In the article, “Free information and the Efficient Market Hypothesis“, I give an example of how valuable free information relevant to investments is available on the Internet and how this is largely unexploited.

This free “open source” information calls for open source intelligence techniques (OSINT) to be useful. See: Open Source Capital Market Intelligence.

Free information has a time cost

The problem with “free” investment related information today is that there is so much of it.

Only a tiny fraction is harvested and published by commercial sources like Standard & Poor’s.

Anyone with a computer can get at this information, but because the data is raw, unfiltered, and buried in mountains of irrelevant, useless text, it takes time and skill to dig out what is worthwhile.

In other words, free information has a “time cost” that is a significant barrier.

To break through this barrier, new technologies are required.

From the assembly line to crowdsourcing

Standard & Poor’s, Moody’s, and the rest of today’s commercial providers are all rooted in 19th century industrial techniques.

Some types of financial information (like financial statements and stock prices) are amenable to industrialization.

This information can be processed with standard tools and forced into cookie-cutter patterns, producing a standard product at a predictable cost.

Take a look at a typical report published by Standard & Poor’s and you’ll see the industrial nature of the product.

However, other types of essential information (securities terms and conditions, contract details, laws, regulations) require tailor-made human intervention and are not amenable to industrial methods.

Furthermore, the amount, location, and nature of relevant information that is worth mining is not known before you start digging.

In other words, you don’t know whether you’ll find gold, iron, truffles, of King Tut’s tomb — you can’t use just one kind of tool and start digging. You don’t know how long you’ll be digging or who might be interested in what you find.

To solve this problem, we must abandon the logical solutions of the industrial age and try something completely different.

What has been shown to work in recent years is to get many people — a crowd of participants — to volunteer to spend valuable time in working together — without a leader — to produce a product that will be freely shared by all.

This seems counterintuitive — even crazy — but when done correctly, it works.

This is the essence of crowdsourcing.

Crowdsourcing explained

Crowdsourcing comes in many forms and is used for many different purposes.

Wikipedia is one famous example. The Apache operating system is another.

Two recent best sellers describe the basics of outsourcing and leaderless organizations at length: Wikinomics: How Mass Collaboration Changes Everything and The Starfish and the Spider: The Unstoppable Power of Leaderless Organizations.

For crowdsourcing to work, the participants in the system — those that are donating their time and effort — must be getting a greater return for their effort than they are putting in.

How and why this occurs is different from case to case — and the “what’s in it for me” factor is often not obvious to the casual observer.

In the case of crowdsourcing investment research, the value of collaborative research is not that hard to discern — at least at one level.

If I want to build a portfolio of 100 securities and it takes me 75 hours to do open source research on each security, obviously the market will have passed me by before I complete my research.

On the other hand, if I invest based only on the limited industrialized information published by Standard & Poor’s, I’ll have no competitive advantage and can expect mediocre returns.

However, if I can get five friends to help doing original research, each taking on a different security, we’ll each get five times the output by collaborating as we would working alone.

Better yet, if I can get 10,000 people to pitch in, the return for each participant will be vastly greater than the input.

The question is, how can I manage that?

The technology of crowdsourcing

In order to get 10,000 people to work together productively for a common purpose, without a leader, takes some very special technology and a solid system.

Most visitors to Wikipedia come away with information, without any real idea of the less visible side of wiki organization that makes it possible. Each crowdsourcing system has its own rules, technology, and motivational structure that is relevant to its particular purpose.

Wikipedia has many policies that make it inappropriate for capital market research. For example, the goal is defined as “neutral point of view” rather than “truth”. Original research is prohibited. Much open source information is not considered as being acceptable. A criteria of “notability” would preclude information about the tiny audit firm that signed off on Bernard Madoff’s accounts — until it was too late. Articles that are deemed to “promote” a particular organization are routinely deleted.

Capital Market Wiki, opened to the public in March 2009, after 30 months of preparatory work, is a crowdsourcing tool designed specifically for collaborative researching of open source investment intelligence.

Superficially, Capital Market Wiki resembles Wikipedia, but it actually employs quite different technology and concepts:

• Capital Market Taxonomy: a knowledge structure that helps to avoid redundancy and organizes information.
• A semantic database: automatic classification systems that allow readers to find answers to such free-form queries as “print a table of all options traded in Asia or South America, based on soybeans, with settlement in currencies other than dollars”.
• Standard formats: a system of suggested outlines for articles, based on a standard taxonomy, that saves time in open source research.
• Parallel motivational worlds: “What’s in it for me” benefits that are embedded in the software and that provide advantages for editors with different objectives: finding employment, teaching open source intelligence in schools, collaborative investors, entrepreneurial side ventures, ethical promotion of institutions, and support for institutional research departments.
• Multi-language support: Each article can be written in any language and will share the same semantic database
• Leaderless governance: Clear policies on wiki governance, finances, and continuity.

These are just some of the high points, but I’ll explain Capital Market Taxonomy and some other features in a future article.

In the meantime, you can click on any of the above links and get an immediate explanation.

Crowdsourcing governance

Perhaps the most mysterious and interesting thing about crowdsourcing and leaderless organizations is how such structures are managed.

The rules, of course, are different in each case, but the following description of the role of editors in Capital Market Wiki, quoted from the “governance” page, is typical:

Most of the people involved with this wiki are collaborating editors, who join voluntarily, are self-directed, communicate with each other as they see fit, and leave whenever they feel like it. There is no fee or cost associated with joining or becoming a collaborating editor, there is no application form, no approval process, and editors are not paid a salary or wage. Collaborating editors are self-organizing, setting up projects on the Community Portal, or acting alone, as they prefer. When editors first join, they find a wiki that has a structure, rules, and behavioral norms, which they are free to modify and improve by communicating with other editors on “talk pages”, working for consensus towards an ever better encyclopedia.

Any editor can reward the good work of others by giving merit awards, or editors may criticize, revise, and even delete what others have done. No one “owns” the articles in the main encyclopedia which all are free to change as they see fit. No one is a “boss” or “expert”, to whom others must bow.

The wiki is set up so that editors may remain anonymous, so that wiki reputations are kept separate from real world identities.

As in all wikis, some editors are vandals, spammers, flammers, exhibiting various manners of “bad behavior”, but many collaborating editors are willing to spend time reverting harmful entries, healing damage, and calling attention to those who would bring down the encyclopedia, so that steps may be taken to ban, blacklist and ostracize harmful editors from the community.

The relative power and influence of collaborating editors depends upon their wiki reputations and the time and effort they are willing to devote to building and improving the encyclopedia.

Although there is no central wiki organization to pay collaborating editors, the wiki is structured to provide economic benefits to participants — outside of the system. Such benefits range from non-monetary advantages such as education and free information, to the practical benefits of anonymous job and business networking based on a semantic database. As the wiki develops, there will be the possibility of cash prizes and free-lance consulting contracts, as the patron system is put in place.

There is no ranking or apprenticeship for collaborating editors; anyone may join the wiki and immediately start to edit articles, without approval from anyone else. The power of individual editors places them at the top of governance structure of this wiki.

You can join Capital Market Wiki today as an editor.

This entry was posted on Wednesday, April 22nd, 2009 at 12:54 pm and is filed under Individual Investors, Technology. You can follow any responses to this entry through the RSS 2.0 feed. You can leave a response, or trackback from your own site.

|