Economic Disruption: We’ve Seen This Before

Live-blogging from the IBM Watson University Symposium at Harvard Business School and MIT Sloan School of Management. Additional coverage is on the Smarter Planet Blog. .

Closing Remarks By Martin Fleming, IBM Chief Economist

Martin Fleming, Chief Economist, IBMIf we look at the global economy over the last 250 years, we’ve seen five waves of technology change:

  • Industrial Revolution
  • Age of Steam and Railways
  • Age of Steel, Electricity and Heavy Engineering
  • Age of Oil, Automobiles and Mass Production
  • Age of Information and Telecommunication
The Next Big ThingThis is one in a series of posts that explore people and technologies that are enabling small companies to innovate. The series is underwritten by IBM Midsize Business, but the content is entirely my own.

We’ve seen similar characteristics of each change. One is that we’ve gone through a stage where technology is applied to existing processes. Early factories were still vertically structured near bodies of water. It wasn’t for years that people restructured factories around the new machines.

I can remember when people were paid in envelopes of cash. Then we had payroll checks and now we manage our benefits via the Web. Technology enables us to change the way we do things.

Another example is that we have changed the way we consume media because of the iPhone and iPad, and this has altered fundamentally the media industry.

A third example is Watson in a health care context. This can allude to the democratization of health care, where the practice of health care is altered because of the technology. The whole concept of evidence-based medicine is an important new development.

Each of these five waves has been interceded by a crash in the financial markets. We’re beginning to understand this phenomenon.

When the global economy was declining in 2008, the decline in world trade and industrial production was actually steeper than in the 1930s. The decline didn’t last nearly as long (I would assert because of more assertive economic policy) but it was very steep.

The increase of public and private debt and the concentration of wealth in the financial sector also correlate to historical trends.

As we came to the end of the 1990s, we had opportunities that were typically high in ROI that were increasingly difficult to find. The financial sector had to look to new ways to deliver high returns to their investors. So you had the creative of things like sub-prime mortgages and derivatives. We also saw a reduction in bank regulation because the memories of the previous episode had faded. We also had financial structures in place, like pension funds, that were unsustainable in the new environment.

So we’re seeing new models appear. We’re creating ways to use the new technologies as the old technologies wind down.

When we look at fixed-investment spending, starting in the late 19th century the wave of investment grew rapidly. Then it slowed in the early 20th century and nearly stopped in the 1930s. But then there was a burst in the 1940 and 1950s, another slowing, and now we’re in a period where there is almost no growth. Labor hours and labor productivity showed similar shifts. Real interest rates also went through a similar increase and decline.

I don’t want to say this is a cycle. There is a very stochastic set of events that are occurring, but there are some similarities.

What does this have to do with Watson? The economy is beginning to open up opportunities for innovation. Demand for skills is shifting. It will likely be the case that in some job categories there are insufficient numbers of workers, which will cause assets to be created that reduce skill requirements.

This set of changes are in line with what we’ve seen with Watson capabilities.

David Ferrucci on Building the World’s Smartest Computer

Live blogging Dr. David Ferrucci’s address to the IBM Watson University Symposium at Harvard Business School and MIT Sloan School of Management. Ferrucci was director of the IBM Watson project.

Additional coverage is on the Smarter Planet Blog. .

Ferrucci tells a story about his daughter’s quote: “Interesting things are boring.” Because that was her frame of reference. Interesting things involve a lot of complexity that makes them boring (to people who aren’t explicitly interested in them).

We’re trying to move from the age of moving bits to the age of understanding their meaning. Meaning is ultimately subjective and we are the subjects. We see a calculator with dice on top and we infer it has something to do with calculating odds. We bring a huge wealth of background knowledge to interpreting what an image means.

The Next Big ThingThis is one in a series of posts that explore people and technologies that are enabling small companies to innovate. The series is underwritten by IBM Midsize Business, but the content is entirely my own.

Don’t expect AI systems to originate meaning, but expect them to infer meaning. We have to get the AI systems to make the right assignments of meaning but we can’t expect them to originate meaning.

You can look at context from a chronology or magnitude or causation perspective. If you want computers to detect human meanings, it takes advanced technology.

Jeopardy was an interesting challenge. The end goal was to get smarter about processing a language that humans use to detect intended meanings. Watson and Jeopardy were milestones in that direction.

You couldn’t play Jeopardy by having thousands of FAQs and linking to their answers. Example: “The first person mentioned by name in  ‘The Man in the Iron Mask’ is this hero of a previous book by the same author.” The answer is D’Artagnan. If you’re going to buzz in on that question, you’d better have an accurate probability that your answer is correct. You have to look at all the evidence supporting that answer and analyze it so that that probability gets better and better.

Example: “This actor, Audrey’s husband from 1954 to 1968, directed her as Rima the bird girl in ‘Green Mansions.’” You need to parse what “direct” means: to guide, to lead, to direct? What is “Green Mansions?” Watson would produce multiple syntactic assignments because there was no guarantee of approaching that problem correctly. You need to parse all of these assignments in parallel to see which are most likely.

In a random sample of 20,000 past Jeopardy questions we found 2,500 distinct types. The most frequent occurred less than 3% of the time. They didn’t help us map to answer, but it did help us identify different ways of understanding how questions are parsed.

Plausible inference varies by context. Example: In the category of Lincoln Blogs: “Treausre secy Chase just submitted this for the third time. Guess what, pal? This time I’m accepting it.” Answer is “resignation.” Sixth-grade class came up with a different answer: “Friend request.”

“Vessels sink.” “Sink” can refer to submerged but you can also sink a cue ball.

We calculated the winning player on Jeopardy gets about 50% of the answers. Those that they got a chance to answer they answered correctly about 85% of the time. Ken Jennings was able to answer 62% of the board, which was phenomenal.

Fundamental bets we made:

  • Large hand-crafted models are too limited.
  • Intelligence from the capability oif many: We had to come up with many hypotheses and algorithms to figure out how to attack the problem. We had to combine those to balance their application.
  • Massive parallelism was a key enabler: We had to pusue many competing independent hypotheses over large amounts of data.

The notion of intelligence from the capability of many enabled us to get a lot of different components to interoperate with each other. We had to build many different ways to combine them.

Watson Architecture

Watson takes many possible alternatives or candidate answers. For each of those hypotheses it gathers a large amount of evidence. So it may take 100 interpretations and gather 100 facts for each, or 10,000 factors. We then run them through filters to calculate a probability of each answer, and if it’s above a certain threshold, like 50%, then we buzz in.

Then we also have to balance in the competition. If Watson is way ahead, it’s less likely to buzz in on a lower probability. If it’s way behind, it gets more aggressive.

The goal of all researchers was can they drive the end-to-end system performance? We changed the cultural incentives. We all go into one room and everyone focused on natural language processing, information retrieval, knowledge representation and more. We produced more than 8,000 documents.

Some early answers:

Decades before Lincoln, Daniel Webster spoke of government “made for”, “made by” & answerable to them.” Watson: “No One.”

“Give a Brit a tinnkle when you get into town and you’ve done this.” Watson: “Urinate”

This system is complex in ways that no one imagined. We did end-to-end integration almost every two weeks, then drove into data to determine which factors were contributing to right and wrong answers.

We got to 87% to 90% precision, which was good enough to compete. We went into the final game with Ken Jennings. Our analytics told us we had about a 74% chance of winning.

It took about two hours to answer one question on a single 2.6Ghz CPU. We used about 15TB of RAM and 2,000 parallel cores.

I fought hard to get an answer panel on TV so people could see Watson computing possibilities. This got people thinking about what was going on behind the scenes.

An answer doesn’t really matter if you can’t back it up with evidence that a human can understand. When we apply this to health care or finance, you need to provide the evidence of why you think this is an important answer, but you have to do it in a way that people can understand. The computer isn’t giving answers but providing evidence.

Applications in health care: You’re finding hypotheses that weren’t clearly evident and gathering information that supports evidence. You can imaging information comine from patient data, symptoms, office visits, and background databases, computing these profiles, providing these profiles to the medical teams and helping them to see how the evidence supports the hypothesis.

What do people intend when they use words: “Geeshe, she was only 10 when she took ohome an Oscar in 1974. She’s 40 now.” Watson’s confidence was low. It knew that Tatum O’Neal won an Oscar at age 10, but it doesn’t understand “take home.” So we asked people to answer that question so that Watson could understand what the term meant.

It’s not enough to assign semantics to a sentence. You want to interact with people who can help you understand meaning.

Question: How long until Watson can program itself?

As Watson is training on prior questions and answers, it’s balancing those inputs and learning how to weight the possibilities. Different problems require different balancing. I can’t give you a short answer because what do you mean by programming yourself? Short answer is I don’t know.

Question: It sounds like you ran out of gas at some point in comparing Watson’s performance to humans. Is there some limit?

It does slow down. At some point, there are things that are incredibly hard for a computer to do. Can we push that line upward? I think we can, but for Jeopardy there was only so far we needed to push it. The game is also evolving. We learned that data prior to 2003 was easier for the computer to understand. After we learned that, we focused only on data post-2003.

Question: What areas are this technology more appropriate for?

We’ve only focused on health care so far. It’s harder for the machine to understand different domains. In health care, you don’t have as much training data and the input is not just a small graph. It’s a huge graph of different factors and relationships. In health care we had to tackle the multi-dimensional input problem. Because of the limited ontology in that domain, it’s somewhat easier to tackle health care than some other areas.


Downed trees in Framingham

Downed trees in Framingham (Photo via Framingham Patch)

Today is my fifth day without Internet service. My connection dropped on Sunday during Tropical Storm Irene, when my cable provider, RCN, lost power. Five days later, my local RCN office remains one of the 3% of utility customers statewide who have yet to get power back.

I’ve accepted this inconvenience with quiet resignation. On the misery scale, my plight is about the same as a mild headache, or running out of milk. Being offline for so long has made me more keenly aware of how central the Internet has become to my work and life, however.

I live online. My work setup includes four monitors powered by two PCs, each running voice recognition software. I administer four different blogs and Websites on a regular basis and stay tethered to my clients and friends through e-mail, Facebook, Twitter and instant messaging. My e-mail, calendar and many of my documents live in the cloud. Even my phone calls often involve screen shares and file transfers. Without the Internet, I’m almost unable to work.

With no travel planned this week, I’ve had to find places with free or cheap Wi-Fi that would let me take up residence for hours. This turned out to be more difficult than I expected. The closest wired restaurants – a Panera Bread, McDonald’s and a coffee shop – have all been without either power or Internet all week. The three Starbucks in the area resemble press rooms: their tables are filled with other afflicted locals hammering away on laptops. Linger too long and you get dirty looks.

A client in the area kindly offered to give me a desk in its offices. Unfortunately, my laptop wasn’t set up to authenticate to its network, so I ended up piggybacking on the cafeteria Wi-Fi for an afternoon. I might as well just have camped out in the cafeteria.

That’s when I hit upon the local public library. Blessed with a fast, free and taxpayer-funded Wi-Fi connection, extended hours and a staff that’s happy to have you there, It’s become my temporary lifeline during this week of digital deprivation. The Framingham (MA) Public Library has private, closed study carrels where I can make phone calls and dictate to my laptop without bothering others. The librarians have even looked the other way when I bring the forbidden food and beverages into my closet.

Offline Challenges

Being disconnected has had some surprising challenges. I never figured out how to set up an FTP connection to my home file server, so transferring files from my laptop has been a tedious process of saving to the local disk and transferring manually or moving files around on flash drives. Version control is nonexistent.

My personal life is also been affected. My wife and I share each other’s calendars and schedule appointments like pediatrician visits through invitations. We’ve had a couple of miscommunications this week because one of the other wasn’t connected. We routinely chat via instant messaging during the day, and with Dana off-line, cell phone texting has been a poor substitute for AOL Instant Messenger. Printing documents from the Web has meant transferring files from the laptop to a desktop connect to the printer.

Lack of TV service has left me unable to watch my beloved Red Sox during a series with the Yankees this week. AM reception in my area is poor, so I usually listen to the game over a live audio stream. Not this week.. Some of our favorite TV programs are showing their final episodes of the season this week, but we’ll miss them because TiVo has nothing to record and Web video is unavailable.

Like I said, mild annoyances. They’ve dramatized to me, though, how vital networks have become to my way of life. Fifteen years ago I got along just fine without any of these tools. Today they’re part of nearly everything I do.

Earth Knowledge is a Great Example of Content Curation

Earth Knowledge takes curation to a new level. The site was conceived as a way to promote a concept called “Reliable Prosperity,” or decisions that contribute both economic and ecological value. Founders Julia and Frank D’Agnese enlisted dozens of content partners like Alternative Energy News, BBC Earth Explorers, The Christian Science Monitor, European Environment Agency, the US Forest Service and the US Geological Survey to contribute content, which is summarized on “knowledge portals” and linked back to the original source.

One of the really innovation features of Earth Knowledge is a Google Maps mashup that delivers selected content contributed by partners in a striking visual style. The view below shows the location of natural and glacial aquifers in the U.S., and that’s only one of many options for seeing where natural resources exist and how they’re changing. There is also an assortment of narrated audio and video “tours” that use Google Maps to show things like the the geography and natural attractions of the Great Lakes.

Earth Knowledge is an example of nearly pure curation. The site operators don’t create any original content. They provide value in the creative ways in which they organize and visualize content created by others.

Earth Knowledge curation - aquifers

Why I’ve Been on Hiatus

If you’re wondering what’s become of me lately (OK, humor me that you actually noticed!), it’s because of these two little darlin’s – Lillian Emma and Blair Isabelle – who were born on April 1 at 9:02 and 9:07 p.m., respectively. They came into our lives following a difficult pregnancy and labor for Dana, who declined pain medication until her 17th hour of hard labor. All came out well, though. The twins are beautiful and healthy. At five pounds each, they could stand to put on a little weight, but with my genetic code, that shouldn’t be a problem.

And yes, the hospital really did put masking tape labels on their little knit caps to tell them apart.

Here’s an assortment of images from the night of their birth and the couple of days thereafter.

ViewMore FromTagsCommentsShareSendFavoriteTwitterFacebook

Lillian Emma (l.) and Blair Isabelle Gillin

How B2B and B2C Marketing Are Different

My fourth book, Social Marketing to the Business CustomerSocial Marketing to the Business Customer came out this week. While the purpose of this post is ultimately to convince you to buy it, I hope to also impart some insight I gained from immersing myself in business-to-business social marketing for six months.

Co-author Eric Schwartzman and I wrote the book because we felt that B2B marketers were getting inadequate advice about how to apply social media constructs to their work. We’ve attended scores of conferences over the last few years and heard lots of wonderful stories about how to use everything from blogs to video games people play using servies like CSGO BOOSTING, to even sell blue jeans, potato chips and fine wine. Invariably, someone stands up and asks, “What does this mean to me as a B2B marketer?”

The response is usually something like, “Well, you can do this, too.” I used to take that answer at face value, but the more I thought about the unique characteristics of B2B buying decisions, the more it struck me as dodge. The fact is that much of what works in consumer markets would fail in B2B interactions. There are plenty of opportunities to apply social media tactics, but the context is different.

Download a sample chapter

As Eric and I began to dig into this topic, we put some thought into how B2B and B2C markets differ. We came up with six major areas of divergence, and we were surprised to realize how really different these two flavors of marketing are. Here are the six points we arrived at. I’m sure this list is not comprehensive, so leave a comment with your impressions.

1. Value-driven decision-making. Probably the most important distinction between business and consumer marketing is that nearly every buying decision a business makes is driven by the need to solve a problem, pursue an opportunity or make or company more efficient. There is no room for sex appeal, status, feeling good, tastes great or less filling. A lot of great consumer marketing campaigns sell at the gut level, but B2B buyers base their decisions upon facts and calculated value. If you don’t deliver that, you don’t get considered.

2. Group consensus. Most businesses are inherently conservative, and decision-makers seek validation from many sources, including analysts and their peers. Part of this is simple risk avoidance, but an equally important factor is that decisions made by a group are more likely to be supported by all of the members. The bigger the purchase, the more people are usually involved. Research by marketing Sherpa and TechWeb found that 41% of technology buying decisions involved 15 or more people in the process. These people typically come from many different areas of the organization, and each has different information needs.

3. “Bet the business” decisions. When Federal Express chooses a vendor of hybrid engines for 1,500 trucks or Ford installs a fleet of welding machines on its assembly lines, the decision has the potential to affect the company’s bottom line and its stock price. Even seemingly small decisions, like the choice of an e-mail marketing vendor, can have far-reaching implications if the supplier can’t deliver. Consumers almost never face issues of this magnitude.

4. Long-term relationships. Business executives buy companies as much as they do products. Most prefer to work with a small number of favored vendors who get a large share of their budget in exchange for high-quality service and “one throat to choke” accountability. Consumers make few choices that involve persistent relationships.

5. Knowledgeable buyers. B2B buyers don’t hesitate to bring experts into the decision-making process. These people may have years of in-depth technical experience, certifications and degrees. They want to talk to the people who build the products they are considering, ask detailed questions and gain confidence that the company is a worthy long-term partner. In contrast, consumers may study up for a bit before buying a car or refrigerator, but they rarely bring people with Ph.D.’s into the process.

6. Intense need for information. A B2B decision usually requires information from a lot of sources about a lot of topics. The CFO, head of manufacturing and CIO all have different questions, and all need to be satisfied. The business buyer’s appetite for information also doesn’t end with the sale (see item 4). Users of call routing or process management systems, for example, may spend days or weeks in continuing education classes or at conferences to keep up with new developments. There is virtually no parallel for this in consumer markets.

For these and other reasons it’s shortsighted to tell a B2B marketer to apply the tactics used to sell blue jeans to the task of selling aircraft engines or sales force automation software. The same tools can be applied – and we devote 250 pages to explaining how – but the tools that B2B marketers differ in some pretty basic ways from those liked by their B2C counterparts. We found some wonderful case studies, lots of innovative people and even some very clever campaigns.

So here’s the promotional message: Buy it! Read it! Post your review on Amazon or tell us what you think here or on our Facebook page. If you’re a B2B marketer, this book is for you. Let us know if we hit the mark.

Remembering Tom

Tom MonahanOf the thousands of people who pass through our lives, few stand out as true originals. The world lost one last week. Tom Monahan died of ALS at the age of 58. He was way too young.

Tom was a treasure. He was both a lone wolf who thumbed his nose at authority and a team player who loved his friends and colleagues and who was loved by them. He was one of a kind.

I met Tom in 1982 when I joined Computerworld as a staff writer. He was a designer, and he made an immediate impression. Tom’s weapon was the one-liner. His timing was brilliant and his wit often withering. His favorite venue for a well-timed wisecrack was a full room, usually during a pause in an executive presentation. Tom could bring down the house, not just because his timing was so good, but because people couldn’t believe he had just said what he said. The more pompous or self-important the speaker, the more devastating Tom was. I was a target many times, as was IDG Chairman Pat McGovern. He made no exceptions for authority or people who could get him fired. No one else but Tom could have gotten away with it. He was just such a great guy.

Tom was an unconventional man with conventional roots. The product of a large Irish Catholic family from Boston, he eschewed religion and demonstrated little interest in relationships during much of the time I knew him. Late in his life, though, he met and married Mary, an incredible woman who nursed him through a terrible disease. Mary softened some of Tom’s rough edges, and he clearly loved her deeply.

Tom was one of the smartest people I knew, yet he never paraded his intellect. It would come out in subtle and unexpected ways: a soliloquy on the history of printing, a verbal essay on the travails of the Pilgrims or a lesson on the finer points of nautical navigation. He once told me he read three or four books a week. I don’t doubt that.

Career success, at least as many people define it, didn’t interest him very much. He worked at IDG for more than 25 years, eventually becoming online director of the website. He had the talent and the smarts to jump to a bigger magazine, join an agency or just start his own business, but visibility and financial rewards seemed to motivate him less than familiarity and the chance to enjoy his many outside interests.

Tom Monahan BookcaseWhen IDG laid him off six years ago, he was devastated, but he quickly pulled himself up and set out on a new career – as a furniture designer. You can see some of his work here. The ALS struck just as Tom was completing a two-year intensive program of study at a furniture-making school. What cruel timing for a man whose hands were his most valued tools.

Tom was an artist. He wasn’t much of an illustrator, but he had a gift for design, especially color and typography. I’ve worked with many designers over the years, but never one who took such an interest in the subject matter of his work. Tom could think like an editor, and that’s a rare trait in his profession. Editors loved working with him.

He played guitar and his band, the Texas Instruments, belted out pretty decent bluegrass rockabilly. He was a photographer and a lover of designs in nature. When he started making furniture, he made beautiful furniture. When he bought a Jeep Grand Cherokee, he had to factory-order the vehicle to get the exact color he wanted. Beauty excited him and his hands had a gift for creation.

Tom was a team-builder and a team player. Although he never liked authority very much, he loved the people around him and he advocated tirelessly for them. He could be blunt, but he was always constructive. Tom would tell you things no one would else would tell you. During my time as editor of Computerworld, I came to trust and confide in him about nearly everything.

Beginning in the late 90s, I joined Tom several times for day-long sailing excursions around Boston Harbor. He took to sailing like all his other hobbies: with passion. Tacking through the Harbor Islands with Tom was like watching the Travel Channel. When he wasn’t relating the history of landmarks along the route, he was explaining the geometry of sails or the complexity of nautical charts. He never lectured; he was just sharing. He always shared.

He was a burly guy with a gravely voice and boundless good humor. No matter how serious the situation, he could find a way to make a joke. His Twitter profile is one word: “Michievist.” Leave it to Tom to invent a word to describe himself.

He had a low-key laugh – “Heh heh heh” – and an enormous smile. He was quick with a wisecrack, but also with a gentle word of reassurance. When times were tough, he was the guy you wanted at your side because he always found a way to remind you that it’s just not that big a deal.

God, I miss him.

Mary is asking that any donations in his memory be made to Compassionate Cares ALS, at Cards and notes can be sent to 10 Allen Place, Sudbury, MA 01776, and you can reach Mary by at,

A Word of Mouth Campaign for Clean Drinking Water

Charity:Water Supergenius promotionNext week’s Supergenius conference in New York, organized by GasPedal and word-of-mouth pioneer Andy Sernovitz, promises to be a fantastic event, at least if last December’s inaugural conference was any indication. Andy is trying to use Supergenius to demonstrate the power of word-of-mouth in support of a very worthy cause: fresh drinking water for the developing world.

GasPedal has partnered with charity: water, an organization whose goal is to bring clean drinking water to the estimated 1 billion people on earth who don’t now have it. I’m supporting the effort and I hope you will too. Equally important, I hope you will tell your friends and help prove that grassroots organization using the tools of social media really can make a difference.  You can sign up for Supergenius here. Use the discount code “paulismyhero”.

The fund-raising goal is $50,000, which can provide clean water to 2,500 people in 10 communities. 100% of public donations go directly to water projects. All operating costs are covered by a group of private donors so every dollar you give can go to people in need.

Charity:water has already funded more than 2,000 projects, but that’s just a start. Deforestation and the rapid expansion of deserts are making access to water the greatest human crisis of the 21st century. Please do whatever you can to help. This page makes it easy for you to contribute and to share the message.

Watch the video. It’s inspiring.

Want 2010 Red Sox Tickets? I Got Some

As my Twitter followers already know, I’m a rabid fan of the Boston Red Sox. In fact, I’ve been a season ticket holder since 2004. I go to 18-20 games myself and sell the rest to family and friends. This year I still have a few pairs of ticket to sell, so if you want to grab any, here they are.

These are right field box seats, Section 7, Box 93, Row MM, seats 9 & 10. They’re about 50 feet past the Pesky pole and about 15 rows off the field. See the photo below.

The price is $100 a pair, which is my cost plus a couple of bucks s/h. That’s still below the $104 price at the box office for comparable seats.

The Red Sox have sold out the last 550 games in a row, so if you’re from out of town and want to get to a game this year, this may be your best bet without paying scalping fees.

To reserve a ticket, simply reply in a comment to this post or send me an e-mail at paul [at] gillin [dot] com. No tweets, please. I have enough moving parts to keep track of!

Date Time Opponent
4/20/2010 7:10 PM Rangers
4/21/2010 7:10 PM Rangers
4/22/2010 7:10 PM Rangers
5/10/2010 7:10 PM Blue Jays
5/11/2010 7:10 PM Blue Jays
5/12/2010 1:35 PM Blue Jays
6/15/2010 7:10 PM Diamondbacks
6/17/2010 7:10 PM Diamondbacks
8/24/2010 7:10 PM Mariners

View from Fenway Park Section 93, Row MM