8Issue 2.0.8, April 2008
Jimmy Guterman, from Money 2.0, page 1
“A year after
first looked at what financial markets and
Money 2.0By Jimmy Guterman
Looking for the New Pond
Advice from Money:Tech conference chair Paul Kedrosky on how to overcome skepticism and how not to look like a “twit.”
The Wider Impact of Money 2.0
Collective money management shows one way Wall Street and Web 2.0 could work together.
By Marc Hedlund
What Do Hedge Fund Managers
Want from Web 2.0?
They’ll want more once they understand it more. By Cathleen M. Rittereiser
Seven Hot Prediction Markets Tips
Best practices for starting and managing a prediction market in your company.
Published six times a year by O’Reilly Media, Inc.,
1005 Gravenstein Highway North, Sebastopol, CA 95472
This newsletter covers the world of information technology and the Internet — and the business and societal issues they raise.
One year later, the connections between Wall Street and Web 2.0
are getting stronger and moving in surprising directions.
Jimmy Guterman is the editorial director of O’Reilly’s Radar group and editor of Release 2.0.
One year ago, we published an issue of Release 2.0, entitled “When Markets Collide,” in which we considered what Wall Street and Web 2.0 might have to teach one another. Quite a bit, it turned out: the key parallels we uncovered include latency (both have to do their jobs more or less instantly), connectivity (that’s the liquidity of Web 2.0), sensors and actuators (and how to use them), and reputation (stockbrokers are no longer curators—they’re rated, in public).
Since the publication of that issue, we’ve seen a tremendous amount of activity as Wall Street and Web 2.0 size up one another, culminating in our inaugural Money:Tech conference, intended to bring the two sides together in the same room, which we held in February. So it’s a ripe time to consider the status of the relationship. What’s new? What’s changed?
On one level, there’s nothing new about Wall Street looking to new data sources to discover alpha, an edge. In No Bull: My Life In and Out of Markets, hedge-fund legend Michael Steinhardt notes that long ago, in the early 1980s, he “constructed my own New York City taxi index relating the percentage with ‘available’ lights to those occupied, hoping for more to be ‘available,’ thereby signaling a slowdown in demand.” He discovered a proprietary economic indicator around the availability of taxis and traded on it. And Peter Lynch, manager of the Magellan mutual fund for Fidelity Investments, famously suggested that individuals buy stocks based, in part, by what they could see for themselves: in particular, how full a store’s parking lot was.
But Lynch didn’t limit himself to anecdotal data sources. As portfolio man-ager Omid Malekan writes, “Lynch also looked at all sorts of fundamental and technical information on top of how full a store’s parking lot looks.” —>
Our original coverage of the collision between financial markets and web markets can be found at http://downloads.oreilly. com/radar/r2/issue188.8.131.52.pdf. You can find out more about the inaugural Money:Tech conference at http://en.oreilly.com/ money2008/public/content/home. The second one will take place in New York on February 5–6, 2009.
Tim O’Reilly, from When Markets Collide, page 1
“Web 2.0, like Wall Street, is a series of markets in which mere milliseconds can make an enormous diﬀerence. The more you look, the more you see what the two sets of markets have to teach—and warn—one another.”
Release 2.0.8 May 2008 Money 2.0 Jimmy Guterman
How do I know Omid Malekan? I don’t. But I do have a good sense of what his portfolio is, because he posts at least some of it online, via Stockpickr, a website that offers many professional portfolios and rates them. He has some interesting investing ideas, some of which I agree with and some of which I don’t. But, because his informed advice is available on the public web, I can decide for myself what’s useful.
The amount of financial data available publicly (that is, not including propri-etary systems like Bloomberg’s) is astonishing. That doesn’t mean it’s all useful, though. Here’s how the Federal Reserve Bank of Cleveland estimated the proba-bilities of various outcomes at the March Fed meeting:
That chart is begging for someone like Edward Tufte or the Stamen Research team to rip it apart, yet it does illuminate one of the key issues that come up when Wall Street and Web 2.0 executives talk to one another: There’s plenty of data out there, but it’s plenty confusing. You can’t extract alpha until you understand what you’re looking at. As Michael Simonsen, president and CEO of Altos Research, puts it, “free data on the internet is a mess.”
If anyone doubts that financial markets and technology markets are deeply intertwined, consider this: the same day that JPMorgan Chase revealed its “pur-chase” of Bear Stearns, a Gartner Group analyst released a report showing that
Source: http://www.clevelandfed.org/ Research/policy/fedfunds/index.cfm
“the financial services industry continued to lead all vertical markets in server revenue, as it accounted for 25.3 percent of worldwide server revenue in 2007.” As goes one set of markets, so will go the other.
In this issue of Release 2.0, we consider the Wall Street/Web 2.0 mashup from a number of angles. We talk to Paul Kedrosky, chair of our Money:Tech conference and an influential blogger on the topic (as well as others), about why some on Wall Street hate Web 2.0—and what Web 2.0 can do to infiltrate Wall Street nonetheless. Entrepreneur Marc Hedlund, now chief product officer for personal finance startup Wesabe, examines what happens when hidden data gets surfaced. Cathleen M. Rittereiser talks to hedge fund managers to discover what they think they want from Web 2.0—and what they’re actually getting. Longtime Radar contributor Nathan Torkington digs deep into prediction markets and spells out both how to manage them and what companies might gain from implementing them.
In search of lost time
The challenges facing those who want to extract alpha from new data sources or new ways of understanding older data sources are immense. And the most press-ing one may be time. Michael Stonebraker, chief technology officer of StreamBase Systems, says, “the minute you store data, you lose.” His company’s system, using a $1,000 generic PC, can, he says, run 300,000 messages per second.
Some time, maybe not so far away, that may not seem so fast. David Leinweber, a Haas fellow at the University of California at Berkeley, says, “To get an edge, you have to get the news before the news people get there.” Leinweber estimates that in the 1980s, before the web, a market’s reaction time to certain market events was measured in weeks. By the late 1990s, as graphical web browsers achieved hegemony, the reaction time to earnings surprises could be measured in mere minutes. Now, as we’ve reported previously, firms are moving their data systems closer to Wall Street so they can shave a millisecond or two off their latency time.
Time is a component of the three variables that measure the value of information, according to John Mahoney, cofounder and chief technology officer of Info(n)gen (http://infongen.com). He says those three variables are quality, relevance, and scarcity. “Quality and relevance have needed to step up as a result of technology improvements,” he says, “but scarcity has been completely redefined.” —>
Two notable data points:
1. Job listings are always a good place to see where an industry is going. So we pass on that the venerable bank Wells Fargo is looking for a “social media developer” who will, among other things, “lead the develop-ment and impledevelop-mentation of complex web sites and web-based widgets and tools on our blogs and popular social networking sites, like Facebook.”
2. Where a company decides to work says something about what it’s doing to avoid latency. From an April 8 Reuters article, “New York more fertile ground for tech startups”: “The pace of technology innovation is now
far more rapid in New York, where a string of local start-ups are working on software prod-ucts for the financial services and insurance industries.”
David Leinweber’s rarely-updated but often-provocative blog is at
Release 2.0.8 May 2008 Money 2.0 Jimmy Guterman
Where is the useful information? It’s easy to say that financial markets are in trouble when, as Robert Passarella of Bear Stearns noted at Money:Tech, people working on trading floors are looking to Wikipedia to find what financial terms mean. But it might also mean that Wikipedia is a more important source of information to traders than originally thought and that brings with it opportuni-ties for data miners (and data hackers). Information isn’t scare, but useful and tradeable information is. Renny Monaghan, senior director of product manage-ment at salesforce.com, says the question is “How do I avoid being a news junkie?” How do we know which of infinite possible inputs is worth our attention? And how do we do it quickly? Traders, after all, tend not to believe in delayed gratification.
Any patient person can, like Michael Steinhardt, note the comings and goings of New York taxis. That’s not scarce information. Indeed, it’s so free that hardly anyone bothers to track it. It’s in paying attention to something so clear that it’s invisible that information becomes scarce—and valuable.
Paul Kedrosky expresses the problem with an ecological metaphor: all the obvious sources of information have long been overfished. So how to find new bodies of water? Ben Lorica of O’Reilly’s research team says there are plenty of ways to track events without measuring them directly. For example, someone can sit outside your house and find out what movie you’re watching on your Slingbox, just by following the packets. It’s easier to just ask, of course, but as Marc Hedlund notes (see page 12), sometimes the information we cast off without intention has the most value.
There are plenty of companies scouring for the value in that information. SkyGrid (http://skygrid.com/) gives hedge fund managers and research analysts tools to measure market sentiment in real-time. Collective Intellect (http://www. collectiveintellect.com/) is an “alternate research” firm that provides insights from social media, also in real time. A future O’Reilly Radar research report will evalu-ate all these companies in detail.
FirstRain (http://firstrain.com) is another one of the companies trying to sell new kinds of research to Wall Street. It is expensive—$10,000 per year per seat, but that’s a little over half what a Bloomberg Terminal costs. Indeed, Bloomberg must be FirstRain’s target, having signed a distribution deal with CapitalIQ, a Bloomberg competitor.
Please write me at firstname.lastname@example.org if you’re interested in learning about this research report as we develop it.
There are plenty of ways
“The disconnect continues,” says Martin Betz, FirstRain’s vice president of technology. “I was surprised when I arrived here at how disconnected Wall Street is from technology. It’s amazing how much the web is untapped or simply uninteresting to many finance people. A big part of what our sales force has to do is explaining to hedge fund managers that there is quite a lot of useful data on the web that they haven’t seen. Those few hedge funds who are using data from the web usually say they’re getting analyst reports or Google alerts. Those are very primitive ways of thinking about data on the web. The perception that new and valuable information is available hasn’t taken hold in a widespread way. But it’s out there. Looking for new data sources is not what we do predomi-nantly. Most of the data we crawl and collect is publicly available. It’s just that much of it is obscure and unfiltered.”
As you’d expect of an executive whose company is in the business of selling data to such a reluctant audience, Betz maintains that “data from the web, if collected and analyzed properly, will give you an advantage. The noise of the blog world has led the conservative people who run Wall Street to think that the whole system is noisy or junky and that’s simply not true. For companies like ours to succeed, we have to deliver the information in a format that’s understood by the program-trading tools these people are using.”
Betz says speed isn’t necessarily the key part of the value proposition FirstRain is promoting. “We don’t promise you that you’ll see something first. What we promise is that you’ll see aggregated and filtered information you wouldn’t have seen otherwise.” So it’s a qualitative edge as much as a quantitative edge that Web 2.0 firms are selling to Wall Street. “Trading is still full of human beings collecting qualitative information and then testing it.”
That qualitative information is what’s next for Wall Street. You can see the astonishing, ubiquitous success of the Bloomberg Terminal as the definitive— for now—solution to Wall Street’s quantitative data problem. And, despite its leading position, those who use it don’t find it bloated (in contrast to, say, recent iterations of Microsoft Windows). The proprietary Bloomberg Professional system is the standard for monitoring and analyzing financial market data in real time. It is speedy , in part because the two-monitor system is text-based rather than built around a graphical user interface. Those in “the business” say it’s a great vehicle for personal messages as well as quantitative information. Traders don’t send you a message; they “send you a Bloomberg.” —>
Release 2.0.8 May 2008 Money 2.0 Jimmy Guterman
But you can also see a double-screen Bloomberg Terminal, however imposing, as merely the finite solution to the first phase of Wall Street’s analytics problem. What’s emerging now is a second phase, one more aligned with Web 2.0 notions such as harnessing collective intelligence, that concerns itself with extracting data—and, ultimately, alpha—from unstructured text, sentiment, mentions, and buzz. It’s Bill Tancer of Hitwise predicting everything from unemployment statistics to American Idol winners by tracking website visits. It’s MarkMail (http://markmail. org/) analyzing mailing list archives in new ways that uncover unexpected patterns. It’s LinkedIn trying to turn its social network, nearly 20 million strong, into a research platform with Wall Street as its prime market. It’s companies like Weatherbill (http://weatherbill.com) building risk-management services around the weather. It’s people like John Seo of Fermat Capital Management, constructing catastrophe bonds around such information. It’s Steve Skiena, formerly best-known for his jai-alai algorithms (http://jai-tech.com/), opening TextMap (http:// textmap.com/), a specialized search engine that analyzes both the temporal and geographic distribution of news. It’s Rick Seaney of FareCompare ( http://farecom-pare.com) extracting new meaning out of air travel data. It’s about people and companies finding new applications for data that wasn’t considered worthy of mining—or wasn’t considered at all.
“Don’t think in terms of new data sources,” says Joshua Schachter, formerly of Morgan Stanley and now at Yahoo! “New data sources are better described as new ways of viewing the data. In five years, I bet people will view the reading of finance articles on the web as a very dated activity.” He thinks some filter will have to come between the data and the consumer of the data. “It’ll grab all
the interesting but obscure sources that you’d like to read, but even an RSS feed can’t keep up with them: local newspapers and industry publications, blogs that you’d never want to read every day but one day every two years something relevant to your business pops up.”
There are limitations to what computers can tell us about qualitative data. APIs, web-page scrapers, and spiders can capture and organize massive amounts of information quickly, but they’re not as swift or effective at giving a bulletproof indication of what sentiment is embedded in text. Computers don’t do irony or sarcasm all that well, at least not yet, and so much of the text on the web, partic-ularly in the blogosphere, is informal and full of all the complex sentiment that humans do so well and machine learning is slowly catching up to. Commerce-related sites are easy to track, because they’re formalized with standardized numbers like ISBNs and ASINs. Those listing products to sell on eBay, for example, have a strong incentive to identify those products precisely and quantitatively. But someone on her blog raving about the record she just bought isn’t so incented. The next Bloomberg will likely be the company that can do for the qualitative what Bloomberg did for the quantitative.
Reuters is one of the companies competing with Bloomberg right now. At Money:Tech, Tim O’Reilly spoke with Devin Wenig, now CEO of the merged Reuters and Thomson Financial, and Wenig highlighted what he considers the two biggest trends hitting financial (and other professional) data: the increasing impact of consumer media on professional media (young traders are used to more diverse data inputs) and—more provocatively—a decreasing importance in latency. Wenig suggested that, over time, Reuters’ business could move from news to insight derived from news, more about making connections than simply gathering information. It’s a move, in a sense, from descriptive data to predictive data. Wenig stated emphatically that semantic markup would be crucial to this happening.
As Tim wrote in his post-interview notes, “Ultimately, Reuters’ news is the raw material for analysis and application by investors and downstream news organizations. Adding metadata to make that job of analysis easier for those building additional value on top of your product is a really interesting way to view the publishing opportunity. If you don’t think of what you produce as the ‘final product’ but rather as a step in an information pipeline, what do you do differently to add value for downstream consumers? In Reuters’ case, Devin thinks you add hooks to make your information more programmable…That’s a really good case for the Semantic Web.” —>
Joshua Schachter, inventor of del.icio.us, suggests that one project worth doing would be a program that spiders the “who we are” pages on a company’s website. That list of senior managers is often changed prior to an explicit public announcement. Schachter says, “When someone is added or ejected from the page and there’s no announcement, that’s an unusual thing. A spider would be a way to find out before you’re supposed to find out, still using public data.”
We covered the Semantic Web in the October 2007 issue of Release 2.0, issue 2.0.5, entitled “Looking for The Web’s Edge,” available at
Release 2.0.8 May 2008 Money 2.0 Jimmy Guterman
We are impressed by Wenig, but we’re not so sure that latency isn’t a problem anymore. As we quoted Haas fellow David Leinweber earlier, “To get an edge, you have to get the news before the news people get there.” Latency is becoming less and less about the raw data and more and more about what people are either paying attention to or have the tools to pay attention to. As Henry Blodgett of Silicon Alley Insider puts it, “The only research that’s valuable is the stuff nobody else has.” And what if everyone “has” it? Do something with it. In the words of Nouriel Roubini, who runs the premium news and analysis site RGE monitor (http://www.rgemonitor.com/), “We filter what’s available from the web for free and people pay us for that.” There’s still an edge to be had, but it’s coming at a different, higher level. And that edge will come from a combination of what machines and humans (who Tim O’Reilly calls “the last mile of extracting meaning”) can recognize and act on.
From buzz to trade
“You need both,” says Kate Niederhoffer, vice president of measurement science at Nielsen BuzzMetrics, a measurement firm. “You gather evidence and a panel of people corroborates the evidence. The panel tells you how people behave. The buzz data tells you why.”
“Buzz” is elusive. It’s something companies and individuals want their offerings to have—and it’s something investors want to understand and predict. We’re still in the earliest stages of leveraging buzz and some of those early findings simply support time-worn notions (i.e., advertising spend is the best predictor of buzz), but companies are learning that, particularly when applied to consumers, buzz itself can have a predictive quality.
Some hedge funds are looking for what Jerry Needel, senior vice president, product, at Nielsen BuzzMetrics, calls the “low-hanging fruit” of buzz: needles than can be extracted from haystacks, such as when an engineer at a large technology company reports to fellow developers that something will be delivered late. But the most interesting practice hedge funds have for this data so far is using buzz to aid in predictive modeling. Much buzz data right now parallels stock movement, but there have been cases in which buzz has offered early warning: Needel reports that the company saw buzz regarding the once-popular Atkins diet peak a full three quarters before sales topped out.
Don’t expect Nielsen to start a buzz fund anytime soon, though. “Buzz is a piece of the equation when it comes to trading, but it’s certainly not the whole equation. And we’ve had our best performance with long strategic investors, not in-and-out daily investors.”
“There are plenty of investors who might be interested in what teenagers are saying about wireless or other brands,” says Niederhoffer. “What our data gives them is access to a leading indicator. Then the people who trade for a living, after they consult our index of overall tonality and buzz, can get a sense of how that relates to stock price. It’s diagnostic information and, most of all, a deep dive into what people are saying on a particular topic.”
Best practices on both sides of the line
As David Leinweber writes, “Whatever raw material you choose, fooling yourself remains an occupational hazard in quantitative trading.” With Leinweber’s warning in mind, here are four rules what those trying to bring financial markets and web markets should keep in mind.
Web 2.0 companies looking to sell to Wall Street should
1. Sell solutions, not data.
2. Be clear about the reach and the limitations of the offering.
3. Make conservative, demonstrable arguments for your solution. Wall Street can be wary of Web 2.0. Overpromises will make you seem like a “twit” (see our interview with Paul Kedrosky, page 10).
4. Understand that Wall Street is notoriously stingy with data. The notion of “collective intelligence” goes against the behind-closed-doors mentality of generations of Wall Street executives. You must be clear as to why more open can mean more profitable.
Wall Street companies looking to take advantage of Web 2.0 should
1. Remember that Web 2.0 is about harnessing collective intelligence, not snazzy-looking web pages. It’s not just MySpace out there.
2. Be confident that the traders are still the traders. What Web 2.0 offers is new data and new ways of considering the data. They’re powerful inputs, but they’re part of a broader set of inputs.
3. Accept that new data sources necessitate new ways of discovering and manipulating data.
4. Accept that new data sources will be shared—just not immediately.
It’s a truism that alpha lasts longest when it’s hidden. That may have been true in the past, but the growing use of Web 2.0 tools means that less data will stay hidden, and what’s hidden will stay hidden for a shorter period of time. We offer the last word to James Altucher of Stockpickr: “When it comes to data nowadays, closed source is a myth.” nn
Release 2.0.8 May 2008 Jimmy Guterman
What doesn’t Wall Street understand about Web 2.0?
At a high level, Wall Street just thinks that Web 2.0 is frivolous. Wall Street, on the whole, equates Web 2.0 with Ajax wizziness or Facebook. The people on Wall Street think that’s all Web 2.0 is. And if that’s all Web 2.0 is, they don’t care. They are very, very hung up on appearance.
At a lower level, Wall Street goes wrong in understanding Web 2.0 when it assumes that people in the Web 2.0 world have the same motivation they assume everyone involved in financial markets has: money. They don’t understand how there can be any value in community. If you know something, the thinking goes, you’re an idiot if you tell someone else. Collaborative behavior is fundamentally antithetical to their being. They don’t just think it’s wrong. They think it’s stupid.
So how can these two groups play together nicely—and profitably?
People talking about Web 2.0 or Wall Street 2.0 have to show that it’s about more than cosmetic changes, hosted apps, and snazzy apps. They have to show that these tools can help you make money, that it doesn’t matter anymore if you don’t know what, say, a synthetic option is. Different tribes speak different languages. So, to succeed on Wall Street, Web 2.0 people have to talk in the language that makes people money. Community, for example, can help you make more money. The logical chain around that has to be made obvious. That’s been missing so far.
Several Wall Street old-timers I talked to for this issue of Release 2.0 said
that they don’t trust any Web 2.0 people. One said, “If the idea is so good, why aren’t they trading on it themselves?”
That’s so true. I’m surprised every person you spoke to didn’t say that. That’s how they think: If you have something that could make people make money, you’re either stupid or wrong or both. That’s the Number One objection any Web 2.0 person going to Wall Street has to be prepared to deal with. You can’t just say you have this fantastic piece of data that will give alpha. Do that and you’ve anointed yourself a twit. It’s more about how you come in and talk about your-self and what you have to offer. You’ve got to show sophistication and nuance.
Paul Kedrosky, proprietor of the Infectious Greed blog (http://paul.kedrosky.com) and conference chair of O’Reilly’s Money:Tech conference, is a leader in understanding what Wall Street and web markets have to teach one another. We asked him how Wall Street and Web 2.0 were getting along nowadays.
“Wall Street doesn’t
understand how there
can be any value in
Looking for the New Pond
Looking for the New Pond
“New data requires
Can you get more specific about that?
There are two ways to show that you’re not a twit. First, at a high level, you have to say something like, “I know that constructing a trade is about finding an information edge—and then constructing a trade around that edge. I have a little information that will give you better ideas of when to buy and sell. I don’t construct trades.” Be clear about what you do and what you don’t do. You almost have to pander. Say something like, “I’m a lonely technology guy with some unexploited data. You understand the market impact and risk control.” Promise less and you’ll show that you understand more.
Second, make it clear that you understand that the information you’re selling them is just part of the mosaic. You’re not telling them that this piece of eBay data alone lets them trade eBay on a daily basis. You’re telling them that this piece of data—when it’s combined with other factors—might make it possible for them to trade eBay more effectively.
If you were selling data to Wall Street, what would you want to sell?
by Marc Hedlund
In Berkeley, California, there are a number of auto shops that will service your BMW. The best-known and best-trafficked of these is Weatherford BMW, the local dealership, which has a prominent location just at the entrance of the free-way leading to San Francisco. A large “WEATHERFORD BMW” sign announces its presence to people zooming by or, as is more likely, to those stuck in traffic. People all over town recognize the building and use it as a landmark, regardless of whether they would ever drive a BMW or need one repaired. You can’t miss it. On average, people who get their cars serviced at Weatherford spend about $1,300.00, give or take, per visit. When asked to rate Weatherford’s service, on a scale of 1–100, 100 being best, Weatherford averages a score of 17.
Across town, at the end of a one-way street in an industrial area, a nearly unmarked cinderblock building houses a small independent auto shop, Bavarian Professionals. You could park your BMW directly in front of it and go to the nearby brewpub — the only reason you’d be likely to wander by at all—and fail to notice that your car was in unusually similar company. You could search Google for “berkeley bmw auto shop” and entirely miss it—even Berkeley Parents’ Network, an extremely well-used local recommendations site, hardly mentions it.
Yet on average, people who get their cars serviced at Bavarian Professionals spend about $600.00, give or take, per visit—less than half the cost at Weatherford. When asked to rate Bavarian Professionals’ service on the same 1–100 scale, Bavarian averages a 96—79 points higher than Weatherford.
Quite a contrast! These two data points tell a clear story, one that would benefit any BMW driver in town. You’d certainly think that Weatherford would have no business at all, and Bavarian would be buying out the brewpub to make room for more customers. That isn’t the case, though; at least, not yet, since those data points aren’t easily available. Where would you go to find the average amount spent at one local auto shop versus another?
This hidden economic data could make consumers into far better shoppers, if only it were not hidden.
Harnessing collective money intelligence
One of the key tenets of Web 2.0 is that extremely powerful tools can be built by harnessing collective intelligence. When one person acts in a way that a web application can observe and record, the potential knowledge in that action can be released onto the web, and made available to everyone, usually permanently and for free. Tools such as del.icio.us aggregate information about web book-marks, revealing the reading and interests of its users. Last.fm, similarly, exposes
Marc Hedlund is the chief product officer of the personal finance startup Wesabe (http://www. wesabe.com) and a contributor to the O’Reilly Radar (http://radar.oreilly.com/marc/). Previously he was entrepreneur in residence at O’Reilly Media. He blogs at “Wheaties for Your Wallet” (http://blog. wesabe.com/).
The Wider Impact
of Money 2.0
Sometimes implicit measurements deliver more useful
information than explicit ones.
the music listening patterns of its members, showing the new hot hits and the old favorites together in a way that a Billboard chart could never approach. Like paths cut in the grass between buildings, the tracks of masses of users on these sites point the way towards clarity.
The Money 2.0 websites, including that of my company, Wesabe, take this approach to collective intelligence and apply it to finances. Looking to the idea that greater transparency in markets leads to greater efficiency, these sites aim to give users financial power through information. Across a range of topics— cash management, investment, retirement, goals, loans, and more—people are starting to manage their finances on these sites, and in so doing, they are contributing to, and benefiting from, the collective intelligence that emerges from their actions.
The impact of this movement goes far beyond cataloging tastes or interests. It has the potential to change markets, to shift the balance of power between consumers and producers, to overwhelm brand association, convenience, and word-of-mouth, and replace those weak signs with more rigorous data inter-pretation and fundamental analysis.
Take the example of the auto shops, above—a real set of data that emerged early in the development of Wesabe. Site members participate in the application by uploading their bank and credit card statements, editing their transactions to categorize and consolidate their spending records, and then giving feedback on the merchants where they shop. Unlike five-star rating sites, which depend on explicit review postings to get information on merchants, Wesabe learns from the shopping patterns of our users, boiling down a set of transactions into a report on a set of merchants. Each new user uploads several hundred transactions at a time, and each transaction speaks articulately about that user’s financial world. They tell us how much users spend, and in what circumstances—was this restaurant purchase tagged ‘lunch’ or ‘dinner’? What is the cost different between those tags?
They tell us when users make switching decisions, such as going from Safeway to Whole Foods, and whether they stick with their choices or go back to their old vendors. And they tell us how a purchase fits into a consumer’s overall budget— where rent falls for this area, or the income of the people who buys clothes at a certain store. —>
When one person acts in a
way that a web application
can observe and record, the
potential knowledge in that
action can be released onto
the web, and made available
to everyone, usually
Likewise, Money 2.0 investment sites such as Cake Financial (http://www. cakefinancial.com) and Covestor (http://www.covestor.com) allow members to share investment decisions with peer groups and the world. Rather than looking at market shifts as a dynamic of prices for holdings, these sites look at bundles of individuals, observing how each person understands the market, and how they react to it.
Looking at financial data differently; looking at the movement of an individual in a market; looking for patterns in spending that suggest sentiment: These common themes have started the Money 2.0 movement, and are leading us towards a new way of understanding finances altogether.
From personal finance to information that moves markets
Our “hidden” fundamental metrics, then, are no longer hidden—or, at least, they are emerging from the fog. All of the Money 2.0 sites are limited by the data users have uploaded and—we hope—edited, tagged, or rated, so inevitably, the Weatherford BMW and Bavarian Professionals prices quoted above are biased by the sample of Wesabeans uploading—a self-selecting and technologically-biased group. Still, with more than half a million users of Money 2.0 personal finance sites, and hundreds of thousands more using Money 2.0 investment and loan sites, these populations are many times larger than those that make up the Nielsen rating panel or the Conference Board’s Consumer Confidence Index.
How, then, should we look at this data—how do we understand the averages shown for Weatherford and Bavarian? Do they tell us anything about larger eco-nomic trends? Are they, essentially, emerging ecoeco-nomic indicators?
The Money 2.0 sites are pointing in the direction of far greater economic transparency for consumers and investors both. While it is too early to draw widespread economic conclusions from the current sites, it is easy to imagine that a site like Wesabe or one of its competitors reaching the scale of an earlier Web 2.0 success story such as Flickr would lead to a profound new data source with significant results for the economy as a whole. If the top Google result for “Weatherford BMW” shows a competitor charging half the price and returning five times the satisfaction, that is likely to change both Weatherford’s and Bavarian’s positions in the market—rewarding the most efficient supplier.
Someone believes in Covestor. At press time, it received $6.5 million in a Series A round of funding led by Union Square Ventures and Spark Capital. Cake Financial is funded by Alsop Louie and various angel investors.
Likewise, if consumers are setting and working towards financial goals, those goals and the progress towards them have the potential to tell us far more than a survey of consumer confidence. How many durable goods were ordered last quarter, and how far along are goals related to those goods in this quarter?
It is not simply the availability of this data that is important—transaction data is only part of the equation. Money 2.0 users affect this data with their edits, tags, reviews, comments, and recommendations, and all of these actions, undertaken in most cases for the user’s own record-keeping or enjoyment, lead to a richer set of data about the transaction: a view into the sentiment behind the purchase. A very low-rated mobile phone transaction, for instance, may suggest that a consumer will bolt as soon as a contract expires. Looking solely at the mobile phone company’s revenues or forecasts would miss the frustration users would be inclined to express towards a bad supplier in whose clutches they are (temporarily) captive. Uploading, editing, sharing, reviewing: These simple acts create a financial data stream richer and more evocative than we have had before. Money 2.0 tells its tales.
The investment community has seen the broad impacts of greater information flow, from Bloomberg screens to automated trading networks and beyond. Until now, however, that benefit has been most broadly distributed for public markets and securities. With the advent of these Money 2.0 sites, tracking of financial fundamentals is moving “down-market,” into the hands of consumers, who are highly motivated to extract more value from each dollar they earn (especially now as the credit crunch takes hold). Show such a consumer a side-by-side comparison between two auto shops, or two plumbers, or two restau-rants at which they might eat dinner, and the understanding that data brings will change the way people shop. No one, it turns out, likes spending twice as much to get a fraction of the value. nn
By Cathleen M. Rittereiser
The inaugural Money:Tech conference, presented by O’Reilly Media in February, featured a lively panel on what hedge fund managers want from Web 2.0. Panelists included JP Rangaswami, managing director for BT Design and previously CIO for Dresdner Kleinwort Benson; Sean Park, founding partner of Sixth Paradigm; and “Finbar Taggit,” a hedge fund manager who blogs pseudonymously. We asked
Cathleen Rittereiser, moderator of the panel, to take the story further. —J.G.
Hedge fund managers have put plenty of energy into dismissing Web 2.0, in large part because they equate it with its most public consumer implementations (see Paul Kedrosky interview, page 10). Indeed it’s hard to argue that “biting” a colleague on Facebook and turning him into a vampire will provide any alpha. But technologies in general—and Web 2.0 technologies in particular—are crucial to the business of hedge funds. Enlightened hedge fund managers do appreciate the impact of technology on what they do. Finbar Taggit attributed the growth and rapid expansion of the hedge fund industry to technology. “New hedge funds can come up to speed now because of technology. For backtesting, I have grabbed data off Yahoo, for free. There are day trader systems that are comparable with those at the banks.”
“It’s a battle between investment banks and hedge funds,” Taggit says. He believes banks unwittingly provide hedge fund competitors with an advantage by limiting employees’ Internet access to specific websites such as Facebook. “Hedge funds want to find market imperfections and trade on them.” With
investment bank employees limited in what they can see, he feels he gains an advantage.
Yet in conversations with hedge fund managers and technology experts, consultants, service providers such as prime brokers and administrators, and vendors, I learned that when it comes to Web 2.0 technology, many hedge fund managers do not know what they want, because they do not know what Web 2.0 means. They do not know what it means, because, in their view of the hedge fund business model, they do not make money for thinking about it. As one leading quantitative investment analyst put it, “Web 2.0 has little to no relevance to anything we or our clients do. I would have nothing to say on this topic.”
He’s wrong—or maybe he thinks Web 2.0 is just Facebook vampires. Yet we know of at least one business in which a hedge fund purchases data and out-sources analysis of that data to a third party, which combines it with other data and returns the new data, with accompanying insights, to the hedge fund. That
What Do Hedge Fund Managers
Want from Web 2.0?
Many of them see Web 2.0 as being about consumer-facing
new technologies. But what it offers them is less flashy and
more likely to help their bottom line.
Cathleen M. Rittereiser is a hedge fund marketing and investor relations expert focused on the institutional investor market. She is the co-author of
Foundation and Endowment Investing: Philosophies and Strategies of Top Investors and Institutions.
Banks unwittingly provide
combining of data from multiple sources is nothing less than a mashup, a hall-mark of Web 2.0 information-gathering and information-dissemination. Yet that analyst is not alone, so it’s worth considering why some hedge funds have been slow to adopt Web 2.0 methods.
Why would hedge funds defer Web 2.0?
Culture: Druce Vertes, a hedge fund technology consultant, explained that
there hasn’t been much adoption of Web 2.0 technologies by hedge fund man-agers, because “Wall Street firms are late adopters. The culture is top-down and hierarchical, not conducive” to Web 2.0 concepts. “The notion of mashing up Morgan Stanley and Goldman Sachs research is fairly laughable—but in 10 years might be commonplace.” He says that, to some degree, regulators are a limiting factor. “Investors collaborating, talking about their book on blogs can be controversial, or even seen as evidence of an investor group or market manipulation conspiracy.” Additionally, he thinks “the many-to-many model is fundamentally disruptive and tilts the game away from the professional.”
Since Silicon Valley tends to be years ahead of everyone else, the slower adoption rate in the hedge fund industry is natural. Hedge fund managers are not Twittering trades en masse. It can take years for companies to agree on a market-leading product and make all their programs work with it. Wall Street’s slowness in adopting Vista, Microsoft’s most recent iteration of Windows, is not unlike its slowness is installing content management systems a decade ago. Technologists are not driving the technology implementation decision in hedge funds, especially for new products and services.
Economics: Gerald D. Mintz is the president and CEO of PerTrac Financial
Solutions, a front-and-middle office platform that provides investment analytics and workflow management tools to hedge funds, hedge fund investors, and other investment professionals. He explained the lack of awareness of Web 2.0 technology in economic terms.
“Most hedge funds start small, do not want to spend money on technology or anything deemed non-essential to the investment process and performance. They usually start out tracking something like investor relations on an Excel spreadsheet and, as they grow, they end up with a huge set of linked Excel spreadsheets. There is very little automation.”
This approach to technology makes sense to them because of the hedge fund business model. Mintz says, “hedge funds are very evolved at using the things that make money. When they get started, they are focused on trading, their track record, building capital. They don’t know what they will make until
they start to earn fees, so they spend as little as possible.” Therefore, hedge funds will not likely think or act on any technology concept such as Web 2.0 unless or until they find it economically necessary.
As hedge firms grow, Mintz says, “they get to be a $3 to $5 billion firm and their technology has not kept pace with the scope of their business.” Hedge funds are “a tough business because of the economics. Some firms may know from the start that they want to spend the money on technology, but there’s so much risk that it is hard for them to imagine spending the money until it becomes important.” And with so many Web 2.0 tools available for free, it’s hard to use cost as an argument against them.
Practical: Hedge funds tend not to pursue new technologies for practical
reasons. Generally, hedge fund firms tend to be small and do not have the time or manpower to integrate standalone or new technology. Whether a piece of technology is built using open source code, Web 2.0 software, or wire and string, they want to use what they know already works.
“There’s a lot of potential but not much worthwhile going on,” Vertes says. Blogs as a source of primary information is one area with potential, as is RSS. “RSS delivery of information makes a lot of sense. RSS is a lot better than email.
It’s an ideal tool for distribution of research.” Although he questions the timing, overall Vertes believes Web 2.0 “will be huge, because the tools simply let you pro-cess more information faster. As an added bonus, lowering the barrier to publish-ing and distribution means you get access to a more diverse set of viewpoints.”
How is the hedge fund industry dealing with Web 2.0?
While Vertes sees Web 2.0 having more meaning and impact for hedge fund managers and firms further into the future, other industry participants have either embraced or accepted the inevitability of Web 2.0 technology and have begun to incorporate it into their businesses to varying degrees.
Ron Suber, a senior partner at Merlin Securities LLC (he was, until recently, president of Spectrum Global Fund Administration), qualifies as an embracer. He says, “Web 2.0 is going to dominate. It gives web designers new freedom. It allows this next generation more creativity and a better user experience.” Granted, design and usability are just the most visible layer of Web 2.0, with its power built on harnessing collective intelligence, but it’s the way in for many businesses, hedge funds and otherwise.
Investment analytics and tools continue to be a big driver of technology spending and hedge funds tend to buy or invest in technology “that gets the job done, not win awards,” says Marc Solomon, director of global markets
As hedge firms grow, they
find that their technology
has not kept pace with the
scope of their business.
financing and services at Merrill Lynch. PerTrac’s Mintz echoes that idea: “The industry is reasonably well served on the trading side with all sorts of trading tools and technology. But as soon as you go from trading to the middle and back office of the firm—investor relations, accounting—there are not that many solutions.”
How are hedge fund service providers deploying Web 2.0?
Suber, Solomon, and Mintz, as providers of technology and technology-dependent services to hedge fund firms, have the most at stake right now in the evolution of Web 2.0 technology and have made the most progress thinking and acting on it. Here’s what they’ve learned:
Customizing and sharing data: Solomon has observed a trend he calls
“sharing of customization” in which hedge funds that have access to online, dynamic reporting are increasingly looking to share customized information internally. Similarly PerTrac is beginning to take advantage of Web 2.0 within its existing products, creating services as simple as a trip planner, essentially a mashup that connects contact information saved within a clients’ system to web-based information (such as Google Maps).
Software as a service: Suber has observed applications with SAS models
like salesforce.com and the niche-oriented service Backstop (http://www. backstopsolutions.com/) making inroads at hedge funds and other investment management firms. Clients have raised concerns about data security, but they have also become more comfortable because large firms like Merrill Lynch have vetted and used these services.
Technology-light models: Web 2.0 technology is lighter on the client side,
making the services easier and faster to deploy. Suber’s previous company, Spectrum, recently built a global portal to serve hedge fund functions such as compliance, investor relations, and fund reporting. The product went from idea to design to deployment in 95 days. “In the past it would have taken six months and cost over $1 million. Web 2.0 shortens the cycle of product development and differentiation.”
Information and contact networks: Suber reports that LinkedIn has
facilitated further networking and at the same time, made recruiting much more transparent. “Fifty percent of the time when you work with an executive recruiter, you get a resume with the person’s name omitted,” he says. “I can go to LinkedIn and search on keywords and find the name of the person. LinkedIn has diminished the time and effort it takes to stay connected to your network and also makes staying in contact that much easier.” —>
What’s the Future for Hedge Funds and Web 2.0?
Cost efficiency: Merrill’s Solomon believes hedge funds will continue to focus
on keeping their costs low. “With alpha getting tighter, hedge funds are looking to get more efficient on the cost side. The more they trade complex products, the more they want to get efficient on the finance, admin, and operations side.”
Opportunity for vendors: Solomon has become aware of new web-based
services that handle tasks like road show management. At the same time, how-ever, he reiterates that hedge fund managers will not pay a lot extra for extra services. Ultimately he thinks it will create huge opportunities for the incumbents— embedded vendors like Bloomberg and Reuters—to push out more services and create new products for their client base.
Change their thinking or your timeframe: Mintz feels optimistic about the
prospects for Web 2.0 technology and thought processes taking hold in the years ahead. As people get comfortable with the technology, he says, “there will be opportunities for sharing information in a closed group. Hedge fund managers are not necessarily tech people. They haven’t done this kind of work. They’re analysts and traders and are not familiar with thinking about how to use tech to run their business.”
Providing solutions: Responding to the observation that there may need to
be more intermediation of technology in the future, Solomon says, “Managers can build proprietary technology or outsource to third parties. We see a trend toward packaging, putting people and technology together to extend and deepen our relationships with our clients.”
Web 2.0 enables better products and services. As it allows service providers to automate and support all aspects of managing a hedge fund, helps build more effective business relationships, and uncovers more alpha, Web 2.0 will undoubtedly create Hedge Funds 2.0. Many (most?) hedge fund managers may not know yet what they want from Web 2.0, but hedge fund managers need the business management, automation, cost efficiencies, support and partnerships Web 2.0 technologies and concepts can help deliver. nn
Meredith Jones, a managing director responsible at PerTrac, says “The amount of information and the flow of information has changed dramatically” over her decade in the industry. She cites the social networking functionality of an industry news and com-munity website, Albourne Village (http:// village.albourne.com/), that allows industry members to post questions and share infor-mation. In the past, investors had to network actively in order to get information about managers. Having the ability to access a network in an open community allows more investors and managers to get information without having to be physically networked to them. Jones sees the proliferation of hedge fund blogs as another example of the type of open information that changes how people in the industry relate, network, and gather information.
One market in which Wall Street and Web 2.0 are undeniably colliding is a prediction market, a speculative market intended to encourage and test predictions. It takes the guessing about future value that is the core of so many financial markets and connects it to the hallmark of Web 2.0, harnessing collective intelligence. If we truly believe that all of us know more than some of us know, markets in which interested (and unconflicted) parties do their best to predict the future may yield much better decision-making. —J.G.
Players in a prediction market bid on shares in outcomes and the price of an outcome’s shares reflect its consensus probability. In recent years, prediction markets have become quite trendy, but it’s one thing to create a prediction market and another to have it be useful and applicable to better understanding real-world markets. To learn what works, Release 2.0 spoke with Professor Robin Hanson, who came up with the idea of prediction markets, and Bo Cowgill, who manages a groundbreaking prediction market, called “Prophit,” inside Google.
Tip 1: Know whether you can take bad news
Hanson draws a distinction between “morale markets” and “decision markets.” A morale market informs decision-making rather than being the primary grounds upon which a decision is made. In Hanson’s words, “morale markets help people feel involved.” The problem comes when the decision is a difficult one and the market points to an unpleasant course of action.
Decision markets are those where management says, “we want to know the answer, even if it’s unsettling.” These can work, says Hanson, but you have to take great care to get those with the relevant knowledge involved. They also require the whole company to buy into the process, to follow the advice of the market even if it goes against a senior manager’s intuition.
Tip 2: Ask questions that will get trades
“Our prime consideration is liquidity,” says Cowgill. “We feel markets are more rewarding for players when there are a lot of trades.” Often this involves asking questions that a lot of people have insight into. The goal is to avoid the situation where an order is offered but not accepted.
To solve the liquidity problem, some prediction markets have a “market maker,” an algorithmic agent that buys and sells stock. The argument for market makers is, “if the bot makes dumb offers, traders can make smart buys.” Prophit doesn’t have a market maker as Cowgill wasn’t happy writing a bot whose job was to lose money. —>
Seven Hot Prediction
How they work, what you can learn from themNathan Torkington has chaired the O’Reilly Open
Source Convention and other O’Reilly conferences for more than a decade. He ran the first web server in New Zealand and was one of the founding Radar bloggers
Some of the most well-known public predic-tion markets are InTrade (http://www.intrade. com/), the Iowa Electronic Markets (http:// www.hsx.com/), and the Hollywood Stock Exchange (http://www.hsx.com/). Release 2.0.8 May 2008 Seven Hot Prediction Markets Tips Nathan Torkington
Release 2.0.8 May 2008 Seven Hot Prediction Markets Tips Nathan Torkington
Tip 3: Ask questions with definite outcomes
Market questions should have a precise date and be measurable so you can pay off the successful traders. The idea is to provide a feedback loop so players know how well they did. “You could get similar results simply by making people look at how well their estimation did in the past,” says Hanson, “but nobody ever does.” So “will life be better under a Democratic president?” isn’t a question for a prediction market, but “will inflation go up 1% in the first six months if a Democrat wins the presidency?” is.
Tip 4: Ask ungameable questions
Cowgill also aims for markets that a lot of people have insight into, and where “one or two people can’t control the outcome”. For example, “will Bob pass $1 million in sales this year?” would incent Bob to bet on the answer and not meet his sales targets. This would be bad.
Tip 5: Don’t use real money
Using real money in a prediction market turns prediction into gambling. Consequently, all U.S.-based prediction markets use “play money” and don’t simply reward successful traders by turning their market gains back into real money. At Google, for example, they trade in artificial “Goobles” but award prizes: cash, t-shirts, books, even small social occasions with guest speakers.
There are hosted overseas markets, such as InTrade, which deal in real money (and make their profit on commissions). Research from Wharton and Yahoo! has shown that accuracy isn’t affected by choice of real or play money.
Tip 6: Some questions shouldn’t be asked
There are also forbidden questions. For example, Cowgill says, “It’s possible that questions like `what will our stock price be?’ can act as conduits for inside information and all those who saw the market would have to be regulated as insiders. We don’t ask those sorts of questions.”
Some questions simply may not be how the company does business. There were Prophit markets proposed about what a competitor was going to do. The team lead said, in effect, “I don’t care what they do and this market would make it look as though we did.” The market didn’t launch. Similarly, the U.S. government proposed a Terrorism Futures Market to predict the likelihood of various types of terrorist attack. It was shelved after public outcry at the appearance of gambling on and profiting from a terrorism attack.
Tip 7: Keep players trading with a leaderboard
While there’s research showing you only need 14 or 15 people to get accurate results, Cowgill points out that it assumes those people trade often. Google doesn’t use reminders or nags to get people to play, only emailing them when their order is matched. “People play because they find it fun,” he says. “There’s some element of competitiveness. We run a leaderboard.”
The leaderboard reveals the identity of the top 10 players for bragging rights, but in general Google traders play anonymously. Cowgill doesn’t see that as inherent to its success: “I think people here think independently and would trade against a founder. There are bragging rights to be able to say `I predicted X, Y, and Z before the rest of you.’”
Sure enough, Google has found that traders are more motivated by reputation than prizes. “At one point I was late giving out the cash prizes,” says Cowgill, “and nobody noticed. But everyone writes in if I’m a day late with final rankings—they want to learn whether they got a t-shirt.” nn
“You could get similar results
simply by making people look
at how well their estimation
did in the past, but nobody
1,463 Google employees have made trades
25 markets each quarter
30% of markets are “fun”; 70% are work-related
Seven weeks out, markets beat a 50:50 guess
$10,000 total prizes each quarter
10.5% lower returns on optimistic (good for Google) securities over neutral ones
Overcoming Bias <http://overcomingbias.com>: from the creator of prediction markets, a blog evaluating ways in which we can overcome our natural human biases
Mercury’s Blog <http://blog.mercury-rac.com>: fascinating blog from prediction market consultants
The Wisdom of Crowds James Surowiecki’s bestseller brought prediction markets to popular consciousness.
Inkling Markets <http://inkling.com>: offers hosted prediction markets to enterprises.
InTrade <http://www.intrade.com.>: public participation markets with real money.
Using Prediction Markets to Track Information Flows <http://bocowgill.com/GooglePredictionMarketPaper.pdf>: Bo Cowgill reports on Google’s prediction market.
The Nantucket Conference (Nantucket, MA)
A group of investors, entrepreneurs, technologists, and executives head to an exclusive island, go off the record, and imagine the future.
Xtech (Dublin, Ireland) http://2008.xtech.org/
The theme this year is “The Web on the Move,” focusing on the emerging portability of data, applications and identity on the Net.
The New Yorker Conference (New York, NY) http://conference.newyorker.com
The venerable magazine promises its second annual event will deliver “stories from the near future.”
O’Reilly Where 2.0 (Burlingame, CA) http://conferences.oreilly.com/where/
If you’re interested in geo, this is the, er, place to be.
D (Carlsbad, CA) http://www.allthingsd.com
Last year Walt and Kara got Bill Gates and Steve Jobs to speak together. Who will they pair up this time?
Flight School (Boulder, CO) http://www.edventure.com/fs08/
Release 1.0 editor Esther Dyson continues to blast off into the world of air taxis and space travel.
O’Reilly Graphing Social Patterns East (Washington, DC)
How is the social graph changing your business?
Supernova 2008 (San Francisco, CA) http://www.supernova2008.com
Kevin Werbach, another Release 1.0 veteran, leads us into the “Network Age.”
Velocity Web Performance and Operations Conference (Burlingame, CA)
O’Reilly debuts a new conference dedicated to Web performance and operations. Are you building at Internet scale?
A selection of significant public events over the next few months.
Personal Democracy Forum (New York, NY) http://pdf2008.confabb.com/
How is technology changing politics? Come to Lincoln Center and find out.
Ubuntu Live (Portland, OR) http://www.ubuntulive.com/
Find out the latest about the most popular Linux distribution. And while you’re in Portland…
O’Reilly Open Source Convention (Portland, OR)
Join more than 2,500 open source developers, experts, and gurus.
Brainstorm Tech (Half Moon Bay, CA)
FORTUNE’s David Kirkpatrick brings together “an invited group of superb tech thinkers and leaders with smart people from other arenas for two days of intense and creative interaction.”
Black Hat USA (Las Vegas, NV) http://blackhat.com/
Find out about the latest in security before it’s too late. Will what you learn in Vegas stay in Vegas?
And Defcon (http://defcon.org/) comes right after, August 8–10.
Web 2.0 Expo NY (New York, NY) http://ny.web2expo.com/
Why should Silicon Valley have all the fun? The megaconference comes to the Javits Center.
Pop!Tech (Camden, ME) http://poptech.org/
This year’s event pivots on the relationship between scarcity and abundance.
Singularity ’08 (be there, wherever you are. It’s a three-day online conference.)
And, if the singularity (http://en.wikipedia.org/wiki/Technological_singularity) occurs by then, we’ll all be there anyway.
city state zip country
email (personal email required for electronic access)
n My colleagues should read Release 2.0, too!
Send me information about multiple copy subscriptions and electronic site licenses.
n check enclosed n charge my (check one) n american express n mastercard n visa
card number expiration date cvs code
name and billing address n same as above n see below name
city state zip country
Please fax this form to 1.818.487.4501 or mail it to Release 2.0, P.O. Box 17046, North Hollywood CA 91615-9588
Payment must be included with this form. Your satisfaction is guaranteed or your money back.
If you have any questions, please call us at 1.800.889.8969 or 1.707.827.7019 or email us at email@example.com.