Unthawing = Freezing
I'm pretty sure that is how it is. Thawing = unfreezing, just to be very clear about what I am saying...
« January 2008 | Main | March 2008 »
February 23, 2008
February 14, 2008
Who ought to tell us what we should know?
Among the latest to get in line for that job is Susan Jacoby who, after over hearing a conversation in New York City after 9/11 comparing the Twin Towers tragedy to the Pearl Harbor bombing that started the Vietnam War, decided she had had enough.
Her first and best idea for combating this? She wrote a book: The Age of American Unreason, criticizing what she identifies as a particularly American Hostility to knowledge. That is an ironic strategy. No one who is hostile to knowledge will read the book. In fact, the curious and informed (turns out the curious always become informed!) have already noted her point and feel her frustrations. Possibly Ms. Jacoby knows that that moment of empathy and identity will help sell the book?
I have not read the book. And my to-read stack is so high it is a hazard to pets and visitors to my home, so I may not. I have been overheard lamenting the lack of knowledge, the ignorance-promoting antics of the current administration, etc. But I have increasing uneasiness with my polemics.
What is the problem with, for example, college students not being able to find Iraq on a map? Should they be planning a trip there soon? Or be preparing to make a return trip, in case they are somehow transported there without a map? A World map or globe is quite abstract with its colors and words and lines—I’m guessing none of which are found on the ground in Iraq. Should we be suggesting topo maps instead of globes?
The only way knowing “where” (in the sense of pointing to it on a map) Iraq is located is a practical question is if you believe a number of other abstractions are also practical questions: Who lives near the Iraqis? What do Iraqis eat? Is it cold or hot in the north? Why do they have a mix of religions, ethnic groups? Why the history of episodic intense interest of the US and Great Britain? And so on…
But these abstractions are only practical in the sense of providing plausible explanations (more abstractions!) for this or that event. Once we make this leap, the rubber finally meets the road when one are asked to do something practical like vote or shoot at someone based on this long, long trail of abstractions about where Iraq is and how it matters.
If the only reason for knowing the location of Iraq is so one can participate in a long narrow trail of manipulation through abstractions, then people will continue to be hostile to knowledge. How can it be otherwise? The curious people I know, experience this knowledge as a rich web of interconnection, explanation, appreciation, beauty. For a long time, the curious among us have taken great trouble to turn abstractions of all kinds into experience. Our trouble with teachings isn’t that we need louder, shiner, more elaborate, better refined or brightly colored abstractions, maybe we simply need to design richer, more accessible experiences.
February 13, 2008
Competition III--Another Iterated Prisoner's Dilemma Tournament
These results are a follow on to my last post on competition and iterated prisoner’s dilemma simulation. In the tournament below, I used the tournament rule that every agent plays every agent at each round. This takes a lot longer to run and the results are different. AvgT4T is still the winner, but T4TForgive beats out T4T.
Tournament 1 was one in which everyone played once each round. This seems more like a competitive environment in which the other players don't get complete information about the how play proceeded in that round--maybe more like competing for jobs or making friends in high school.
Tournament 2 seems to give more complete information at each round as a public auction or public disclosure pricing might provide. I am sure the merits of using one style of tournament over another have been debated plenty.
The difference in outcome illustrates an important characteristic of complex interactions among agents: The initial conditions and the rules of play make all the difference in the world. For those who are committed to free markets, there seems to be a parallel assumption that the ultimate achievement in free markets is one with now rules at all. Citizens of a connected and crowded planet may choose to design the rules of interacting based on our initial conditions. When I play with these simulations for awhile, I start to understand Case's argument against depending on the simplistic dogma of free-markets to answer all questions of our common wellbeing.
Data tables ...
Match Win-Loss-Tie
| Random-T4TForgive: | 1.000, 0.000, 0.000 |
| Defect-Random: | 1.000, 0.000, 0.000 |
| Random-T4T: | 1.000, 0.000, 0.000 |
| AvgT4T-Random: | 0.537, 0.460, 0.003 |
| Random-Pred1: | 0.701, 0.299, 0.000 |
| T4TDefect-Random: | 0.946, 0.006, 0.048 |
| Defect-T4TForgive: | 1.000, 0.000, 0.000 |
| T4T-T4TForgive: | 0.000, 0.000, 1.000 |
| AvgT4T-T4TForgive: | 0.000, 0.000, 1.000 |
| Pred1-T4TForgive: | 1.000, 0.000, 0.000 |
| T4TDefect-T4TForgive: | 0.998, 0.000, 0.002 |
| Defect-T4T: | 1.000, 0.000, 0.000 |
| Defect-AvgT4T: | 1.000, 0.000, 0.000 |
| Defect-Pred1: | 1.000, 0.000, 0.000 |
| Defect-T4TDefect: | 1.000, 0.000, 0.000 |
| AvgT4T-T4T: | 0.000, 0.000, 1.000 |
| Pred1-T4T: | 1.000, 0.000, 0.000 |
| T4TDefect-T4T: | 0.981, 0.000, 0.019 |
| Pred1-AvgT4T: | 1.000, 0.000, 0.000 |
| T4TDefect-AvgT4T: | 1.000, 0.000, 0.000 |
| T4TDefect-Pred1: | 0.970, 0.015, 0.015 |
Match Scores
| T4T-T4TForgive: | 0.500, 0.500 |
| T4TDefect-T4TForgive: | 0.510, 0.490 |
| Random-T4TDefect: | 0.490, 0.510 |
| Defect-AvgT4T: | 0.506, 0.494 |
| T4TForgive-Random: | 0.490, 0.510 |
| Pred1-AvgT4T: | 0.522, 0.478 |
| Defect-T4T: | 0.506, 0.494 |
| T4TDefect-AvgT4T: | 0.516, 0.484 |
| Pred1-Random: | 0.415, 0.585 |
| Defect-Random: | 0.857, 0.143 |
| AvgT4T-T4TForgive: | 0.500, 0.500 |
| AvgT4T-T4T: | 0.500, 0.500 |
| T4TDefect-T4T: | 0.504, 0.496 |
| T4TDefect-Pred1: | 0.511, 0.489 |
| T4T-Random: | 0.499, 0.501 |
| Pred1-T4TForgive: | 0.510, 0.490 |
| Random-AvgT4T: | 0.484, 0.516 |
| Defect-T4TDefect: | 0.506, 0.494 |
| Defect-T4TForgive: | 0.541, 0.459 |
| Pred1-T4T: | 0.503, 0.497 |
| Defect-Pred1: | 0.506, 0.494 |
February 11, 2008
Competition II
I couldn't let this one alone. While reading Case's book on Competition, I decided to simulate the theoretical game Iterated Prisoner's Dilemma. The explanation on Wikipedia is great. I use the canonical PD payoff matrix {(3,3), (0,5), (5,0) (1,1)} where the scores represent points awarded to (agent1, agent2) for combinations of play {(mum, mum),(mum, snitch),(snitch, mum),(snitch, snitch)} respectively.
The simulation runs 200-game matches for 40 rounds of play. I have coded both everybody-plays-once-per-round and everybody-plays-everybody-per-round tournaments, although the results here are only from everybody-plays-once-per-round tournaments. Tournaments start with 25 agents of each strategy type.
I programmed 7 agent strategies:
Average T4T seems to be the population winner in my simulations, settling in at about 40% of the population. I haven't seen any documentation on this effect. It may be an artifact of my particular mix of agents when the tournament starts. Or I may have discovered a new, effective agent strategy. This doesn't seem likely as it is such an obvious strategy. Has anyone seen this effect before? Please drop me a line.
The game win-loss-tie record between agent's strategies was a little surprising to me. Defection wins most of the matches. But it is clear that winning games doesn't correspond closely to achieving the highest scores per round. Coming in a close second against a wide range of strategies is what explains a particular population's survival and growth.
Match Win-Loss-Tie
The simulation runs 200-game matches for 40 rounds of play. I have coded both everybody-plays-once-per-round and everybody-plays-everybody-per-round tournaments, although the results here are only from everybody-plays-once-per-round tournaments. Tournaments start with 25 agents of each strategy type.
I programmed 7 agent strategies:
- Defect (always snitch)
- Random (fair coin toss)
- Tit-for-tat (Do what your opponent did last play--T4T is best overall according to the literature)
- T4T with occasional snitches
- T4T with occasional forgiveness
- Prey on cooperators
- Average T4T (T4T based on the average of all plays to date)
Below are the results of 40 rounds of play. After each round, the total points earned by each strategy are tallied, then the population is redistributed according to the point breakdown. Predatory strategies tend to become extinct after 15 to 20 rounds.
Average T4T seems to be the population winner in my simulations, settling in at about 40% of the population. I haven't seen any documentation on this effect. It may be an artifact of my particular mix of agents when the tournament starts. Or I may have discovered a new, effective agent strategy. This doesn't seem likely as it is such an obvious strategy. Has anyone seen this effect before? Please drop me a line.
The game win-loss-tie record between agent's strategies was a little surprising to me. Defection wins most of the matches. But it is clear that winning games doesn't correspond closely to achieving the highest scores per round. Coming in a close second against a wide range of strategies is what explains a particular population's survival and growth.
Match Win-Loss-Tie
AvgT4T-Random: 0.492, 0.508, 0.000 T4TDefect-Random: 1.000, 0.000, 0.000 T4TDefect-T4TForgive: 1.000, 0.000, 0.000 AvgT4T-T4TForgive: 0.000, 0.000, 1.000 Defect-T4TDefect: 1.000, 0.000, 0.000 Defect-Pred1: 1.000, 0.000, 0.000 T4T-AvgT4T: 0.000, 0.000, 1.000 Pred1-T4TForgive: 1.000, 0.000, 0.000 T4TDefect-T4T: 0.974, 0.000, 0.026 Pred1-AvgT4T: 1.000, 0.000, 0.000 T4T-T4TForgive: 0.000, 0.000, 1.000 Defect-AvgT4T: 1.000, 0.000, 0.000 T4TDefect-AvgT4T: 1.000, 0.000, 0.000 Random-T4TForgive: 1.000, 0.000, 0.000 Defect-Random: 1.000, 0.000, 0.000 Pred1-Random: 0.250, 0.750, 0.000 Defect-T4T: 1.000, 0.000, 0.000 Defect-T4TForgive: 1.000, 0.000, 0.000 Pred1-T4T: 1.000, 0.000, 0.000 Random-T4T: 0.750, 0.000, 0.250 T4TDefect-Pred1: 1.000, 0.000, 0.000
If you compare the percentage of points earned in the (7!)/(2!)(5!) = 21 possible pairings, it is clear that the point leaders fairly evening split the points with all opponents, while the agent strategies the result in rapid extinction lose to some strategies by wider margins.
Match Scores (Fraction of Points)
T4T-T4TForgive: 0.500, 0.500 T4TDefect-T4TForgive: 0.510, 0.490 Random-T4TDefect: 0.492, 0.508 Defect-AvgT4T: 0.506, 0.494 T4TForgive-Random: 0.491, 0.509 Pred1-AvgT4T: 0.522, 0.478 Defect-T4T: 0.506, 0.494 T4TForgive-AvgT4T: 0.500, 0.500 T4TDefect-AvgT4T: 0.518, 0.482 Pred1-Random: 0.397, 0.603 Defect-Random: 0.858, 0.142 T4T-T4TDefect: 0.496, 0.504 AvgT4T-T4T: 0.500, 0.500 T4TDefect-Pred1: 0.512, 0.488 Pred1-T4T: 0.503, 0.497 Pred1-T4TForgive: 0.510, 0.490 Random-T4T: 0.502, 0.498 Random-AvgT4T: 0.503, 0.497 Defect-T4TDefect: 0.506, 0.494 Defect-T4TForgive: 0.537, 0.463 Defect-Pred1: 0.506, 0.494
You can download the code (zip, gzip) here to play with ideas for strategies or see how the dynamics of the system change based on the mix of agents. There is a Readme file explaining how to create new agents with your favorite strategies. Adding new agent types is fairly straight forward and requires only a few lines of new code in most cases. Python 2.5 is required to run the simulations. The analysis programs require MatPlotLib to create the plots. This software is available under a non-commercial Creative Commons License. Have fun!
February 10, 2008
Rivers and Tides (Thomas Riedelsheimer)
Rivers and Tides is a documentary about artist, Thomas Riedelsheimer, and his work in natural, transient sculptures. He makes snaking lines of leaves, grass, ice, contrasting colors of sticks and flowers in circles, loops or floating down a stream.
Thomas works a meditative shift and a new mode of perception. I won't see sculpture, nature and the flow of water in the same way again. And I will think twice before I give up making something because I feel self-conscious about how it fits my social context or somehow doubt its value.
Two concepts in particular struck me:
Utility. What is Thomas' are for? In the current American way of thinking about career, economy and even art, Thomas answer is not adequate. It is for a personal understanding, for a moment of perception. It is there to make fleeting ties between elements to move our frame reference. You can't take it home and hang it on the wall; you can't buy it for museum collection. It can't really be justified by any of our common "explanations" for art.
Attachment. Thomas is remarkably unattached to the product of his work. A stream of leaves in a creek is transient. In fact, none of his art seems intended to be permanent. In one scene we see a line of yellow flowers flow down a swift stream, passing near a yellow circle of flowers in the rocks--a scene designed to be experienced once and only if you the observer doesn't look away at the wrong moment.
Further, the documentary shows many of Thomas's works falling apart with his hands still on them, before they are complete. His detachment in these circumstances is unusual. Thomas sometimes expresses disappointment, but never appears to experience an exasperation the drives him to revise his outlook. He simply starts again.
Thomas's equanimity seems to come from feeling as if he is a particular part of something large. And Thomas expression of his part seems severely critical of many popular claims of what that "something" might be.
Thomas works a meditative shift and a new mode of perception. I won't see sculpture, nature and the flow of water in the same way again. And I will think twice before I give up making something because I feel self-conscious about how it fits my social context or somehow doubt its value.
Two concepts in particular struck me:
Utility. What is Thomas' are for? In the current American way of thinking about career, economy and even art, Thomas answer is not adequate. It is for a personal understanding, for a moment of perception. It is there to make fleeting ties between elements to move our frame reference. You can't take it home and hang it on the wall; you can't buy it for museum collection. It can't really be justified by any of our common "explanations" for art.
Attachment. Thomas is remarkably unattached to the product of his work. A stream of leaves in a creek is transient. In fact, none of his art seems intended to be permanent. In one scene we see a line of yellow flowers flow down a swift stream, passing near a yellow circle of flowers in the rocks--a scene designed to be experienced once and only if you the observer doesn't look away at the wrong moment.
Further, the documentary shows many of Thomas's works falling apart with his hands still on them, before they are complete. His detachment in these circumstances is unusual. Thomas sometimes expresses disappointment, but never appears to experience an exasperation the drives him to revise his outlook. He simply starts again.
Thomas's equanimity seems to come from feeling as if he is a particular part of something large. And Thomas expression of his part seems severely critical of many popular claims of what that "something" might be.
February 09, 2008
Competition-Mobius Thought Loop
James Case's Competition: The birth of a new science caught my eye while I was browsing the stacks in Powell's Bookstore over the holidays. The Amazon reviews seem to hit most of the book-review points I might try to make so no use in reproducing that here.
Case has an interesting background as a pro baseball pitcher and a professional mathematician. In Competition, he is making the argument that classical economics is in the process of maturing toward a "real" science by embracing experiments and increasingly holding theories to the standard of allowing the possibility they may shown false by successful explanations of new data. This is something he both observes and documents at the edges of economic academia. And it is also a cause Case promotes.
Here is one additional thought: This discussion is moving along an ironic and amusing thought loop-maybe, an argumentative Mobius Strip?
Starting on one side, Case explains that it is time for a paradigm shift (Kuhn's kind) in Economics, that data and new analysis abilities are building a compelling argument against the classical economic axioms: Efficient Markets, Gaussian randomness, the tenant that free-agent-self-interest leads to optimizations, free trade always creates domestic benefits (Principle of Comparative Advantage-David Ricardo [1817]--no Wikipedia hits on this), etc. Case is not trying to prove this is happening in one short book; instead he lays out the background science, framing the argument, and citing original works.
Case spends a lot of energy explaining how experiments in actual and simulated competition show that monopoly behaviors (characterized by predatory actions on competitors executed in order to keep markets already taken lucrative) ubiquitous. As you continue to follow the loop, Case goes on to explain some of the ills to society these misconceptions (idealizations) cause.
To show the nature of the resistance, he quotes many criticisms of the "new science" by the classical economists. He catches them ignoring new data, new ideas behind the analysis of competition and real agents (people with partial knowledge, greedy, colluding, etc).
Now in a sense, we seem to be back to where we started. Case may be overlooking the irony or he may have dulled his sense of absurdity, or he may feel irony is not useful in this discussion. It seems that he might have dispensed with the straight face and ended arguing that economic classicists will change slowly or not at all because, with regard of their careers as academics and government advisors, the current monopoly, that is, their current economy, is working great. Why use theory and data to ruin a perfectly good system?
One can imagine a day in the future when some new explanation or technology from the "new science" (empirically falsifiable, complexity, self-organizing complex systems)--or some newer science--is so compelling and valuable that the market decides in its favor over the classical views. Don't expect Microsoft to start feeling uneasy that it might be a monopoly and contemplate breaking itself up. Likewise, why would classical economists give up their central positions to make a few model-generated lines run a little closer to a few experimental data points?
February 02, 2008
Moleskine Fan II
I write in notebooks constantly for projects at work and at home. I write essays, do calculations, take notes from books and keep work notes in them, so I have one or two with me all of the time. As I have confessed earlier, I am a Moleskine fanatic.
I don't get this fussy about much else. But when I got a glimpse of leather Moleskine covers from GFeller Casemakers in Meridian, ID, I couldn't resist. These are beautiful covers!
The pics here show the cover for the small Moleskine notebook. This is number 215 and was made by leathersmith Steve Derricott.
GFeller Casemakers offers beautiful field cases too!
