on the July letter, jobs remain at center stage. How much will this recession prove to be an interruption in the way things were, and how much will it prove to mark a shift in underlying forces of globalization, the balance of product- and services-based work, or long-term costs and benefits of modes of agriculture, consumption habits, and population pyramids? What are the odds that GM, Citibank, or Sears -- and the industries they represent -- will return to their positions of past dominance? More likely, but similarly daunting, is the question as to how entrepreneurs could possibly generate tens of millions of new jobs, on any continent.

Kindlemania is in full flower, driving the publishing industry to confront some long-held assumptions. Back in May, CEO Jeff Bezos announced that Kindle sales had hit 35% of book sales when Kindle editions are available for a given title. In a matter of months, Amazon has disrupted 500 years of relatively stable technology that dated to Gutenberg. The implications will be all around us. At Princeton, for example, a trial using Kindles for textbooks was problematic insofar as page numbers (and footnotes to page numbers) needed to be rethought. Searching a textbook is useful; not being able to use sticky notes requires getting used to.

Long tails make the list -- no surprise, in the age of YouTube and eBay. What's interesting is the Economist's assertion that fat tails (hit movies or blockbuster drugs) are remaining as vital as ever. The surprising conclusion appears to be that the middle market could turn out to be no-mans-land, as the Harry Potters and Transformers movies (the latest of which merely grossed over $400 million) dominate the mass market while endless, hard-to-serve niches proliferate elsewhere.

In the developing world and the OECD countries alike, mobility is not only redefining the telecom sector, as major as that may be. In addition, the notion of always being reachable, or becoming accustomed to connecting to people rather than fixed locations, is becoming commonplace so fast that we may not realize all that is happening. Worldwide, the number of cellphone subscriptions per 100 people has soared from just over five in 1998 to nearly 60 in 2008. In the midst of it, this change can be lost in fashion wars (RAZR vs. iPhone vs. Blackberry Pearl, or whatever), but eventually, in hindsight, we will see the magnitude of what we lived through.

News is moving in new ways to new people. The broadcast model is augmented (not replaced) by millions of electronic conversations. The utility of owning a big antenna, a printing press, or a television studio has dropped precipitously as lightweight digital equivalents proliferate. Even though free societies need reliable news, at a time when such countries confront complex debates over everything from immigration to climate to aging to employment, the business model for news is highly unsettled. The conundrum of the need for news and the problem of organizations' being able to afford to report and provide it must be resolved, and such efforts as Google's Living Stories experiment with the NY Times and Washington Post will, I hope, spawn still more innovation.

O is for open book, shorthand for the myriad of issues relating to privacy and scrutiny. Open records, or open meetings, laws were never intended to broadcast local, paper-based information to the entire planet. At the same time, "sunshine is the best disinfectant," as Louis Brandeis so aptly put it. How and where will different people and groups trade off voluntary and involuntary exposure of private information for what perceived benefits? How will generationality play out, especially as data turned loose in one's early years may be uncomfortably or even dangerously revealing later, with different attitudes, tools, and agendas in play 10 or 20 years from now?

Given the speed and magnitude of the changes afoot, and given the essential characteristics of "being digital" as Nicholas Negroponte titled it, competition is playing out not just between products (Dell's PCs versus HP's or Lenovo's), it is also evolving to situate competing platforms (the P word). The choice between Nintendo Wii and Playstation 3, between an iPhone and a Nokia, or between a Chevy Volt and a Toyota Prius are more complicated than merely deciding on features and price. What are the two ecosystems -- of accessory makers, of software developers, or of product owners (and so of a current or future secondary market)? How will future innovations be incorporated into today's purchase? Will Google establish a beachhead in the browser as a quasi-operating system, on the mobile device, or in mapping? Where is Microsoft (see S)? Will still more industries begin to exhibit platform dynamics?

While the phrase "real time" is not new, the advent of people-powered notification means that rather than coming from capital-intensive air-traffic control, equities trading, or medical monitoring systems, real-time data is now the product of real people. Whether in natural disasters, social movements, or just a dozen families attending an out-of-town soccer tournament, the spread of lightweight, mobile coordination mechanisms will soon make many of us wonder how we ever got along without them.

The software industry is at a crossroads: enterprise vendors still work on adjusting the mix between license and maintenance revenue, between hosted and premise-resident installations, and between consumer, middle-market, and large enterprise sectors. Software as a Service sounds great in theory, but Salesforce still has bugs to work out (regarding scale, for one thing), and the industry is still in search of other viable exemplars. In consumer markets, meanwhile, the days of CompUSA or Computer City being the dominant channels for distribution of diskettes or CDs are over: Apple's app store model has redefined developer programs and consumer software distribution essentially overnight. Open- and closed-source models are still being sorted out. With so many dimensions of the business up for grabs, who will emerge in the coming years? Who will be left behind? What further surprises still await?

T is for thermostats, a proxy for an entire class of inanimate objects and devices that are increasing the reach and complexity of the global network. Whether implemented for energy savings, human comfort and well-being, or security reasons, building automation joins health monitoring, security cameras, and a vast number of other devices in a quietly but rapidly growing "Internet of things." While this domain frequently lacks glamour, the possibilities for drone vehicles, for dramatic cost and energy savings, and for increased human welfare (via care-giving robots for instance) verge on the realm of science fiction.

Whether in the U.S. or elsewhere, the place of universities is being questioned. While California's 32% tuition increase grabbed headlines and motivated nearly nostalgic building takeovers, the fact is that California education remains underpriced. The University of Texas, by comparison, has raised fees 60% in the past five years whereas California has a cumulative increase of only 20%. Such numbers appear to be unsustainable, raising the question of what will be cut when dramatic spending decisions will have to be made in the coming decade. One-time budget relief from the stimulus package is similarly unsustainable, while long-term curriculum directions scream out for reassessment. As desirable as it might be to add labor relations, African-American studies, or forensic science to the course catalog, how can universities simultaneously a) steer resources toward the future, b) respect their role as custodians of the past, and c) keep expenses under control? Classics is a frequent target for programmatic termination, but what about sociology, recreation management, or broadcast journalism? Does the U.S. need more than 200 law schools? Who decides? How? At both public and private institutions, the next decade will force tough decisions to be made.

While virtualization is a widely used term of art among computer architects, my sense here is broader: Webster's Second defines virtual as "being in essence or effect, but not in fact." Not only are computing resources not resident at the point of use, neither are people for more and more tasks. Very few people could work by telephone from their homes, yet today one's physical presence and one's "essence or effect" can be many miles and time zones apart. Whether in dating, or education, or telecommuting, or elder care, we are seeing the start of a particular kind of disembodiment: just as Descartes split mind and body for the individual, will some latter-day philosopher distinguish physically co-located groups and digitally "present" assemblages?

Whereas in M we discussed what it means for people to be mobile, the W refers to the coming demand for wireless bandwidth. On every populated continent, we're seeing dramatic increases in mobile data and telephony. AT&T is confronting the problem of the iPhone's success as its data networks are at times showing signs of overload. Countries from Pakistan to Estonia are leapfrogging wireline infrastructure, at which they never reached mass-market penetration, and getting the majority of these country's households connected via wireless in less than a decade. By contrast, it took nearly 100 years to bring 100 million wired phones into service in the U.S., at the time a nation of 200 million. As usual, there is no free lunch, and we will be seeing radio spectrum continue to be a political hot potato. Whether in regard to suspicions (not yet confirmed) about heath issues, to spectrum auction formats, to "interference" with other activities on other frequencies, wireless demand is driving a shortage that is invisible and intangible - until the call drops or the application crashes.

What is an electronic game? Despite the success of Modern Warfare ($550 million in sales in five days), console platforms such as the X-box find themselves in competition not only with each other but with unlikely channels: Electronic Arts (maker of Madden and other category-leading titles) laid off 1,500 people in November, while web-hosted low-resolution, lightweight games (often running in Flash) can command vast audiences. The Scrabble knock-off Scrabulous help drive Facebook's early growth, while more recently Farmville counts 73 million players per month. Put another way, Farmville grew to 11 million daily users in two months; World of Warcraft took four years to hit the same figure. Just as MP3 files convinced listeners to trade convenience for fidelity, perhaps the game industry will see further segmentation between low-resolution (but heavily social) Flash games and high-fidelity, computationally-intensive titles.

There you have it, minus entries for Q, Y, and Z (Scrabble value: 24). Additional questions of course remain, particularly in the areas of nutrition (water is a likely battleground), health (obesity, medical education, step-function gains in bureaucratic efficiency, and pharmaceutical risk/reward allocation), and aging: the time is due for an honest debate about age-65 retirement, and the role of families, villages, and societies in the care of elders.

Before any of these issues unfold further, I send my personal best wishes for a peaceful holiday and a prosperous new year. The community of readers has become virtual (see V) family over the years, and I take it as a solemn responsibility that so many of you keep reading and commenting. Thank you, and blessed holidays.

Monday, November 23, 2009

November 2009 Early Indications: Prediction Scorecard

As is our custom, every November we revisit the previous year's predictions. Given that one of the dominant themes of 2009 has been stagnation -- of reform legislation, of job growth, of the housing market, how well did last December's outlook see into the future?

The predictions were divided into global and domestic spheres. Taking the former first, we noted that "a globalized world creates a new category of issue that requires multi-lateral response well beyond the scope of traditional definitions of sovereignty" and predicted that "with so much room between the cracks of law, enforcement, and reporting, expect to see more global equivalents of dropped fly balls in 2009."

Score: Hit

On the climate change front, expectations for the Copenhagen Conference next month are being managed downward. At the juncture of international crime and terror, as we noted last year, Mexico's particular mix of drug trafficking, organized crime, and para-military groups leaves it with a unique problem blending terror, crime, and unstable diplomacy to both north and south. Civil institutions including hospitals, schools, the press, and of course law enforcement are under brutal and continued attack. Given that the drugs are destined for the U.S. and Canada and originate, in many cases, in South America, a multilateral solution is needed but has yet to take shape.

The new reality of global information flows that transcend jurisdiction was illustrated vividly -- in strikingly different ways -- by the role of electronic media in the Iranian protests and the Indian elections. So-called social media are continuing to challenge the role of the nation-state with newly-mobilized ethnic and/or virtual communities with impressive powers of persuasion and coordination.

In the use of unmanned aerial vehicles (UAVs) such as the Predator to launch missiles at individuals suspected of terror connections, the U.S. is opening a new kind of warfare made possible by emerging technologies. The technologies and policies also illustrate the challenge of aligning innovation with written and unwritten rules of engagement. In the same year that Bush-era plans for targeted assassinations drew sharp rebukes, the fact that Obama-era UAVs often both miss or misidentify their targets and cause civilian casualties is a hot issue in Pakistan but little discussed in the U.S.; the attacks have been defined as "executions without trial" by Pakistanis and others.

The cost in political support and stability versus the purported benefit of killing terror leaders has yet to be debated or justified. The lack of debate notwithstanding, the situation amounts to a moral, legal, strategic, and tactical no-man's-land. In Pakistan last month, Secretary of State Clinton was confronted by angry audiences decrying the attacks, but she did not respond except to refuse to comment. The point here is that the combination of asymmetric warfare, new technologies, and new military tactics is raising entirely new sets of issues. For example, what is the U.S. liability for erroneous attacks? Under what jurisdiction and what laws do the operators and commanders of robotic assets fall? What are the human costs to contractors and other non-military personnel who in the morning drive to a nondescript air-conditioned facility, watch people (including civilians and children) die by their remote control thousands of miles away, then drive home at night to their families?

The Domestic Conundrum

In the aftermath of the Obama election, we "expect[ed] to see some combination of strong efforts that will have the effect of attacking boundaries between problems. Four key areas in particular are often attached: health care, demographics, consumer spending, and asset markets."

Score: Not much has happened.

We have more and better analyses outlining why health care reform is so necessary and so hard, but zero legislation to date. Regarding demographics, the high unemployment rate will affect all age groups, but falling fertility will counteract immigration in important ways, at least in the U.S. Even with the run-up in equity prices, I'm still quite worried about how the baby boom generation will retire: "It's not a 2009 prediction, but I believe a bail-out will eventually be required to address a massive shortfall between long lives and small retirement accounts. Unlike health care reform, or bank bail-outs, or wars, demographic change typically takes decades to unfold."

On consumer spending, retail continues to be depressed, nudged up temporarily by the Cash for Clunkers subsidies. Housing remains soft. Given high un- and underemployment and the sharp falloff in home equity cash-outs from the days of the real estate bubble, the pattern of change in personal consumption expenditures, as measured by the Bureau of Economic Analysis, is impossible to extrapolate.

As for soft asset markets, the climb in stock-market indexes remains somewhat mysterious. Systemic issues of risk and reward, executive pay, and bankruptcy are still live, and will not be settled for some time. I agree with The Atlantic's Andrew Sullivan: the massive burst of stimulus money that will be expended next year ($9 million for a pedestrian bridge at the [privately owned] Gillette Stadium complex outside Boston, to take a random sample) could well play a significant role in the midterm elections.

Five secondary questions

Score: too early to tell

I asked five questions that have longer timetables than one year. We got a few hints in 2009, but no clear answers.

-What will the civil rights movement of 2012 look like?
**While court-ordered recognition of same-sex unions was overturned by popular vote in Maine, heavily-Mormon Salt Lake City passed anti-discrimination legislation in the off-year election, and the state of Washington (led by Seattle-Tacoma) upheld Referendum 71, the "everything but marriage" guarantee of gay and lesbian rights. Might "everything but marriage" serve as a template for other same-sex-union ballot initiatives? The place, rights, and role of Latin Americans in North America, meanwhile, are all changing rapidly, and 2012 could well feature substantial debate over immigration, education, health care, and other issues as they unfold in the Latino/Latina community.

-How many non-profits will lay off social workers, administrators, and the like, adding to the unemployment rolls next year?
**Apart from the toll of the Madoff scam on non-profits, there's little regular reporting on the economic health of this sector. Food banks, job training centers, and alternative energy groups appear to be busy, but I have no sense of the long-term directions here.

-Can this semi-private philanthropic (e.g. Gates, Omidyar) sector outperform the NIH in finding a cancer cure, or the WHO in mass inoculations, or big pharma in breakthrough drug discovery?
**The global response to the H1N1 virus has been instructive. Social media, mash-ups (most including mapping), and online video are all in the arsenal of the CDC, the better to counter potential hysteria. (the philanthropic arm) is measuring search terms to extrapolate on influenza patterns. Thus far, activity appears to have peaked in October, whereas in six previous flu seasons, activity spiked in December (much more sharply) in 2003 and in February very other year. If it proves reliable, that kind of real-time tracking will introduce new elements to the practice of public health.

-What forces can reinvigorate American manufacturing?
**Many analysts propose clean energy as the key driver here: a short list might include windmills, smart electrical grids, mass transit, nuclear plants, batteries and charging stations, and energy-efficient building practices. For strategic or logistical reasons, all of these might be economically produced and deployed on home soil rather than be imported from China and elsewhere: shipping costs and the other implications of long, energy-intensive supply chains keep near-shore locomotive production, for example, viable. How government can best encourage this trend, however, remains to be seen: existing interests, such as GM, appear to have occupied far more attention than innovators and entrepreneurs.

-If present-day blogging isn't capable of replacing formerly great newspapers, what comes next?
**The key problem here was the formulation of the question: blogging and social media will not replace newspapers or cable networks. Rather, multiple media architectures will work in tandem. We saw this phenomenon last month in a particular college football broadcast: a University of Florida player gouged the eyes of a Georgia player at the bottom of a pileup, and the commentators saw and said nothing. Instant replay and high definition broadcasts, however, allowed a viewer in Pennsylvania to see the action away from the play, record it, and forward the clip. Thousands of Tweets and rants later, the action made it onto the highlight shows, the league and team took disciplinary action, and the party formally known as the audience, as Dan Gillmor put it, controlled the news cycle in one domain, for a few days. Big questions remain about investigative reporting, about foreign bureaus, and about credibility, but it's clear that the new media landscape will alter reporting, and entertainment, and leisure time, and institutional memory, and many other sectors besides.

Overall, the predictions that depended on the presidency were too high, underestimating the time it takes to form an administration, align congressional forces, and balance day-to-day crises with long-term vision. At the same time, the continuing rise of non-nation-state actors is facilitating new kinds of action, causing new kinds of problems, and challenging existing entities (whether the UN, the Indian navy, or the U.S. Department of Justice) to evolve.

Leaving behind the aughts, or whatever we end up calling this decade, what lies ahead in 2010? Watch for the annual predictions next month.

Saturday, October 31, 2009

Early Indications October 2009: The Exploding Mobile Web

This newsletter is about numbers, specifically these:







We'll take these in turn.

According to Morgan Stanley's Mary Meeker, whose Web 2.0 presentation should be required reading, the iPhone and iTouch surpassed 50 million units shipped in 9 quarters after launch. This would make it the fastest technology adoption, as measured by zero-50 million, in recorded history. Netscape's Navigator reportedly had 38 million downloads in 18 months, but that could include double-and triple-counting. In addition, Netscape's Internet distribution model allowed it a substantial advantage over conventional logistics, while Apple physically moved all those devices. Actually, an even faster technology adoption was a wireline phone feature: as I wrote in 2005, the federal Do Not Call registry surpassed 55 million users in less than a year, but no software or devices changed hands.

The iPhone has spurred a vast ecosystem of software developers. According to App Shopper, more than 103,000 applications for the iPhone have been approved. While about 20,000 are free and the average selling price is $3.25 for paid apps, GPS add-ons from MobileNavigator and TomTom sell for nearly $100. Other top sellers include mobile editions of both conventional (Uno) and electronic (Madden) games. By contrast, the Google Android Marketplace has 10,000 applications, followed by Blackberry App World at 3,000, Nokia's Ovi Store at 660, and Windows Mobile Marketplace with 246 -- less than 3/10s of 1 percent of the leader.

Those applications are helping drive truly staggering demands on bandwidth. Cisco estimates that global mobile bandwidth demand will increase 66 times (!) in the next four years. Based on AT&T's experience, that number is fully believable:
mobile data traffic has increased 4,962% (essentially 50 times) in less than three years. The wide dissatisfaction with iPhone performance is often blamed on AT&T's network, but provisioning that kind of growth would tax any organization.

As much as the iPhone has stressed the cellular network, the picture would be far worse if wi-fi, which is essentially ten times faster, had not picked up so much of the load. According to AdMob, between 40 and 60% of iPhone data transfer occurs over these ad hoc networks, which were not built with a government stimulus package, a spectrum purchase, or a conscious deployment plan. This offloading of bandwidth may explain why Verizon is allowing its new Google phone, the Droid, to connect to wi-fi, but it is not clear under what conditions or with what fees.

That last number, 100,000, approximates the size of the character set to be allowed by ICANN for Internet domain names in character-based (non-Latin) languages. Right now every root server can look up millions of domain names based on 37 characters: the Latin alphabet, 10 digits, and the hyphen. Starting next year, Cyrillic, Arabic, and other character-based languages will begin to be included.

Given how much of the world a) uses character-based language and b) how fast wireless Internet is penetrating the developing world, the implications of character set for handset design will be fascinating to watch: a RIM Blackberry with 6,000 Chinese characters is not in anyone's future, I don't think.

The expansion to character-based languages may have a deeper implication for the mobile Web. The essential structure of semantic programming is based on a subject (Ridley Scott) - predicate (directed) - object (Blade Runner) model. The semantic triple allows data to be handled more flexibly than in relational databases, where relationships need to be known at the time of the schema's creation. Triplesets, by contrast, can be expanded to form graphs (Blade Runner - grossed - $33,000,000, Ridley Scott - directed - Harrison Ford) in order that web data can become queryable (how many Oscar-winning actors did Ridley Scott direct?). As smartphones become the Internet access device for much of the developing world, how will the various semantics of their many languages inform the deeper structure of Web data and data retrieval? Down the road, the non-Latin Web may have implications for Oracle, Amazon, and IBM at the same time that it challenges carriers and device companies.

On the handset front, meanwhile, Apple has the Graffiti experiment to learn from, along with strong developer momentum as it confronts the question of how to reach the next 50 million users, and the next 50 million after that. Motorola, HTC, Samsung, and Nokia, meanwhile, each bring a distinctive package of strengths and weaknesses to the table as they fight for market share in a global contest for hardware supremacy in a new order. Whatever happens, we will be confronted by growth rates the likes of which no manager (or capital market) has ever seen, each with their own raft of unintended consequences.

For more on semantics, see Toby Segaran, Colin Evans, and Jamie Taylor, Programming the Semantic Web (Sebastopol, O'Reilly, 2009).

Wednesday, September 30, 2009

Early Indications September 2009: Universities and an Information Economy

Apart from asserting, with Emil Faber, that "Knowledge is good," there
is little one can say with certainty about the role of the university
in an information, rather than industrial, economy. The connections
are many and complex, with as many exceptions as rules. My examples
will all be from the U.S., which is based on my merely anecdotal
understanding of promising developments in Australia, the Nordics, the
Arab world, and elsewhere.

In the light of 50 or so years that mark the rise of information and
services and the decline in the role of manufacturing in the U.S.
economy, what can we say about U.S. colleges and universities?

1. American colleges and universities have grown faster than
population growth. According to the U.S. Department of Education,
758,000 people enrolled in college in 1960. The post-war "baby boom"
kicked in shortly thereafter, as the number of high school graduates
rose by 64% in the following decade. As high schools turned out more
graduates every year, a greater percentage of them attended college.
In 1960, 45% of high school graduates attended college. Every year
after 1980, at least half did so, and since 2005, 2/3 of high school
graduates attend college. Accordingly, college enrollment has soared,
from 3.6 million in 1959 to 17.8 million in 2006, an increase of
almost 400%. In addition, the number of institutions, mostly branch
campuses of existing institutions I suspect, more than doubled in
those same years. By comparison, U.S. population in that period rose only 67%.

2. In the past half-century, American higher education has grown more
female. In the academic year 1959-60, men outnumbered women among the entering fall enrollment 2.3 million to 1.3 million. Only 20 years later, women had taken the lead 5.9 million to 5.7 million -- and note how much larger the entering class became in only a generation! In 2006, women outnumbered men 10.2 million to 7.5 million. Women have held or currently hold the presidency at every type of institution, including MIT, Duke, Harvard, Penn, Chicago, Michigan, and Princeton. Faculties, however, are still more male than female.

3. Tuition has grown much faster than inflation. According to the
College Board, since 1958 tuition has risen by less than the Consumer
Price Index in only 6 years, all in the 1970s, including one year when
the CPI jumped 11.5% In a typical year, tuition rises by twice
general inflation, and in some years it has gone up 4, 5, or 6 times
the CPI. Numerous explanations have been proposed, but the situation
is so big and so complex, it's likely nobody knows for sure.

Some considerations in that spending growth:

-Universities are heavy in human resources, and so have to pass along
increases in health care and related benefits costs. At the same
time, in part driven by a glut of Ph.Ds in the 1980s and '90s, many
universities have shifted teaching loads away from tenure-track
faculty to fixed-term, adjunct, and other types of instructors, which
should deliver some measure of cost savings.

-Headcount growth over the past 50 years has included many
administrative positions to oversee student life, minority affairs,
grant compliance, record-keeping (including IT posts), and alumni
affairs/fundraising. In contrast, teaching faculty has grown more
slowly than student populations: the ratio of American students to
faculty in 1959 was 9.6 to 1, while today it is over 13.5 to 1, nearly
50% higher.

-Medical schools and teaching hospitals are vastly complex and almost
universally lose money. Salaries can be astronomical - a Columbia
University dermatology professor makes over $4 million a year, and
heads of health systems can also earn seven figures. The university's
role in the emerging health care funding model has yet to be
determined, but it will be significant.

-To compete for students, universities have invested heavily in
facilities such as student unions, dorms, and fitness centers. To
compete for grants, universities have invested in researchers,
post-docs, and facilities; big science, in particular, is expensive.
Capital spending is an area of major growth, as a trip to your local
campus will reveal.

-Finding real estate to hold new buildings can be an issue,
particularly at urban institutions: Columbia and NYU are tightly
constrained, while Harvard's expansion into Boston across the Charles
from Cambridge is both expensive and controversial. Given the
performance of the university's investments, it is also on hold.

Harvard's Ivy League rivals in New Haven have had better fortune. In
2007 Yale purchased 137 acres of land home to 17 buildings totaling
1.5 million square feet from Bayer; three of the largest research
buildings were less than ten years old and would have cost about $700
per square foot to build. Large warehouse spaces, meanwhile, are
climate-controlled, making them potentially attractive for
conservation of art, scientific specimen, and manuscript collections.
Like many university expansion projects, such as Carolina North, which
is two miles from UNC's main campus, Yale will have to work through
the integration issues of separate facilities.

4. By definition, universities' responses to job supply and demand
lags the market. PhDs continue to graduate in fields with few or no
job openings. Nursing schools cannot fill demand, in part because of
a shortage in nursing professors. Some majors will surge in
popularity -- crime scene investigation and forensic science rode the
wave of TV detective shows, for example -- while others such as
accounting are perennially strong. Efforts to connect higher
education to job growth have had only mixed success: the number of
moving parts, and the lead times involved, and the degree of foresight
required on all sides of the table, make such a project a daunting

5. Private universities are increasing their research role. For
years, Texas, California, and Michigan led the nation in Ph.D.
production. According to Washington Monthly's tables, the leaders are
now Berkeley, MIT, Stanford, Michigan, and Illinois. In terms of
research dollars won, Johns Hopkins is on top, followed by University
of Washington, Penn, Michigan, and Stanford. Powerful medical schools
tend to be the constant on this list.

6. Universities drive high quality of life. In such listings as
Cities Ranked and Rated, Money Magazine's Best Places to Live, and
others, college towns consistently score well. Pollution tends to be
low, cultural activities abound, teaching hospitals provide excellent
health care, and commutes tend to be short. Many such towns are so
desirable, however, that real estate prices drive a high cost of
living. Boulder, Charlottesville, and Hanover, NH have seen this
phenomenon, and Palo Alto is off-the-charts expensive. The impact on
adjunct faculty and university support staff usually means that food
service workers, administrative assistants, and HVAC technicians must
live far from campus in many college towns.

7. Documenting university economic contributions is difficult.
Stanford's record of launching companies is probably unmatched in
recent history: roughly 4,700 companies have been founded by graduates
and faculty, including tech heavyweights Cisco, Google, HP, Nvidia,
Sun Microsystems, and Yahoo!. Reaching farther back, MIT's company
list includes Gillette, Texas Instruments, McDonnell Douglas,
Raytheon, Bose, and Genentech. In terms of royalties deals,
Northwestern sold its rights to the pain drug Lyrica for $700 million
in 2008. Gatorade has earned between $80 and $100 million for the
University of Florida since its invention in the 1960s.

Economic mobility is harder to track than IP licensing or
entrepreneurial impact. Washington Monthly ranks colleges and
universities by the percentage of students receiving low-income Pell
Grants, and by the predicted graduation rate of those recipients. On
this measure, the leaders are not Penn or Chicago, but South Carolina
State, Penn State, and UC-Davis. Harvard, Stanford, Johns Hopkins,
Duke, and MIT have negative scores on this index, suggesting they
function more to reinforce class rigidity than drive social mobility.

8. Universities are not particularly transparent. Unlike public
corporations, government bureaus and universities (both public and
private) do not report their activities in any standardized form. The
University of California, for example, presented 180 pages of detailed
budget documentation to the Board of Regents that is public. The
University of Michigan's public budget documentation, meanwhile,
consists of a one-page summary of all major units that add up to $1.4
billion. For all of that state's economic turmoil, the university
budget is up over 3% for FY2010, including an increase of 12% in
spending on "ceremonial and presidential events."

9. University athletic programs are big businesses on their own, most
of them operating in the red. A few athletic departments turn a
profit: Michigan's brought in about $90 million last year and passed
$1.6 to the University's general fund. The Southeastern Conference
(state universities from South Carolina to Arkansas to Kentucky, plus
Vanderbilt) recently signed a contract with ESPN for $2.25 billion
over 15 years. Including another deal with CBS, the league will
receive over $200 million a year in media rights, not counting bowl
games. Payrolls at the top end are in line with entertainment scale:
the University of Alabama football coach makes an estimated $5 million
a year, including cars, country club memberships, use of the
university airplane, etc. A rough rule of thumb at football-centric
schools is that the head football coach earns four times what the
university president does, but at Iowa and Oklahoma, the multiplier is
about nine.

10. Going forward, colleges and universities face multiple challenges
that will shape the society and economy in both short and long terms.
Given the nation's front-burner issues with wars and terror, budget
deficits, economic regulation and stimulus, and health care policy,
education is not the stuff of blogs, talk radio, or Senate hearings.
A long list of macro-level questions will demand action at some
juncture, however:

-Tuition cost increases have priced private education beyond the
middle class, not just at elite institutions: cost is not proportional
to quality. Total annual fees at St. Olaf college in Minnesota, to
take a good school at random, are estimated at $45,600 compared to
$50,600 at Princeton -- and Princeton's massive endowment ensures that
its aid packages are better. The Bucknells, Loyolas, and Oberlins are
faced with a severe challenge in that most costs are fixed: dorms have
to be heated and professors paid whether the entering class is 100% or
80% full. State universities face a different set of issues given
their vast scale and the changing revenue patterns for state tax

-High tuitions paid by wealthy families are to some extent a subsidy
to those on financial aid. Elite universities do a good job of
running that process as a "black box," but assuming continued tuition
growth, there will be increasing dissatisfaction with the arrangement
among some of those who are (on paper, if not in wallet) rich enough
to pay full freight. Once again, issues of social class are deeply
ingrained with educational institutions.

-Curricula may change more slowly than markets, but they must change
nonetheless, and the process is (for some good reasons and some
worrisome reasons) slow and politically sensitive.

-The customer satisfaction mentality, in which students are
"consuming" their education, has both upsides and downsides. As
tuition soars, it's impossible not to take seriously student concerns
about return on their and their parents' investment. At the same
time, education is not a vacation with professors serving as
recreation directors, and the customer is not always right.

-How will universities cope with globalization? Population growth and
market need are most vigorous in the developing world, and a wide
variety of partnerships is in place for U.S. institutions to tap into
such countries as India, Spain, and China. Will top universities
franchise their name, open branches overseas, dramatically expand
online offerings and experiences, or develop entirely new business

-Demography is destiny, and a decline in the college-age cadre appears
to be coming coming in a few years: the U.S. population pyramid shows
fewer 10-year-olds than 15-year-olds.

-Do 2/3 of graduating seniors really need to attend college? If
tuition is priced out of reach for the middle class, and if the
education premium is shown to be lower than believed, colleges and
universities will be further pinched between high costs and decreasing
demand. The wild card here is of course international students, who
are already an increasing presence at many institutions.

For all their challenges, American colleges and universities are
over-represented among the best in the world. According to the Center
for World-Class Universities at Shanghai Jiao Tong University, U.S.
institutions occupy 17 of the top 20 positions, and 36 of the top 50.
Such excellence occurs despite low math and science test scores among
high school students, despite troubling patterns in research funding,
and despite foundational shifts in U.S. social, regional, and economic
patterns. Going forward, the biggest challenge may be balancing pride
in the magnitude of our achievements with forward thinking to meet the
needs of the next generation of research questions, employment trends,
and entering classes.

Monday, August 31, 2009

Early Indications August 2009: Informational Geography

As the U.S. economy has shifted away first from agriculture and then
manufacturing as its core activity, numerous side effects have
emerged. In this newsletter we'll look at some of these, particularly
in regard to land and space issues. As we have witnessed, the
information age closely parallels the emergence of services as the
primary economic driver: the first commercial computer application
(payroll) was installed at GE's appliance operation in Louisville in
1955, so we'll look at roughly the past 50 years.

Macro Trends in Employment
As for the shift to a "services economy," the Bureau of Economic
Analysis numbers (sampled at 5-year intervals) tell several stories,
two of which focus attention on the 1970s. First of all, if we look
at "personal consumption expenditures," which are separated from
investment, the U.S. crossed over from spending more on goods to more
on services in 1970. From that 1:1 ratio, the momentum stayed with
services, to the point where in 2005 services spending approached a
2:1 advantage over goods. Second, at about the same time the U.S.
swung from a trade surplus to a trade deficit of roughly the same size
in only 5 years, a 181% swing between 1975 and 1980. Finally, and
significantly, we still export more products (70% of the total) than
services (30%).

From the employment side, was manufacturing ever dominant? As of
1919, according to the 1950 Statistical Abstract of the United States,
mining, construction, and manufacturing constituted only 47% of the
non-agricultural work force. Transportation and utilities, retail,
finance, services (as in auto repair but not domestic servants) and
government (excluding armed forces) added up to the remaining 53%. As
of 1959, according to the Bureau of Labor Statistics, the 19 million
goods-producing jobs unsurprisingly lagged services jobs, which
totaled 34 million, 8 million of which were government positions.
Since that time, several trends bear mention:

-In the goods-producing sectors, three distinct cadres emerge. The
number of miners and loggers in 2008 was nearly identical to the 1959
total of 789,000, yet wild swings can be seen repeatedly in that
50-year span: 658,000 was a two-decade low in 1971, then a 50-year
high of 1.2 million was achieved only 11 years later. In
manufacturing, current numbers are 2 million lower than 1959, but
population has nearly doubled, from 177 million to over 300 million.
The final goods-producing sector, construction jobs, grew even faster
than population, from 3 million to 7 million.

-Services-sector job growth reveals some truisms and some surprises.
Government employment more than tripled in 50 years. Leisure and
hospitality quadrupled. Education and health jobs (the category is
bundled) have multiplied six-fold, some of which adds further to the
government total. The financial sector grew 330%. The biggest
surprise since 1959 among services sectors is information, which
barely doubled by 2000 and has been shrinking since, presumably led by

The American geography reflects these changes in many ways, some of
them subtle. The boom in both construction and leisure sectors, for
example, helps explain Florida. Financial services consolidated first
in a few cities (Boston, metro New York, Charlotte, Atlanta, Dallas)
then in a handful of firms in those areas; other cities, most notably
Philadelphia, declined in banking prominence. The rise in
education/health and government sectors makes Austin, San Antonio, and
North Carolina's Research triangle logical beneficiaries. As the New
York Times pointed out, Detroit's steady decline contrasts sharply
with the rise of Washtenaw county, only 45 minutes away, powered by
the University of Michigan.

Other regions are seeing decline, of course. Buffalo may have been
the country's first victim of globalization as the St. Lawrence seaway
bypassed the former grain gateway and then the steel industry also
left town: the city's population has fallen by half in less than 60
years. Now, only 75 miles away, Rochester copes with the falling
fates of information industry pioneers Kodak and Xerox, with only
partially compensatory development in education and health care:
together, Xerox and Kodak employ fewer people than the University and
only 3,000 more than Wegman's grocery stores and its headquarters
operation. Population has declined by 37% since 1930.

Timber industries are in retreat as newsprint consumption goes down
and furniture-making moves offshore; home construction, while
cyclical, is not enough to compensate even in the boom years. Since
the 1960s Maine has lost shoe-making and other manufacturing jobs, and
even though there's no shortage of trees, the lack of saw- and paper
mills contributes to a downturn in both logging and the companies that
sell capital equipment to the paper industry. Pennsylvania is the
nation's leader in hardwood lumber, meanwhile, but that business is
hurt both by Chinese furniture factories (and the accompanying new
sawmills) and the housing bust: demand for oak and maple cabinets and
floors has fallen.

With the decline of the vast integrated steel mills that used to
dominate the industrial heartland in favor of lightweight minimills,
several things are happening. First, some cities including Pittsburgh
have reclaimed the real estate formerly occupied by mills to build
biotech and other facilities. The contrast between Boston/Cambridge
and Bridgeport or Worcester with regard to brass, leather, and similar
factories is playing out again, with Pittsburgh able to use education
and health care for economic growth while cities like Allentown and
Reading still struggle.

At the same time as steel mills have changed shape and location, coal
production for electricity generation has surged. West Virginia leads
the nation in coal mining employment, but productivity per miner is an
order of magnitude higher in the western open mines of Wyoming and
Montana. West Virginia also provides a fascinating contrast with its
close, intertwined neighbor (the states' northern borders overlap for
over 100 miles). Maryland is home both to unemployed crabbers and the
most highly-educated employees in the country. They work in such
units as the Department of Energy, National Security Agency, and the
biotech research complex at Fort Dietrich, where pioneering research
was done on surgical robots, for example. West Virginia, meanwhile,
ranks third in the nation in obesity, 49th in per capita income
(Maryland passed Connecticut to become #1), and in the bottom five in
educational attainment.

Sticking with commodities, no element is as intricately involved in
the information age as copper. Whether in microprocessors,
transformers, power lines and cables, or data networks, copper is
essential for both computing and communications. Michigan's upper
peninsula used to be an active producer, but starting in the 1960s the
emergence of open-pit mines in the American west, and more important,
Chile, made deep-shaft mining uneconomical. While iron continues to
be mined about 100 miles away, the "Copper Country" was transformed by
tourism and education as the former Michigan College of Mines emerged
as the area's economic engine, employing 1,600 people with an annual
budget of roughly $185 million. Nonetheless, both Houghton and
neighboring Keweenaw counties have been losing population in the 2000s
after staying flat the previous 20 years, according to Census Bureau

Policy Implications
What generalizations can we make about the past 50 years, dominated as
they are by the dual (but mysteriously correlated) forces of
information and services? In the northeast, deindustrialization means
more forests, closed mines, decommissioned railroad tracks, and
smaller tax rolls. It also provides the necessity and the possibility
of repurposing of urban industrial real estate. The simultaneous rise
of the South and West, meanwhile, stress-test the infrastructure of
the states that are growing. The budget crises in Pennsylvania and
California that have such different origins, Atlanta's water
emergency, and Florida's real estate collapse all indicate how
internal migration is creating structural difficulties in both the
stagnant and destination states. At the state and federal levels, new
policy issues proliferate.

1) Who Rules?

In addition to reshaping Congress and thus domestic spending
priorities, the 2010 census will tell a fascinating story about
internal migration. Given that there was substantial activity in
certain information industries in the 1990s, and that manufacturing
certainly suffered its share of setbacks, are there migration patterns
that we might expect to intensify when the next measures are taken a
year from now? Between 1990 and 2000, no states shrank, but 11 states
grew by less than half the rate of U.S. population growth, which was
about 13% for that period. Note the presence of Electoral College
heavyweights New York, Ohio, and Pennsylvania on the list, possibly
portending a shift in campaign strategy for 2012 as they lose
congressional impact:

-New York
-North Dakota
-Rhode Island
-West Virginia.

It will bear watching to see if such states as Nevada (66% population
growth in the 1990s) and Arizona (40%) will exhibit any population
artifacts of their role in the mortgage crisis. California,
meanwhile, grew at scale in the '90s, essentially adding the
equivalent of the entire state of Minnesota circa 1990. As the
state's financial and physical infrastructure creak under the load,
where will people move next?

2) Who Pays?
One facet of the change in employment patterns emerges in the need for
tax revenues. Three stories illustrate the extreme difficulties
imposed by the shift to a more virtual economy. First, telecom taxes,
which formerly contributed roughly $30 billion a year to federal and
municipal budgets, will continue to fall as a result of VoIP and
cellular substitution. The city of Boston is attempting to recoup
some of this loss by taxing telephone poles as property. Confusingly,
electric poles are already taxed, but Verizon's predecessor company
got a tax exemption in 1915 for the encouragement of universal
service. In any case, today's reality would seem to require new
arrangements all around, and the discussion continues.

The second tax attempt came this summer in North Carolina, which
joined Rhode Island and several other states in attempting to tax the
transactions of small and medium businesses in the state that used
Amazon's storefront and payment engine to scale their market. Amazon
responded by cutting ties with the businesses to preserve the lack of

Third, I have not seen any estimates of the tax impact of offshoring
jobs, but if a given company's call centers or IT shop shrinks by
thousands, the savings in payroll will undoubtedly affect state and
local tax collections even as demand for unemployment compensation and
social services typically rises.

3) Who Regulates What?
A final policy question relates to the role of broadband. As sparely
populated areas shed jobs and possibly population, the Obama
administration continues to stress the role of network connectivity
without yet specifying who, how, or how expensive. Stringing copper,
coax, or fiber can get extremely expensive, particularly as population
density decreases: part of the pressure on wireline businesses comes
from the cumulative effect of so many defections. A social good whose
costs were shared broadly becomes incrementally more expensive as the
infrastructure costs get shared by fewer and fewer people.

Note that in the "revestiture" of the U.S. telecoms industry, neither
SBC (later ATT) nor Verizon (originally Nynex plus Bell Atlantic)
wanted Qwest and its vast, mostly sparsely peopled service area
extending from Washington to Minnesota and south to Arizona. Verizon
is selling off landline businesses in states such as Maine. Sprint,
meanwhile, spun off its heavily rural wireline business entirely in
2006 as Embarq. Saddled both by customer defections (10,000 a week)
and heavy debt, Embarq was subsequently bought in 2008.

Meanwhile, wireline providers have limited options as universal
service provisions remain in place: firing unprofitable customers is
rarely possible. The universal service provisions could conceivably
be augmented to mandate broadband, but rules that applied to
monopolies have yet to be overhauled to fit a competitive landscape:
in our small town, I could obtain some flavor of telephone service
from at least seven providers. How many providers does a network
require to be called competitive, and thus lightly regulated? Sprint,
AT&T, and Verizon have laid off nearly 30,000 people the past 12
months alone, yet how many regulatory entities are shrinking

The shift in emphasis to services, along with global competition, has
changed the American landscape in myriad ways:

-People can move south and west as a) air conditioning makes the
climate tolerable and b) jobs no longer connect to natural resources
and rivers or ports: compare Atlanta, Dallas, or Phoenix (defined
instead by their interstates) to their northern counterparts. In
historical terms, meanwhile, when will construction activity and
employment slow down their rapid growth?

-As agribusiness grows in scale, farmland in the northeast is
returning to (mostly unmanaged) forest. Deer populations now
constitute a serious problem in many states as habitat increases while
hunting and predation do not. The lack of timber industries allows
these forests to become recreational and ecological resources.
Reforestation and healthy watersheds raise the question: when will
fresh water turn again to become a major economic driver, particularly
in industries where virtual work is a possibility?

-Universities and hospitals have replaced factories as economic
engines in many localities, but as non-profits pay different kinds of
taxes compared to factories.

-In many ways the people who most need broadband, for distance
learning and telemedicine for example, are the least likely to have it
because of the high cost to serve remote populations. Remedying the
access paradox may not be the challenge that health care reform is,
but that does not make it a simple matter.

-As Maryland's physical proximity to but economic distance from West
Virginia shows, the world is not fully flat: local advantage can still
matter, and matter decisively. Detroit, meanwhile, shows just how
"local" local advantage can be.

-Competitive forces that reshape industries are not reflected in the
government portion of the services economy. Even seeing the true size
of state and federal government agencies is a challenge, much less
altering them. The Japanese automakers successfully challenged the
Big 3 on quality, for example, but taxpayers can't similarly defect to
a better performing government. The population of metropolitan
Washington, DC, meanwhile, grew 29% between 1990 and 2007, giving it
the final word on the reshaping of the American landscape in the
information age.

Monday, July 27, 2009

July 2009 Early Indications: Love & Work part II

Taking as our cue Freud's famous dictum that people are defined by
their love and their work, last month's letter looked at online
dating. This time around, the focus turns to work, prompted by both
the current economic situation and a few books on the topic.

What have computers and the digital revolution done to work? Answers
vary considerably. In 1992 Robert Reich (later Bill Clinton's
Secretary of Labor) devised a tripartite schema to classify the
workers of the world, seeing global work forces as already having been
divided into three groups: routine producers (e.g., call center reps
or assembly-line workers), in-person servers (waiters or nurses), and
symbolic analysts who manipulate pure information for large profits
(Wall Street quants). Digitization in the service of high leverage
made the "symbolic analysts" rich, and skewed income distribution.
Seeing the relation of rich to poor less than 20 years later, Reich
may have been onto something crucial, but his tepid solution --
training and education -- has failed to shift the terms of the debate,
partly because school systems change incredibly slowly and require
levels (and types) of investment that are for a number of reasons
politically impossible in the U.S.

A decade later, Richard Florida defined the engine of the new economy
as the "creative class," 38 million of whom comprised 30% of the
workforce. For the winners, digitization empowers flexible work that
gives great meaning:

"In this new world, it is no longer the organizations we work for,
churches, neighborhoods, or even family ties that define us. Instead,
we do this ourselves, defining our identities along the varied
dimensions of our creativity. Other aspects of our lives -- what we
consume, new forms of leisure and recreation, efforts at
community-building -- then organize themselves around this process of
identity creation." (Rise of the Creative Class, pp. 7- 8)

Surely 30% of the workforce can't work at ad agencies or Disney. No,
says Florida, "I define the core of the Creative Class to include
people in science and engineering, architecture and design, education,
arts, music and entertainment, whose economic function is to create
new ideas, new technology and/or new creative content. Around the
core, the Creative Class also includes a broader group of creative
professionals in business and finance, law, health care and related
fields." The core and the donut are linked not by geography or income
or skills but by a value set: "all members of the Creative Class --
whether they are artists or engineers, musicians or computer
scientists, writers or entrepreneurs -- share a common creative ethos
that values creativity, individuality, difference and merit. For the
members of the Creative Class, every aspect and every manifestation of
creativity -- technological, cultural and economic -- is interlinked
and inseparable." (p. 8)

Whatever its relation to life as most people know it, Florida's book
resonated. It led to a thriving consulting business helping cities
attempt to become more economically competitive. How? Not with tax
incentives for auto plants but by luring more of those 38 million
people with more tolerant attitudes, better mass transit, more
authentic espresso bars, and the other factors that separate Toronto
from Topeka or Minneapolis from Modesto.

In the intervening seven years, however, much has happened to cast
doubt on Florida's vision of the future. What exactly do those
creative people do to help the U.S. balance of trade deficit? Movies,
M&A deals, and Microsoft all contribute to exports, but not to the
degree that farm goods do, and none approaches the aerospace sector's
international impact. What happens when offshore competition
threatens large numbers of those 38 million jobs? Legal research,
programming, equity analysis, and even movie-making and distance
learning are already being produced and delivered from afar in
lower-wage settings -- what will be next?

More fundamentally, just how creative are those 38 million people?
Job titles can be deceiving: a good friend of mine was an architect at
HOK, the sports division of which has given us such modern monuments
as Camden Yards in Baltimore or ATT Park in San Francisco. What was
our young Howard Roark's creative contribution? Bathrooms for the
Hong Kong airport.

Matthew Crawford, in a new book called Shop Class as Soulcraft,
raises similar doubts. Beginning with the observation that many high
schools are dropping shop class because it fails to train people to be
symbolic analysts, Crawford challenges the reader to think deeply
about the value of work. Because it often lacks real output, modern
bureaucratic life, defined largely by office automation, can be
unfulfilling. In contrast to the carpenter whose windows can't leak,
or the farmer, who feeds people with tangible crops or livestock, the
office worker (creative or not) lacks physical boundaries to define
the real from the artificial or the possible from the impossible.

As Crawford notes, quoting Robert Jackall's Moral Mazes (now 20
years old), office memos are crafted to be unincriminating no matter
how subsequent events play out. Taking a firm stand is often seen as
career-limiting, so most eventualities remain unforeclosed; every
position is hedged. Along similar lines, after receiving a Ph.D. from
the University of Chicago, Crawford works for a think tank generating
position papers that begin not with the facts but with a position,
reasoning backward to convenient truths. It is intellectual bad faith
of the first order, and he quits. Worse yet, in his circles of
occupational hell, are jobs built on teams with their indeterminate
appropriation of credit and blame, along with the HR-driven
trust-building games that frequently pass the point of self-parody

In contrast, the author points to his work as a motorcycle mechanic.
No symbolic analyst he, Crawford confronts physical limits every day,
and pays a steep price for failure. If he drops a washer into a
crankcase, there are times when he must tear down the engine block to
retrieve it, and cannot in good conscience bill the customer for all
of the hours involved. Mistakes, stupid or otherwise, have concrete
consequences. On the positive side of the ledger, when he fixes a
broken fork, returns a dead bike to life after ten years off the road,
or hears the particular sound of a well-tuned engine, he derives great
satisfaction. He also contends that mechanical work can be more
intellectually engaging than "knowledge work," implicitly challenging
Florida's new world order.

In some measure, we are fighting a new stage of the battle joined by
Descartes, who separated thought from emotion and thereby physicality.
Craft work (fixing or building things) joins the practice of medicine,
certainly, but also full-throated singing as moments where mind and
body unite. Sport constitutes another similar realm, as does cooking,
the recent enthusiasm for which might be seen as a reassertion of the
satisfaction that can only come when head, hands, and palate unite in
a primal act -- that of feeding another person. Compare the gestalt
of today's many cooking shows to the treatment of the modern workplace
in current television programming and the contrast is obvious: Julia
Child, enshrined at the Smithsonian, is a hero while cubicle America's
cultural icon has yet to transcend Dilbert.

Shop Class as Soulcraft also makes the pragmatic point that fixing
things cannot be offshored; one can make a healthy living as an
electrician, for example, or an auto repairman. Last time I was in
for an oil change my mechanic was telling me about one manufacturer's
switch to a fiber-optic system bus -- he knows more computer
networking than I ever will. To service appliances or furnaces today
is to have studied hundreds of hours of digital control and monitoring
technology. High schools, however, generally operate under the
principle that college-bound students will have better careers than
those who work in jobs that require mere training. But what
economists call the education premium can no longer be assured today,
much less in 50 years when today's high-school graduates will almost
certainly still be working.

There's also the matter of permanence. As Crawford notes, many of
today's appliances are built to break and not be repaired. How does
today's work give people the opportunity to build something that will
last beyond their life span? For teachers, this is one of the true
joys of the profession. For most "knowledge workers," the answer is
less clear. True craftsmen raise a red flag. As Michael Ruhlman,
known more for his books on chefs and cooking, reported in a book on
wooden boats,

"I asked Gannon why wooden boats were important to him -- why had he
devoted his life to them? Ross seemed surprised by my apparent
ignorance regarding what to him was plain, and his blazing eyes burned
right through me.

'Do you want to teach your daughter [then 3 years old] that what you
do, what you care about, is disposable?' he asked. 'That you can throw
your work away? It doesn't matter?'"

Whether in passing down the family farm or painting "& Sons" on the
service van, craft work is often connected to future generations that
bureaucracy cannot sustain. This lack of long-term continuity may be
another reason why the modern office lacks heroic images in popular

A final reason for the current deep unease with the prevalent model of
work lies in the fact that we are undergoing several foundational
shifts. Global competition and offshoring are familiar, but the
aftermath of digitization changes so many aspects of life so quickly
that some sense of vertigo becomes unavoidable. As Carlota Perez
asserts in her book Technological Revolutions and Financial
Capital (p. 57), every major technology breakthrough in the western
world since 1750 progresses through two major historical phases,
installation (ending with a bubble popping) and deployment.

Between the two often lies a financial crisis: 1848 for the age of
steam, 1893 for the age of steel, and 1929 for the automobile and
associated industries. Her positioning of 2001 as the start of the
computer age's financial crisis is persuasive, meaning that if the
past is a guide, that deployment of these telecommunications and
computing technologies will bring both financial growth and structural
change. Work will undoubtedly be very different 20 years from now.

Tom Malone of MIT explores that future landscape through the lens of
its institutions. In his 2004 book The Future of Work, he lays out
various scenarios primarily concerned with the coordination and
collaborative facets of organizations. He sees the future as more
decentralized, less hierarchical, and more democratic. If it comes to
pass, Malone's vision foreshadows the demise of "The Office's" Michael
Scott and his kin. Pettiness and incompetence are eternal, however,
so it is worth pondering both what will happen to a Michael in a
Malone-ite world, and what manner of successor will emerge instead.

In the end of any analysis, work cannot be categorized with any
precision. It is both universal and specifically grounded in time,
place, and individual. It offers both rewards and challenges (some of
which may overlap), utilizes groups and solo contributors, and defines
us in multiple ways. The diversity of the perspectives mentioned here
is itself incomplete, missing, for example, the perspective of the
Japanese salaryman, the unionized autoworker, or the classic
professions of law or clergy (both of which themselves are in the
midst of deep change). We have made no mention of wages, which are
retreating in many settings. The appeal of Dan Pink's vision of Free
Agent Nation (2002), for example, has been replaced by the reality of
the less glamorous name for continuous partial employment, "temping."

As to the question, "what have computers done to work?" the answer is
probably less clear than it will be in another 25 years, when the
changes to economies, workplaces, and individual performance will
separate themselves from the end of the oil/automotive/steel age that
wound down in the late 20th century. The exciting news comes in the
realization that the future of work is not yet defined, making it
contingent on the attitudes and actions of many people, professors and
motorcycle mechanics hopefully among them.

Tuesday, June 30, 2009

Early Indications June 2009: Love, Online

Love and work are the cornerstones of our humanness.
-Sigmund Freud

How have the technological changes of the recent past affected these two facets of our existence? As economies around the globe attempt to generate "good" jobs in the face of steep declines in such traditional sectors as news and media, automotive, and even law, the nature of work is at once a policy, economic, and existential question. As for love, we have seen misplaced romantic e-mail damage the careers of public figures including CEOs and a governor. Finland may show us the wave of the future: in 2006 the prime minister met a woman through an online dating service then broke up with her via SMS a few months later, stating economically, Että se ("that's it").

Let's take the love question first; we'll tackle work next month. I was struck by the number of anecdotal cases among my acquaintances involving structured dating services such as or eHarmony. A little research shows how big these services have become, not counting the vast amount of flirting within the big social networks: paid online dating sites were essentially a billion dollar industry in 2008, according to Forrester, putting it ahead of pornography and making the industry slightly more than half as big as digital music and gaming.

Countries around the world are getting involved: in Japan, dating sites must register with the police, and over 1600 such companies did so. Between January and September 2008, eHarmony and Match combined to spend $140,000,000 on advertising. According to New Media Age, UK traffic to online dating sites grew 13% between September 2008 and February 2009: total visitors number about 5m, reaching 13% of the total UK online population. Harris Interactive estimates an average of 236 eHarmony users get married - every day., part of InterActiveCorp, was recognized in 2004 as the largest site in the world and reported 1.35 million paying members as of May 2008.

Online matchmaking has many variations. One can search for potential spouses, for religiously or culturally similar partners, for friends, for same-sex prospects, for uncommitted physicality, or, at Toronto's Ashley Madison, be guaranteed an affair - or your money back. The various market segments each have multiple providers, varying by geography, matching method, and revenue model. Based only on online comments from users rather than any personal experience, claims of differentiation between different sites' matching accuracy and inventory may be inflated: many people use multiple sites and find the same people matching their profile.

Late last month, the American Press Institute convened a meeting of newspaper executives to discuss the state of their industry. They placed substantial blame on Google for being the "atom bomb" to the news industry. Left unlisted were the papers' many losses in the bigger content picture. The report did not acknowledge how many readers are defecting from the bundle model in favor of specialty providers: ESPN et al for sports, the likes of Edmunds for car-buying, Yahoo Finance and many other money sites for stock quotes, to house-hunt, eBay for used cars and household goods, Craigslist for apartments, etc. Personal ads are clearly a part of this erosion: Match, eHarmony, and the rest did not cannibalize all of that $957 million from newspapers, but clearly papers have lost some of their mojo in that department.

As for magazines, college alumni periodicals' listings appear to be shrinking. The New York Review of Books has long featured personal ads that were almost a parody of themselves. I found that, through the magic of the Internet, people can still browse for such appealing specimens as:

LOS ANGELES: bright, playful professional/academic MWM, 50s, tall, fit, good-looking, warmhearted, engaging, open, and present. In emotionally untenable marriage. Seeking woman, 40–58, with good heart, curious mind, sensual, open to exploring possibilities with like-minded good soul.

Some independent newspapers maintain a strong singles presence, as witness the Chicago Reader or The Onion. Mainstream papers, meanwhile, take a variety of approaches. (the Globe's online operation) franchises singles from Yahoo. The LA Times points readers to eHarmony. Many papers, including the New York Times and Dallas Morning News, have no personal ads.

Just within my casual contacts, two adults have married based on their use of the services. One is an innkeeper in a small coastal town, while the other is a medical professional in a relatively remote market. In both bases, geography-based dating is problematic both in terms of sparse "inventory" and in terms of privacy given these individuals' relatively public day jobs. As a divorced charter boat captain told me of his use of Match, "It's great for people like me: in small towns like these, everyone knows everyone else's business. And while you can date the tourists, it's a bit tough starting a new relationship every week when the rentals turn over." The decoupling of physical location from the search process is a very big deal for market "thickness," not to mention the overall sense of romance and adventure in the process.

Secondly, the use of algorithmic matching tools is enhancing the matching process: eHarmony's "scientific" survey instrument includes 400 questions, far more than I ever answered on any unsuccessful first date (or the successful one, for that matter). As we will see, however, the comprehensiveness of the surveys has many implications.

Not surprisingly, the online dating phenomenon has generated sometimes hilarious commentary in the form of vast numbers of blog entries and a few books. Such titles as MatchDotBomb: A Midlife Journey through Internet Dating, Millions of Women Are Waiting to Meet You: A Memoir, and numerous how-to volumes (including a Dummies guide) testify to the pervasiveness of this cultural phenomenon.

The unintended consequences are fascinating to watch.

-Is it ethical for pay sites to count non-paying (former) participants in a match panel?

-How sustainable are the various business and operational models? Might one technology, celebrity endorsement, or other factor prove decisive in a particular market?

-What happens to my profile after I quit the service, either because it worked or because it failed? What rights do I have to my profile on either free or paid services a) after a month, b) after a year, or c) after the company goes bankrupt or gets acquired?

-What are the de facto (when people meet in person) and de jure (in court) standards for truthfulness? eHarmony, for example, insists that applicants be single: legally separated individuals are excluded, and could be banned if they lie to get on. We have a family friend who's a tall woman using some Internet dating resources, and her stories of men's various versions of their height and weight are funny and troubling. "Truth in advertising" has many nuances in this domain.

-What exactly are people paying for? What are the guarantees, warranties, or lack thereof?

-How can and will various systems be gamed? Some services have been accused, without proof, of employing "ringers" (professional first-daters) to exaggerate the quality of available singles.

-What will my profile be used for? Cross-selling opportunities, for example, are numerous and more than a little spooky.

-While the nightmare blind date has become a cultural stereotype, the prospect of meeting truly dangerous people online is more than a little scary, as the Boston Craigslist crimes suggest. It's also possible for bad first encounters to facilitate stalking. I have a colleague whose "thanks for coffee" and the implied "have a nice life" after a meeting drove the candidate to look her up using available search methods. He later turned up on her doorstep unannounced.

The role of such civic institutions as churches, service clubs, and bowling leagues in the wake of suburbanization, television, and more women in the workplace has changed slowly but significantly over the past 50 years. The matchmaking process has changed as well, and the state of online dating businesses will bear watching. In addition, the place of Facebook in 20-somethings' lives is undoubtedly generating its own set of changes to courtship.

As my former boss once said, "digits never die." The prospect of a lengthy profile, and potentially e-mail communications within a dating site, being brought into a congressional confirmation hearing or other process 10 or 20 years from now might give some of those millions of users a pause before they declare yet another preference or "fact" about their life.

Monday, June 01, 2009

May 2009 Early Indications: Clouded Over

The proliferation of so-called cloud computing platforms has been rapid. Because there is so much material available that defines the phenomenon, we'll move here to an examination of some of the unexpected consequences and complicated implications of moving some or all of a computing environment to offsite, third-party environments.

To get the problematic and inevitable definitional question out of the way, here is one from Information Week's John Foley: "Cloud computing is on-demand access to virtualized IT resources that are housed outside of your own data center, shared by others, simple to use, paid for via subscription, and accessed over the Web."

There are of course other contending definitions, but Foley's is mercifully brief. Even so, it begs the questions of private clouds, how small a cloud can be before it starts being something else, and how individual uses of clouds (I don't own a data center but hit on most of the other conditions) vary from and overlap corporate ones. It does get us started in more or less the right direction.

First, here are some resources to get up to speed:

The Economist special report from last October

Accenture Cloud homepage

Amazon Elastic Compute Cloud

Google App Engine

HP Cloud Assure

IBM Cloud Computing

Microsoft Azure platform

Rather than handicap the vendors, or the vendors' definitions, I'd like to focus a bit farther out. In a series of conversations with our corporate and university advisors, a number of questions have surfaced. In particular, I'm building on remarks by Mssrs. Smith and Parkinson at our member meeting earlier this month.

1) What is a vendor's profit path? What can be differentiated and thus generate margins? Compared to the conventional model of data centers, which is often measured in $10,000 or $100,000 units, cloud computing usage at Amazon is measured in dimes.

2) How will incumbents respond? If I have an established business selling hardware as capital expenditure, and a competing model shifts MIPS to an operating-expense model, presumably I don't stand still. Oracle's plans for Sun will be relevant here.

3) How does cloud lock-in vary from existing software (a la classical Microsoft) or hardware (the vintage IBM model) variants?

4) As with so much of the world's infrastructure, what is the incentive to invest in "pipes" when the value-add lies elsewhere, or nowhere?

5) If for legal or other reasons I need performance, security, and/or reliability guarantees, how do I get them if I cannot see or physically access my assets?

6) There are no free lunches. Every one of the Web's elite destinations has suffered from major outages at some point. Just weeks ago, Google suffered a technical breakdown about which the company released few particulars, but it managed to slow down service to millions (or more) of users for several hours on May 14. Gmail also failed at scale in February. In light of that history, what does a fault-tolerant cloud environment look like, require, and cost? (For an amazing graphic of the "Great GoogleLapse," see here)

7) Can there be "one throat to choke" in a virtual environment? Just as outsourcers are arbitraging labor rates by shifting contracted work to other shores, so too will cloud vendors assemble services from multiple entities to create bundled offerings. What will be the unexpected consequences for customers?

8) How does optimization work in a cloud? The vendor may be managing to power consumption, say, while customer A wants stable (not necessarily fast, but predictable) transaction times for a shopping cart scenario. Customer B needs fast compute capability despite bag and frequent reads and writes to disk. How can all three parties go home happy at the end of the day?

9) How can virtual, hybrid environments be tested before major real-world events: a quarterly close, a consumer promotion, a currency meltdown? While there will be some greenfield successes, a big question relates to how well clouds can integrate with existing data centers and other assets. (What constitutes unit testing in a cloud?)

10) What can I as a customer ask for by way of customization? Who can and will provide it, and at what costs in money and performance? The price points reflect commodity economics, but sooner or later most of us stumble upon needs that surpass plain vanilla.

11) Long ago, factory layouts (and locations) changed as power shifted from waterwheels that drove a central shaft around which looms were arranged, to individual electric motors for each machine. White collar offices after the rise of the PC no longer feature typing pools. What will be the organizational innovations that cloud computing makes possible? Focusing on power savings in the data center is a useful first step, but the technology will have many other implications for the ways people come together to achieve goals.

12) The PC architecture flourished in part because of its interoperability: I could choose a big Maxtor hard drive or a faster Seagate, a Dell LCD or Sony CRT display, and my hardware maker could buy the cheapest CD drives, RAM, and power cords on a given day. USB made the platform more flexible yet. Once I choose a cloud provider, how must I choose my ISP, my system management vendor, my billing system? In short, what are the dependencies introduced by a cloud instance?

13) Companies don't switch casually from CA Unicenter to BMC Patrol, HP Openview, or IBM Tivoli, much less a promising startup, because the complexity issues are enormous. Will my Tivoli/Openview/whatever console be able to instrument both my owned hardware and my virtual assets, or do I rely only on the cloud vendor -- who will have good reasons for not exposing too much operational information? The various answers here will have implications for lock-in, for innovation, for risk management.

14) Cloud computing is a coherent-sounding phrase, but computing in turn has many facets. Think about the different time scales relating to

-network latency
-the laws of physics regarding hard drive access
-the laws of physics regarding hard drive failure
-various data structures (think of MapReduce versus SQL)
-load-balancing, failover, and other necessary housekeeping
-core vs. edge workload allocation.

At the end of the day, orchestrating all of those sets of events, each with their own timescapes, in a virtual world is a really, really tough technical and managerial problem. Getting the systems to work doesn't even scratch the questions of profitability, liability, audit and related requirements, etc.

The question is not, will cloud computing happen, but rather, how will this tendency unfold, and how will organizations, regulators, and other actors respond? Until the rhetoric and more important the base of experience moves beyond the current state of pilots and vaporware, the range of potential outcomes is too vast to bet on with any serious money.

Monday, April 27, 2009

Early Indications April 2009: Reexamining Offshoring

When U.S. firms replace onshore technical and other resources with lower-cost labor in offshore markets, the logic is typically financial. Five years after some of the biggest such decisions, however, it has become clear that the calculations were incomplete. As some jobs repatriate (albeit on a small scale), we can suggest some additional decision criteria for future justifications.

Let's start with a generic decision to shift 3,000 applications programmers from onshore to offshore in 2004. The calculation assumes a 10% cost of capital, a 34% tax rate, and 2% savings per year as salaries both in the U.S. and abroad grow at similar rates. The base case presumes a net present value of $400,000 savings per job, times 3,000 workers, for a $1.2 billion projected cash saving. Given the realities of activist shareholders and relentless cost-cutting, it would be difficult, and perhaps an invitation to a shareholder lawsuit, to decline those kinds of cost savings.

As the past five years have unfolded, however, some incompleteness in the analysis has emerged. Eight additional factors would be worth addressing in future considerations.

1) Inflation
Compensation growth in India in the past five years significantly outpaced that of the U.S. to the point where Indian wage inflation ran in the double digits for some of those five years. That compares to flat wage growth (but not benefit costs) in the U.S.

2) Employee loyalty
The years with high wage inflation coincided with high turnover at some offshore firms. The resulting instability contributed to lower performance gains than some onshore clients were expecting.

3) Coordination costs
Another factor widely underappreciated in many cost projections was the increase in coordination costs. Highly compensated, and extremely busy, financial services experts at various firms, for example, have told me that they overestimated offshoring's value. In particular, they and their teams spent far more time generating and refining requirements documents for a team of programmers on the other side of the world as compared to the in-house resources across the hall who knew the baseline terminology and assumptions of the firm and industry in question. Producing code and generating business value through technology are not the same thing. Put another way, Brooks' law applies to offshore resources, albeit in new ways.

4) Technology changes in enterprise applications
The rise of software as a service, virtualization, and cloud computing is challenging old models of application development and deployment, as Oracle's play for Sun Microsystems would imply. In addition, scripting-based programming practices have the potential to transform software still further. 10-year payback scenarios with a traditional computing model held constant are likely to prove problematic as the late years of the model roll around.

5) Public perception
The loss of public goodwill in the performance of offshore call centers (less so in programming) has been unexpected. In 2009 alone, United Airlines, AT&T, Sallie Mae, and Delta Air Lines have pulled back from offshore call-center contracts. As Delta's CEO stated, "The customer acceptance of call centers in foreign countries is low. Our customers are not shy about letting us have that feedback." Dell experienced a broad wave of backlash in 2007; such companies as Royal Bank of Scotland's Natwest unit go so far as to advertise that they do not offshore customer care. Even in programming, offshore resources are prohibited for certain public-sector contracts.

6) Currency dynamics
Both the Indian rupee and the U.S. dollar have undergone significant currency fluctuations, dwarfing that 2% cost savings assumption. In one year alone, a dollar went from buying 39 rupees to crossing the 50 barrier, a swing of about 30% (in this instance, in the U.S. firm's favor).

7) Corruption
The cost-saving calculations implicitly assumed an apples-to-apples comparison of contract law, financial accountability, and other facets of firm governance. But when Satyam, one of India's leading offshore firms, disclosed that its founder and CEO had orchestrated a billion-dollar accounting fraud, attention turned to the differences between Indian and U.S. models of corporate governance. Auditors from PricewaterhouseCoopers, who have been suspended from the firm and jailed by authorities, earned "exorbitant audit fees" and are alleged to have falsified key discrepancies between sales figures and bank deposits.

8) Risk
The 2004 NPV model apparently priced risk at zero, a flawed assumption when India is experiencing political tensions with its nuclear neighbor Pakistan, itself a potential "failed state" in the words of the U.S. Joint Forces Command. The Mumbai attacks of 2006 and 2008 provide further evidence that the country is facing a significant threat, in this case from non-state actors. The story continues to unfold, up to the present day. Various insurgent factions are staging attacks connected to India's monthlong election; 17 deaths were reported on the first day of polling alone.

Even given these additional factors beyond the pro forma case, the economics must continue to be compelling as IBM recently continued its practice of laying off U.S. workers (somewhere between 5,000 and 10,000 so far year to date) and increasing staff in developing markets: employment in Brazil, Russia, India, and China totaled 113,000 in 2008 with the majority of those jobs in India, where headcount doubled (to 50,000) between 2005 and 2007.

Going forward, it will bear watching what happens next. India is becoming expensive, so its firms are in turn offshoring to the Philippines, Vietnam, and elsewhere. U.S. firms are revisiting the cost-benefit equation of onsite programmers as wages have declined, risk has increased, and user dissatisfaction has mounted. China, as in so many matters, bears watching. India's elections will by definition have unexpected results.

At the least, the past five years of offshoring have proven that the logic of the business case depends as much on what one leaves out as on the numbers assigned. The process of globalization will continue to amaze, frustrate, and surprise, despite the best predictions of smart people. Unexpected consequences, for both good and ill, will continue to challenge firms -- and individuals -- on all sides of the equation.