Like I mentioned yesterday, I finished Steven Johnson’s book on Emergence. Following are various thought provoking tidbits I found:

5 Fundamental Principles to follow if you’re building a system to learn from the ground up:

  1. More is different
  2. Ignorance is useful (..better to build densely connected systems with simple elements and let the more sophisticated behavior trickle up)
  3. Encourage Random Encounters
  4. Look for patterns in the signs
  5. Pay attention to your neighbors

(page 77-78)

Paul Krugman’s (another MIT guy) 1995 lectures entitled “The Self Organizing Economy“, which relies on two primary axioms: 1) There must be a tension between centripetal and centrifugal forces, with neither too strong, and 2) The range of the centripetal forces must be shorter than that of the centrifugal forces: business must like to have other businesses nearby but dislike having them a little way away.

Alexa is a software application that ‘recommends’ websites based on what site you’re currently visiting, “… the ‘intelligence’ of Alexa is really the aggregated wisdom of the thousands — or millions — of people who use the system. The computer churns through the millions of ratings in its database, looks for patterns of likes and dislikes, then reports back to the user with its findings.” Further, “It’s worth noting here that Alexa is not truly a ‘recommendation agent’; it is not telling you that you’ll like the five sites that it suggests. It’s saying that there’s a relationship between the site you’re currently visiting and the sites listed on the pull-down menu.” (page 124) Curiously, it looks to me like Alexa is the technology that powers a large part of Amazon’s personalization engine.

Official Emergence magazine

Marco Dorigo’s Ant Colony Optimization project

Further reading:

Weaving the Web: The Original Design and Ultimate Destiny of the World Wide Web

How Buildings Learn: What Happens After They’re Built

A History of Civilizations

War in the Age of Intelligent Machines

The Computational Beauty of Nature : Computer Explorations of Fractals, Chaos, Complex Systems, and Adaptation

Alan Turing: The Enigma

Gödel, Escher, Bach: An Eternal Golden Braid

Emergence: From Chaos to Order

The Death and Life of Great American Cities

Dynamic Patterns : The Self-Organization of Brain and Behavior

The Age of Spiritual Machines: When Computer Exceed Human Intelligence

Bots: The Origin of New Species

Turtles, Termites, and Traffic Jams : Explorations in Massively Parallel Microworlds

The Race for Consciousness

The Clustered World : How We Live, What We Buy, and What It All Means About Who We Are

Tim Berners-Lee on the Semantic Web

Today I T’ed it to MIT Building NE43, took the elevator to the 5th floor and sat in a cozy room with 50 or so other geeks & academics taking in a presentation by Tim Berners-Lee on the Semantic Web. Wait… Tim Berners-Lee! The creator of the World Wide Web for goodness sakes! Wow. After getting past the fact that I was sitting not more than 10 feet away from the guy who created the WWW AND the first web browser, I realized two things: a) Tim Berners-Lee is *really* excited about the stuff he does (isn’t it interesting to listen to someone who LOVES what they do?!) and b) the Semantic Web will be a big thing, although it’ll take awhile.

I won’t try to explain the Semantic Web, the W3C site does an excellent job of describing it technically, and there are other sites that cover it as well. Some interesting tidbits I wrote down (for myself):

· N3 or Notation 3 “… a quick notation for jotting down or reading RDF semantic web information, and experimenting with more advanced sematic web features.” [source]

· HTML is to the WWW as RDF is to the Semantic Web

· The Semantic web is a technical problem, but also a philosophical one. In order for massive adoption to occur, we (how quickly I claim participation!) must have some sort of specification that explicity describes what this is and what that is (see: ontology), which is why there is a working group devoted solely to the problem of describing objects for the web.

· Humourously, at one point Tim mentioned that he had wanted a ‘why’ button in his browser.

· The word ‘trust’ came up multiple times during the conversation, something that we’re more and more likely to see as the Semantic Web makes ‘meaningful’ information 24×7 365. (interesting paper @ Caltech on this subject)

· Various software tools mentioned during the presentation:

Blindfold (“… a semantic web software toolkit. It can be used by programmers as a library of handy functions or at the command line to perform various data transformations.”, written by Sandro Hawke, who was at the presentation),

Algae (“Algae is a constraint-based query interface based on algernon. Aside from syntax, the pricipal difference between algae and algernon is that algae is not tied to an accessor (algernon assumes triples are accessed by subject only)”),

Cwm (“… a general-purpose data processor for the semantic web. It is a forward chaining reasoner which can be used for querying, checking, transforming and filtering information. Its core language is RDF, extended to include rules, and it uses RDF/XML or N3 serializations as required.”),

Annotea (“.. a LEAD (Live Early Adoption and Demonstration) project enhancing the W3C collaboration environment with shared annotations.”, which I immediately began thinking of using along w/ a wireframe implementation. The idea behind Annotea is that you could/can leave “..comments, notes, explanations, or other types of external remarks…” on any document available on the web, thereby increasing collaboration among teams. It’s built into the Amaya browser, you can download Annotea here or use the demo w3c.org annotea server),

Haystack (“The Haystack project makes use of Semantic Web RDF technology to permit maximum flexibility in describing information. On top of this we have created a user interface paradigm capable of displaying various kinds of information in many forms. We provide a powerful infrastructure for incorporating machine learning algorithms that implement personalization and per user adaptation. Our framework also facilitates and fosters collaboration between Haystack users.”), check out this screenshot of Haystack, reminds me of some of the Flash portals I’ve seen demo’ed by Macromedia.

Zakim, “An irc bot created by Ralph Swick that interfaces with W3C’s Zakim teleconference bridge and provides meeting support services such as notification of participant arrival and departure, requests to speak, and requests for agenda items.” Tim talked about this bot briefly, reminded me of Jabber’s Chatbot.

Tim mentioned some of the challenges that the Semantic Web faces in his presentation (which was supposed to be on the web, but I don’t see it here, anyone know where it is?). Anyway, after you get by the technical and philosophical issues, there remains the small tiny matter of UI, how do you display all this information to the user so that he/she uses it? Haystack tries to address this… but there’s definitely a business opportunity there for the person(s) that can effectively present vast amounts of information to a human being in a finite space.

After the presentation was over, someone asked a question about patents relative to the Semantic Web. Tim replied something to the effect that if everyone would keep their hands off the patent machine, we’d probably get alot farther faster (he referenced the VoiceXML area which is supposedly mired in patent hell and thus developers aren’t flocking to the standard(s) as they should)…

That’s all! Thanks MIT, W3C and Tim for a great seminar and for opening it up to the general public. The Web continues to be fascinating, largely thanks to your non-selfish efforts.


LOGO (mentioned in the Emergence book I just read), has been re-released in Java as StarLogo by a group at MIT [slashdot]. According to the StarLogo site, it is a “… programmable modeling environment for exploring the workings of decentralized systems — systems that are organized without an organizer, coordinated without a coordinator. With StarLogo, you can model (and gain insights into) many real-life phenomena, such as bird flocks, traffic jams, ant colonies, and market economies.” If you’ve ever wondered why traffic jams form, you can now model them! 🙂

great introduction to using SOAP, ASP and VB

After posting an order on the ecommerce site I’m working on, we have to get tax information and authorize the users credit card. Both of these functions (Vertex for tax and viaWarp for Credit Cards) live on a different machine than the webservers in the cluster, so I thought it made sense to use SOAP to send and retrieve messages re: Credit Cards & Tax information (especially since the IT department would like to re-use whatever we build with other pieces of their infrastructure). This is a great introduction to using SOAP, ASP and VB.

The Only Guide To A Winning Investment Strategy You’ll Ever Need

Finished “The Only Guide To A Winning Investment Strategy You’ll Ever Need” on Thursday. If you’re not working in a bank and you didn’t pay attention during Finance during school, you’ll appreciate this book. I wanted to have a deeper understanding of stocks, bonds, mutual funds, and all the related financial jargon. This book did the trick. I tried summing up the book a couple posts ago, but I think that Jonathan Clements does it best in the WSJ:

“The case for indexing rests on a piece of unassailable logic. Investors, as a group, cannot outperform the stock market, because together they are the market. In fact, once costs are figured in, investors collectively are destined to lag behind the market average.” (source)

Various quotes and tidbits I want to remember/record:

Re: trying to pick the top stocks/mutual funds vs. investing in an index fund: “Jonathan Clements, a columnist for the Wall Street Journal, said it best: ‘I believe the search for top-performing stock funds is an intellectually discredited exercise that will come to be viewed as one of the great financial follies of the late twenthieth century.” (pg. 38)

Re: market timing: “During the 3,541 trading days of 1980-1993, an investor who built and held a portfolio consisting of the S & P 500 would have realized annualized returns of 15.5 percent per annum. If, in an attempt to time the market, an investor missed out on just the best 10 days, the annualized return dropped to 11.9 percent. This investor, by being absent from less than 0.3 percent of the trading days, would have lost over 23% of the returns available for the entire period. If the same investor had had the misfortune of missing out on the best 40 days (or about 1 percent of the total trading days), his or her annualized returns would have dropped to 5.5 percent, a loss of almost two-thirds of the passive investor’s returns. Another way of looking at this is that the returns of the investor who missed out on just the best 40 days could have been matched by owning risk-free certificates of deposit at a local bank.” (pg. 69)

Re: volatility and average annual return: “… These two examples point out the powerful impact that volatility can have on a portfolio and the relative unimportance of ‘average annual rates of return.’ Investors can’t spend average rates of return. What should be of greatest concern to investors is the compound growth rate of an invested dollar. Next time you read an ad for a mutual fund, check to see if they are disclosing the average annual return or the compound growth rate of each invested dollar. The latter is the real gauge of a fund’s performance and the only way to compare that performance to any other’s.” (pg. 138)

check out the sample portfolios on page 160, the Asset Allocation Time Horizon Guideline on page 174, review the chapter on rebalancing and style drift (page 193).

summary in chapter 11:
” · Markets are generally highly efficient!
· While it is not impossible to outperform an efficient market, the odds of it being done, even by professionals, are very low.
· Because past performance is not a reliable predictor of future performance, it is impossible to forecast which money managers and market gurus will be the lucky few.
· The question of whether or not markets are efficient is an interesting academic question. However, the real question facing investors is whether the correct strategy is active or passive management. I’ve given clear evidence that even when market ineffciencies may exist, markets are not so inefficient that active managers can overcome the costs of their efforts and the taxes generated by their trading activity.
· With institutions now controlling a market share approaching 90% of all trading, the competition among professionals is too tough, thereby making active management nonproductive.
· Efforts to time the market are doomed to fail, because so much of the action occurs over very brief time frames.
· The use of active managers causes an investor to lose control of the asset allocation decision, the single most important determinant of the expected return and risk of a portfolio. Active managers expose investors to style drift, which can have unanticipated and nasty consequences.
· Most of what is published by trade publications and the rating services and aired by so-called financial experts is really nothing more than investment pandering. To attach anything other than entertainment value to such reports can be dangerous to your economic health. These experts are all part of the 6% solution — the loser’s game — hyped by Wall Street because it is in its best interest, not yours.” (pg 226)

Further reading:

Capital Ideas and Against the Gods by Peter Bernstein
The Portable MBA in Investing edited by Peter Bernstein
Bogle on Mutual Funds by John Bogle
Investment Policy by Charles Ellis
A Random Walk Down Wall Street by Burton Malkiel
Investment Strategies for the Twenty-first Century by Frank Armstrong (available on the internet)