Sunday, December 28, 2008
For those not familiar with the OLPC project, the idea is to create $100 laptops that can be given to children worldwide to try to bridge the technology divide and empower young people to use technology. The purchase of our OLPC was actually paired with the gift of a second OLPC to a Third World nation.
My son already has a laptop, it turns out, but I was bound and determined to set-up and try the OLPC. My fun began with trying to get build 706 of the Linux OS to recognize our WiFi network. The problem is that I have a hidden SSID (non-broadcast) for security reasons (this is security through obscurity, which is not the best policy, but when paired with strong encryption reduces the threat of easy compromises). So I began the Linux hacks that I am far too familiar with having five Linux servers in my stable.
I could never get the OLPC machine to see the network for more than a moment, unfortunately, so I eventually relented and just exposed my WiFi node. Then I hunted down the upgrade procedure from within a command line shell and spent hours doing an upgrade. The upgrade radically re-arranged the UI's already cryptic iconic interface (I still can't figure out a few of the button functions!) but also brought some improvements to the WiFi connectivity UI.
Overall, though, my wife complained that I spent 6 hours hacking on a computer that was supposed to be useful to African tribesmen. I also have something like 15 years experience with Linux, even running early versions on laptops with experimental X windows servers by 1994 as an alternative to purchasing a Sun "luggable" at the time. I also have innumerable hours pouring over the Linux FAQs/wikis that are always out-of-date or pertain to earlier builds, trying to unravel the correct changes to make something work. I think OLPC has some more work to perfect.
I contrast OLPC with my IPhone 3G that cost only marginally more. If the claim holds true that WalMart may start offering $99 IPhones (8G), the cost would be comparable to the OLPC while providing substantially greater memory capacity. Some of the educational aspects of the OLPC are missing (built-in graphical Python programming, for instance), but everything else is easier and more stable.
Thursday, December 18, 2008
Here he is in the most recent Newsweek building a rickety historical scaffolding to re-explain how an originalist (political originalism to distinguish from legal originalism) interpretation of the Constitution takes a dim view of government involvement in almost anything, how government involvement in almost anything inevitably leads to a request for more government involvement in our affairs (repeating Hayek), and how regulation drives lobbying because regulation drives defensive politics by business entities.
The odd thing about Will's unending exposition of political and economic history is that it minimizes or ignores the complex problems that led to our modern state. Madison's America was a smaller, emptier and more racist place where notions like the tragedy of the commons were only intellectually interesting; the vastness of the wilderness was also a vastness of available resources for exploitation. It took a hundred more years for economic interactions to become complex enough that monopolistic inequities began to become obvious and regarded as unfair. It took longer still for environmental pollutants to be understood and pervasive enough that regulation and environmental law was shaped. It took almost as long for American law to begin to shed itself of racism and sexism and expand that originalist notion of equality to all people.
For American conservatism to move forward there must be something more than raw originalism crossed with Hayek's fears of serfdom. There are too many counterexamples of each in the modern world for those arguments to carry much weight. Instead of fixed ideology, what is needed is targeted creativity. As an example, while environmental regulation is a thorn for conservatives like Will, a targeted improvement might be to focus on streamlining and using technological means to speed the process of regulatory approval. Partisan efforts that broadly deny the value of such regulation can't possibly succeed.
The same is true of health care reform which Will mentions in passing. He rings the topic with Grover Cleveland's assertion concerning originalism but doesn't penetrate to exactly what might be wrong with the specific programs that Obama hinted at in his platform. It's enough for him to proclaim it might be bad because the Constitution failed to mention it in succinct language (unlike, say, the existence of the Navy).
Fair enough. We are warned. Now I want actual policy that goes beyond a single note.
Wednesday, December 3, 2008
There was soggy architecture, rain-soaked and oversubscribed at the California Academy of Sciences. Two giant spheres sandwiched between a living roof and a subterranean aquarium, pushing the roof outward with a graceful femininity and seemingly mocking the functionally irrelevant twisted tower of the De Young Museum directly across from it.
There was a sunset dinner cruise on the Bay, with detailed discussions of the implications of new theories concerning IGF2, imprinting and the spectrum of mental disorders from autism through to schizophrenia as the swells poured in through the Golden Gate Bridge.
Then, finally, there was SFMOMA, Martin Puryear and participatory art projects. Katharina Fritsch's visually stunning Kind mit Pudeln (concentric circles of black poodles surrounding a baby) invoked the embrace of kitsch (yes, Fritsch invoked kitsch) by, say, Jeff Koons, while simultaneously reminding me that the singular requirement of art is discovery of the art by critics and museums. Advocacy is the requirement. It's not that Fritsch's piece is not interesting, but it has the postmodern conceit of denying conceptual depth with a gentle tongue-in-cheek amusement. As observers we spend time interpreting it, but the interpretation is at best a reference to Faust, at worst a reference to Koons.
The social aspect of art was reinforced while browsing a volume on the roots of cyberart in the 90s (and before). The examples and the narratives built around explorations in coastal metropolitan centers and major universities. The author spun a narrative history in a web of advocacy connecting together their friends and acquaintances from that era. There was not, nor could there have been, an appreciation of anything other than what passed through the author's sphere.
Could it be otherwise? Disintermediation and long-tails might hold a key with the internet supporting better access of artists to art markets and observers, but they also promote an unfiltered view of content without sieving by non-amateurs.
Thursday, November 6, 2008
First, I want to speculate on the origins of the phrase "bat shit." There are many contenders as this discussion points out. The most likely case is that the creative spatial metaphor "bats in the belfry" (spatial insofar as the belfry is atop the building and that the bats are like random thoughts colliding about "in the head"--our mental metaphors do tend to be spatial) was co-opted through "ape shit" and "batty" to become "bat shit."
So "bat shit" is an excellent description of some of the memes floating around in Right Wing circles during the run-up to the presidential election. Recursivity chronicles a few of them, including the remarkable claim that Obama was hypnotizing his audiences using some weird variation of neural linguistic programming. To reduce the obviousness of the scatological quality of the phrase, however, I will coin guanocholia to describe a general susceptability and belief in bat shit ideas.
Extraordinary Popular Delusions and the Madness of Crowds demonstrates that guanocholia is far from a recent phenonema, but what concerns me is that the internet now makes it remarkably easy to transmit and sustain these delusions. What once took years of word of mouth to spread, now can spread in minutes from blogs like so much fertilizer mined from a cave.
Is there any inocculation for guanocholia? Likely not, but the American public seems to be tired of erratic temperments and most guanocholics are, by definition, bat-shit erratic.
Sunday, October 12, 2008
I should join in since the first of my September statements just arrived showing a 10% drop in a diverse basket of international growth, domestic plodders and domestic technology funds. I have a feeling that is just a pinprick compared with next month...
How could this have happened?
One line of toxic reasoning is that Fannie Mae/Freddie Mac were forced to take-on subprime mortgages by Congress desiring to spread homeownership. But that appears to have been only a small slice of the entire pie since FM/FM only held 40% of the subprime mortgages (and many of those bundles were acquired late in the game in an effort to shore up the broader market).
There is also the moral/economic analytic dimension based on moral hazard theory that the S&L bailouts of the late 80s combined with the hedge fund debacles of recent years gave a sense of cushion on the downside. But I don't tend to think that fund managers look much at worst case downside; upside is where the profit is and moral hazard reasoning is meaningless when golden parachutes will automatically deploy in contractual severance packages.
Instead, the best available analysis (thanks, Ted) divides across two arguments: (1) risk was hidden (information loss/information assymetry) due to the complexity of the security instruments, lack of regulation and restraint, and short-term profit objectives; (2) quants and algorithms screwed up resulting in (1).
The latter argument is summed up in the NY Times: "Beware of geeks bearing formulas" begins the article with a quote from Warren Buffet. This is the same bugbear that attacked in the 80s with automatic trading; deploying technology results in unexpected outcomes.
Perhaps. And perhaps the current election is unduly influenced by the flattening of information resources in the internet-driven world.
But there are inevitable corrections to extremes that result in people losing money or power, and the technology will continue to be pervasive while the users of the technology will get smarter about its impact. On the political front, factcheck.org demonstrates how there is already an evolution from pre-internet rumor-driven political feelings to partisan exploitation of technology channels, and then on to sensible corrections to those partisan swings.
Perhaps. I just hope my son's 529 plan recovers enough in the next 8 years to make it better than just a wash.
Friday, September 19, 2008
The essence of the liberal outlook lies not in what opinions are held but in how they are held: instead of being held dogmatically, they are held tentatively, and with a consciousness that new evidence may at any moment lead to their abandonment. This is the way opinions are held in science, as opposed to the way in which they are held in theology.There are other results that seemingly bear this out, including Jonathan Haidt's findings that political conservatives simply value tradition and fairness at different levels from liberals. Preservation and stability trumps flexibility and risk.
Other recent interesting finds this week include Pyschiatric Times reporting that adult ADHD sufferers have lower educational and professional outcomes than non-ADHD individuals, even when IQ was held constant:
Adults with ADHD are not achieving the educational and occupational successes that they should be, noted researchers in the Journal of Clinical Psychiatry.
In a case control study, Dr. Joseph Biederman and colleagues looked at 222 adults with and 146 adults without ADHD to determine if educational and occupational functioning in ADHD represented low attainment or underattainment relative to expectations based on intellectual abilities.
In the control group, educational levels were significantly predicted by IQ scores, and, in turn, employment attainment was significantly predicted by educational levels. However, in the ADHD group, patients did not achieve successes as expected based on IQ and educational levels. In fact, only 50% of patients with ADHD were college graduates, yet based on IQs, 84% should have been. Similarly, only 50% achieved semi-professional or major professional levels, although 80% were expected to achieve such based on their education. Most importantly, the researchers noted, ADHD was associated with significantly decreased educational obtainment independent of IQ.
“These findings stress the critical importance of early identification and aggressive treatment of subjects with ADHD,” the researchers concluded. “Appropriate intervention could be highly beneficial in reducing the disparity between ability and attainment for individuals with ADHD.”
The take-away to me continues to support the notion that an evolutionary effective brain that trades-off risk aversion with creativity (and the kinds of transcendant and even randomizing cogitation that is essential to creativity) is in a wide valley of contributory genetic and environmental inputs that are easy to get just slightly wrong, whether we are looking at dysfunctional and excessive behavior among artists, interference with educational success for ADHD sufferers, or enhanced mental capabilities among borderline autistic individuals. The continued maintenance of this diversity of types suggests that the diversity is or was more adaptively useful than the obverse.
Finally, Marc Hauser works the landscape of moral decision making in a recent Newsweek article, once again describing how remarkably uniform a "moral grammar" we seem to share, regardless of ethnic background or political affiliation. In discussions among friends and family on this topic, I always come away with a more complicated picture of the moral dilemmas. How can you guarantee that dropping the fat man onto the railroad tracks will stop the train? How can you be certain additional help will not arrive before cutting the woman out of the cave mouth? How can you be certain that the death row inmate really committed the crime?
In every case, the trade-off is not between what is morally permissible and obligatory, but between the individual's level of certitude and the killing of one or many. I can therefore almost always answer the dilemmas with a refusal to act until the situation is so dire that action is required. Flipping the switch to divert the trolley car is the exceptional case that demonstrates pure utilitarian moral reasoning, but almost all others require persmissibilty to be modified to something like "permissible only given a lose lose situation where there is the strongest signal that other lives may be lost."
That kind of reasoning requires a low startle response, of course.
Thursday, September 4, 2008
The machine is impressive enough: a Sony VAIO FW (no, not the recalled model) with 4GB of RAM, 250GB of disk, and a Blu-ray device with HDMI output built-in. Although large, it was a concession to a round of deliberation about how I use computers. My old machine, a Gateway with a 13 inch screen and 1GB of RAM running XP, was simply too slow and lacked sufficient screen real estate for effective software development. With 10 Firefox windows open and a running Eclipse instance, things would start to drag and switching became ponderous. Part way through my deliberations over what to get, I seriously considered a Mac Air, which would have not met my requirements at all but was just so delicious I had to give it consideration. I thought briefly that the Air would work because I have five Linux servers hosted in a high-rise in San Jose, California and could use them remotely for my development needs. Almost--but not quite--due to networking speeds and the need to sometimes work offline. The larger Macs were also considered, although the price points to get serious bang were too steep. In the end, the VAIO was a good trade-off, with the Blu-ray add-on a concession I made to myself because I was not going for the high-end Mac.
And then the work began.
Luckily, I have cultivated a model of continuous holographic reflection of all work-related materials through the use of a source code control system called Subversion (SVN for short). In this model, every document, note, source file, image, etc. I create is checked-in to a repository hosted on one of my servers through an HTTPS connection and WebDAV. Change logs are maintained on the server, and periodic backups are created to other machines in the cluster as well as to a portable USB drive and, soon enough, to Blu-ray writeable media tossed in the trunk of a car.
So the first thing I did was install SVN on my new machine and check out everything to my local hard drive. Nice. But to get everything working took 4 days of software installations and configurations. I configured 10 different POP and IMAP accounts, PHP5, Apache HTTPD, MySQL, PHP5 plugins, Eclipse, Subclipse, ITunes, FabFORCE DB Designer, The GIMP, Microsoft Office, Microsoft Visual Studio 8, Adobe Flash CS3, Cygwin, Inkscape, Firefox, Propel, Xemacs, and many more. I updated everything to the latest versions and got automated and manual patches. I rebooted many times (wasn't that supposed to be fixed back when I worked on Windows 2000?)
All the complexities were smoothed out gradually and incrementally, of course, and I am fairly happy with the screen real estate and performance of the new machine after I disabled most of the security features of Vista. I even picked up the Blu-ray edition of Blade Runner, The Final Cut and ran HDMI to our LCD TV to confirm it all worked (note: no start-up lag unlike some BD console players; also, HDMI and LCD can't be running simultaneously due to HDMI-based DRM policies, which seems like ridiculous overkill).
But I wondered why I can't have a computing universe where the ease of the SVN management of my own resources was replicated in the software installation world? There is a hint of that capability in recent Linux installations that can download and install software packages and their dependencies with a single, short command. Still, configuration and customization remains daunting and can even be exacerbated because the installation process doesn't communicate all the details about where resources go (and the destinations change with some regularity).
Ideally we can imagine a computing cloud where apps are no longer installed locally, just web-based, and all of our configuration settings and password management is remote (and trustable) as well. Hints of that have been emerging with Java Web Start, Adobe AIR, Microsoft Silverlight and, to a lesser degree, Google Chrome. Each is an attempt to move web-based applications away from the limitations of a browser-based model and support more sophisticated interaction models. One company I work with has shown the model can work for specialized enterprise computing needs, so I think there is hope, but the evolution is nevertheless slow and may even require re-imagining the computing platform itself.
I'm guessing I have several dozen more gray laptops, virtual tablet devices, holographic mental interfaces and whatnot to go before everything becomes as neat and easy as I'd like, disconnected from individual platforms and universally available on-demand from some luminous computing cloud where replicants slave away maintaining and upgrading software in those tilt-up buildings in Silicon Valley.
Monday, August 25, 2008
Rawls' core principle is that given a "veil of ignorance" about the capacities, needs and external facts surrounding a decision-making event about resources, an individual will choose a fair distribution of resources since they can't be certain how the resources will ultimately be distributed. By being fair, the decision maker will maximize their own chance of a fair outcome.
This is a game theoretic resolution that parallels ideas like tit-for-tat and reciprocal altruism. It is also used to further justify a model of economic interventionism that assigns fair distribution of resources in a society. Hence the concern of Libertarians, though I see David Gordon's roll-up as an incremental and technical fight within the Libertarian community and about which I have little comment.
Instead, I find it interesting that if we acknowledge the implications of limited knowledge we arrive at inherently specialized models for decision making: in this case implications about the fairness of government policies concerning welfare or health care policy. This closely parallels the veil of ignorance expressed in Quine's notion of "gavagai" where we have to accept the limitations of strong notions of meaning and instead rely on decision procedures to achieve a tentative understanding of the immediate epistemological landscape.
One aspect that the purely philosophical treatment of this problem does not address, however, is how to handle the problem of procedural inefficiency. What do I mean by that? Well, we can take the standard critique of command economies as interfering with efficient allocation of goods and services. We can also take the apparent administrative inefficiencies of modern American health delivery. Both reflect differences in economic theories and have strong historical exemplars, but are polar opposites in delivery and informational mechanisms. If these are inefficient models compared to other examples, is there a principle of ignorance that can guide us in how to apportion resources in the face of these failures? For example, let's say that equality of health care is a good because if we could divide-up health care resources evenly and equitably it would result in that equality. Yet, if the mechanism of achieving equality also costs enough that it compromises other aspects of quality, what is the cost/benefit inflection point that recommends one approach versus another? We could seek to maximize some measure of person health outcomes/dollar using a basket of measures like those measured by the Commonwealth Fund's analyses of the United States' problematic health delivery system, but at what point do we agree that government is the best mechanism for delivery?
I tend to think that a principle of ignorance must remain in effect. Given uncertainty about efficiency in allocation, it is best to create a basket of solutions that try to achieve the same outcome. So, given the massive administrative costs of US health care delivery, we can create mandates for integrated billing and informatics for some domains while supporting tax incentives to improve delivery for other domains. There is then competition among outcomes in the basket of solutions and our ignorance is reduced through small, evolutionary outcomes.
Sunday, August 17, 2008
Still, while wandering through the National Gallery while on a quick hop into DC, and just coming down from being mesmerized by the Afghanistan collection, I sought out the Rauschenberg at right. RIP, Robert. With hope, the vaults of the nuclear enterprise in your collage are now just cold warrior memories.
The National Museum of Kabul collection was amazing, though, showing how Alexander's conquest came quickly to be reflected in the artistry. A different era, perhaps, where resistance to foreign occupiers was tempered by an acknowledgment that power is more important than fairness or respect for self-determination.
But truly sad was the gigantic gap in the timeline view of Afghan art and culture from the 1200s up through the discovery of graves during the Soviet occupation. That the analysis and remains stayed in Kabul from the Soviet era contrasts sharply with the later destruction of the Buddhist statues at the hands of the Taliban.
Sunday, July 20, 2008
The vast stillness of the valley, the faint X of airport runways, the thin green ribbon of the river and its fragile accompaniment of cottonwoods and salt cedar, the layering of atmosphere as the monsoonal cumulonimbus began to darken—all inspirational in the same way that they were when I was growing up in New Mexico.
But how does that translate into a crossover of science and art? My other data points were that Jaron Lanier grew up about a mile from me (I would later get to know his dad, Ellery, when Ellery was working on his doctorate in psychology), and that Alvy Ray Smith was also a southwesterner who, in turn, grew up just across the New Mexico/Texas border from Jim Clark of Silicon Graphics and Netscape fame. Then there is Robert R. Wilson who was a Manhattan Project physicist, sculptor and Fermilab architect. And recently, when visiting Southern New Mexico, I saw paintings by a bio-informatics professor and former member of The Institute for Genetic Research (J. Craig Ventner’s original effort) hanging in the lobby of a grocery store.
The tramway ride, it turned out, was part of a mini vacation within a vacation combined with work for DOD in the previous several days. It was my first time monitoring an experiment being conducted on live test subjects and it was fascinating, although somewhat tedious. Bookending the work phase with beautiful scenery, meals under evening thunderheads, historical markers along the Jornado del Muerto, and a quick flight back to the Bay Area, was invigorating and inspirational.
Thursday, June 26, 2008
In United States v. Miller, 307 U. S. 174, 179 (1939), we explained that...
Now this is interesting in that it paints a picture of unity in the way a court operates. A reasonable alternative might have been:
...the majority decision concluded that...Instead, Scalia is both time traveling and invoking a certain personification of SCOTUS wherein its often bicameral nature is subordinated to the body as a whole. Unity out of many, indeed.
Wednesday, June 25, 2008
So how should we treat fringe science and pseudoscience as a social phenomena? For instance, there is a fairly long history of investigations of supposed psychic phenomena dating back to the late 1800s, but with institutional investigation in the US beginning in the 1920s or so. Since that time, there have been many university laboratories created to try to study anomalous experimental findings. Projects at SRI in the 60s expanded the use of our tax dollars on this topic in pursuit of Cold War weapons.
As per the basic revulsion I felt at the treatment of the poor mother in the Canadian case, there certainly is not any evidence that suggests social policy should be influenced by psychic “discoveries” but there are enough findings that further research may be warranted (though funding that research should be a private—not public—matter). I say this with some trepidation, but I would say the same thing about cold fusion or other anomalous hints at findings.
But then what makes these two cases different from so-called “intelligent design” (ID)? For cold fusion, there is a proposed physical mechanism, for one, so that theory and experiment can progress together. For psi, there are some experimental results that look sufficiently interesting that additional experiments could and have been done. A mechanism is still wanting, however, though some ideas have been proposed based on virtual particle interactions that date to the beginning of the quantum science era.
Both of these are examples of what Irving Langmuir called “pathological science” and both were subject to enormous scrutiny and brought with them substantiated examples of dishonesty on the part of some (but not all) of the participants.
But ID is an odd pathological bird. It doesn’t really propose a physical mechanism but instead claims that there is insufficient evidence for a physical mechanism of evolution (though apparently only at the level of speciation). It also lacks experimental evidence (even in a historical sense) but uses a principle of insufficient information as the core of its methodology. It does share with cold fusion and psi powers the unenviable status of being practiced by at least some dishonest charlatans. In the case of ID, they have been actively trying to insert creationism into school curricula. Interestingly, there has never been a concerted effort by psi or cold fusion advocates to try to insert their ideas into the schools.
The odyssey from pathological to mainstream is not unknown (plate tectonics, for instance) but is only survived by relatively few. My guess is that ID will never make it through that filter and be welcomed to the lands of rationality.
Sunday, June 15, 2008
Yet slightly more intriguing to me was the use of language and titles in many of the more contemporary works at SFMOMA. Check out Robert Bechtle's irreverent Watsonville Olympia from 1977:
And anyone with a painting or art history degree (or, in my case, a wife with one) will recognize the inside joke (at Manet's expense). Here's Manet's Olympia:
What mountains an artist must climb. Happy Father's Day!
Friday, June 13, 2008
See Halliday & White, The Suspension Clause: English Text, Imperial Contexts, and American Implications, 94 Va. L. Rev. (forthcoming 2008) (hereinafter Halliday & White) (manuscript, at 11, online at http://papers.ssrn.com/sol3 /papers.cfm?abstract_id=1008252 (all Internet materials as visited June 9, 2008, and available in Clerk of Court’s case file) (noting that “conceptually the writ arose from a theory of power rather than a theory of liberty”)).
See History of Guantanamo Bay online at https://www.cnic.navy.mil/Guantanamo/AboutGTMO/gtmohistorygeneral/ gtmohistgeneral.
Times change, even at SCOTUS.
Saturday, May 31, 2008
There is nothing terrifically new here, though. Neil Postman decried television’s numbing and dumbing effect in “Amusing Ourselves to Death” in the 80s. If only kids had the time and inclination to read deeply they would be able to recall more facts about culturally importance events. If only they set down the cell phone and remote they would be able to recall the major players in World War II. If only they stopped text messaging they might grasp the compact beauty of poetry. If only…
I’m reminded of my first encounter with Goethe or Bellow or Pynchon, where the notion that writing could be more than just the transparent conveyance of a storyline first started to fixate in my youthful mind. I certainly hope that young people get to encounter the art of writing that is skillfully great, whether in prose, poetry or even information design.
I should also wave my hand over my copies of Claude Levi-Strauss, Jane Jacobs, Jung, Heinz Pagels, Chomsky, Jacob Bronowsky, Hillary Putnam, David Chalmers, Doug Hofstadter, Pinker, Dennett, and a myriad other popularizers and integrators of complex social and scientific ideas. There is equal artfulness in integrating disparate facts and ideas into a cohesive view that lends itself to the successive narration of chapter upon chapter.
But I would like to suggest that far too many of the latter are quaintly wrong or irrelevant today as science has unfolded further. They were derived from abduction over limited new facts to cover murky unknowns and were therefore largely doomed to factual attack as the murk was clarified by new experiments, observations and theory. Even their very narrative structure is subject to examination in that it forces the alignment of multiple research paradigms into a single narrative thread that always serves to reinforce the central thesis, regardless of how many other implications arise that fall outside the narrative flow.
In some ways it is the scientific research paper that serves as perhaps the best example of how ideas can be expressed with the maximum respect for the limitations of understanding because they force an exceptional economy on the narrative to avoid speculation, or at least to provide speculation without substituting narration for fully supported propositional constructs.
So there is nothing inherently dumbing about a diminished role for the long-form narrative (except perhaps in conveying the art of narrative itself). Whether a similar effect will save the text-messaging generation remains to be seen.
Friday, May 23, 2008
Three weeks back my wife and I went to a doctor’s appointment in
Weeks passed until I overheard the news of the arrest. Good for her! The next phase of datamining impressed me with the thoroughness of the picture. I was able to use the television station archives combined with Google to find the mug shots, the original sketch of the suspect done by police sketch artists, the suspect’s arrest status in the county courts system, the location of the alleged perpetrator’s house, the suspect’s father’s name and place of business, the suspect’s mother’s name and place of business, a previous citation of the suspect for a moving violation (infraction) in a neighboring city, the county records concerning the amount and type of mortgage held on the suspect’s home, and a satellite view of the home as well.
Amusingly, also, was that the reporter in the news piece actually drove by our house and coincidently filmed our various vehicles. I could likely have read the license plates if I wanted from the footage.
Overall, I had managed to scour out all the corners of ambiguity concerning when, where, who and how, leaving only the strange question of why left in fuzziness. Why was this 24-year-old still living at home, jogging at midday and preying on middle-aged women? Why was he living in this neighborhood where even a Megan’s Law offender is fairly hard to find?
But strangely, it was the suspect’s last name that was the key to developing the search picture because the last name was so unusual. Had he been “Jim Smith” or “Joe Sanchez” or “Mike White” it would have been virtually impossible to make as much headway in extinguishing ambiguity. George Miller is quoted something like “There is only one problem in Artificial Intelligence: words have more than one meaning.” (And I can’t resolve the ambiguity of the source of that quote because George Miller is too ambiguous). This problem is amplified for searching across identities of places and people, or when special identifiers are introduced as placeholders in a single document (this happens quite often in technical literature where an acronym is used locally as a technical shorthand but is ambiguous outside that document or domain). Moving to the level of folksonomies for, say, labeling pictures on the web, we see the problem exasperated by the natural telegraphic shorthand that any labeling scheme suggests to the user purely by dint of the size of the entry fields.
Clever approaches to trying to apply context to help address these limitations start with statistical co-occurrence-based disambiguation and linkage analysis, and then run all the way through to using complex ontologies to try to infer the best relabeling of the ambiguous entity or concept as a canonical identifier. None of these methods can hope to achieve any level of perfection but a basket of them can enhance the process of information discovery and disambiguation.
Thursday, May 8, 2008
American Journal Psychiatry 165: 560-561, May 2008
Some interesting follow-on questions for investigation: Would exposure to multi-ethnic faces early enough reduce the emotion discrimination deficit? Is there a learning window? Are there differences between subgroups (more integrated communities show lower deficits)? What is the impact on eyewitness subject identification across racial lines?
Recognizing Each Other and the Effects of Racial DifferencesCarl C. Bell, M.D.
In this issue of the Journal, Pinkham et al. report findings on own-race and other-race effects in their research on facial recognition in schizophrenia. They highlight the troubling finding in previous studies that African Americans with schizophrenia had greater impairment in recognizing and remembering emotions in faces, compared with Caucasian schizophrenia patients. The authors point out that these previous studies showed only Caucasian faces as test stimuli—a major procedural flaw that no one had anticipated. Pinkham et al. designed a better study in which test subjects—both Caucasians and African Americans with schizophrenia as well as comparison subjects of both races—were shown both Caucasian and African American faces as test stimuli. The authors found that the capacity to recognize and remember emotions in faces is no different between Caucasians and African Americans, whether they have schizophrenia or not. But memory for the faces and discrimination of emotions in them are higher when the study subject is of the same race as the person expressing the facial emotion.
Saturday, April 26, 2008
Still, I consider it a vindication of sorts, though I think they will have a hard road business-wise if they believe twine.com has a massive audience appeal. I will just mention that I interviewed with a startup called Backflip in 2000 that went nowhere as well. The business proposition for these kinds of technologies is when they provide real business value to specific knowledge worker communities--when they serve vertical domains to address specific problems.
A minor footnote: I chatted with Radar Network's Nova Spivak several times about collaboration several years back and suspect that there was a bit of, ehhhh, influence of Ofamind on Twine's development (based also on watching my weblogs). I'm just amused that I beat them to the plate with one consultant and three other ongoing engagements simultaneously. Are there diminishing returns beyond a few agile minds in what Bessemer VCs refer to as our new capital efficient web marketplace?
Thursday, April 24, 2008
But are folksonomies useful and consistent? Some studies suggest they are useful under some circumstances. For instance, querying across the titles and descriptions using tag keywords on del.icio.us bookmarks results in a precision-recall of only 50%. In other words, the tags are not also in the texts around 50% of the time, and so provide an additional channel of information for retrieval. People appear to think differently about tags than they do about titles and descriptions.
In terms of consistency, however, a very large number of tags are used only once or are used in differing and inconsistent ways that indicate ambiguity over multiple user subcommunities. Examples might be “architecture” used to refer to computer architecture and building design, or “camp” referring to drama or outdoor recreation.
A couple of interesting questions emerge about how to refine the power of folksonomies.
For instance, can the title and description (or full blog content) be used to automatically suggest tags that are based on other tagging schemes? The Ofamind system partially does this by automatically categorizing web content among your “views” or collections as you surf. It does a fair job, too, for a great deal of content. This can be seen as a personalized metadata tagging filter, since the view association to content is essentially a categorical tag.
Similarly, business taxonomies, controlled vocabularies, full ontologies and other mechanisms could be used at authoring time to try to suggest or overlay more consistent tags onto web content, enhancing searchability and even supporting reasoning about content. For Ofamind, a subproblem that we are currently working on is how to disambiguate extracted people, places and organizations in order to produce high-quality metadata using a combination of human tagging and automatic methods.
Then the folksonomy becomes more of a folksontamasticon, combining folksonomy, ontology and onamasticon in a rare new tag.
Thursday, April 3, 2008
“Girls always grow up to marry someone like their father.” “You just don’t understand what it’s like to grow up as a redhead.” “My father hates me because I had medical problems as a child.”
Now, I try to be sympathetic of even bad self-analysis insofar as I only have to encounter it in small doses. What good can come from challenging people under those circumstances? But I increasingly wondered as the conversation continued whether there are limits to our natural capacity to overcome the patterns of folk psychology we use to attach meaning and explanation to our lives. I tried lightly challenging one of the statements and suggested that there was little or no evidence to support a given claim. Her response was that she just gave me evidence—a single example. There was some hemming and hawing about how she acknowledged that that wasn’t scientific evidence, but it was enough for her.
Isabel Allende was recently interviewed and she confessed that her entire writing method and inspiration emerged from trying to construct narratives and folk psychologies to explain her characters. Everything had an explanation and there was little difference for her between magical, religious and everyday occurrences—they all had elaborate explanatory narratives that involved mystical forces, and frameworks for punishment, reward and retribution. Magical Realist at core, but also reflecting the need for fiction to tie together into a structural form that is without the weak sense of doubt that pervades our everyday lives; people are complex and do things for complex and sometimes unexplainable reasons best regarded as tendencies.
But is there any sense in which people can change their cognitive styles? Somewhat, I think, but there are also other factors like dopamine and it’s relationship to magical thinking that are likely more resistant to active attempts at change. Still, the goal of liberal education has always had at its core the notion of refining the mind to enhance our ability to think and process information. It’s the best tool we have.
Tuesday, March 25, 2008
In his summary defense in the Boston Massacre trial, he claimed that the British soldiers had every reason to be afraid of the crowd, "a motley rabble of saucy boys, Negroes and mulattoes, Irish teagues [pigs], and outlandish Jack Tars."And here:
In fact, one of the causes of the revolution was the Quebec Act, which gave religious protections to Catholics in Canada. This infuriated the colonists. "Does not your blood run cold to think that an English Parliament should pass an Act for the establishment of arbitrary power and Popery in such an extensive country?" wrote Alexander Hamilton. "Your loves, your property, your religion are all at stake." Sam Adams told a group of Mohawk Indians that the law would mean that "some of your children may be induced instead of worshipping the only true God, to pay his dues to images made with their own hands." Fortunately, George Washington realized that it would undermine the colonists' efforts to win support from Canada and France if they were perceived as being anti-Catholic, so he banned the "monstrous" practice of burning effigies of the pope on "Pope Day."
Thursday, March 6, 2008
Still, I remain a student of the general theme and so am intrigued when people like David Sloan Wilson characterize the role of religion in social cohesion as providing unique evolutionary advantage at the group level (Darwin’s Cathedral: Evolution, Religion and the Nature of Society). While arguing about group dynamics in evolutionary circles is somewhat heretical,
From some cross-over of libertarian and paleo-conservative thinking, comes another intriguing data point from The American Conservative. Yes, that’s Pat Buchanan’s magazine. Pat regularly brings up the notion of “balkanization” both in reference to the Balkans as an example of a geopolitical mess, and as a broader metaphor for the problem of diversity in modern societies, so it is not surprising to see his magazine latching onto Robert Putnam’s discourse on changing American civic involvement (Bowling Alone) and related research on the potential drivers for civic strife.
The TAC article by Steve Sailer is somewhat fragmented, jumping around through some sloppy generalizations about ethnic identities (Hispanics and Italians don’t build large organizations because they only trust extended families, for instance), and dipping into wag-the-dog-style political fervors driven by a common enemy. It ends with some minor discussion of how both religion and mandates can improve cooperation between people, with the latter example being the Army limiting career advancement among officers who discriminate.
Now I suspect Sailer and Buchanan consider this grist for a policy mill that aims to reduce immigration to the US (or perhaps be simply more selective about it), but in some ways it works against their more cherished cause of small government and limited government because, given an uncooperative, pluralistic and diverse population, one remaining channel to achieve grander visions is through government action. Government and law become the conduits for coordination by transforming distrust in others into (perhaps grudging) acceptance of institutions. Even Sailer admits as much in noting that:
So I suspect that we have achieved a stalemate of sorts, with the benefits of diversity (I’ll just start with my restaurant options today and leave it at that…) balanced against less social cohesion, but perhaps propped up by institutions that are trusted enough that we are not always suspicious of corrupt abuses of power. That seems like a gentle enough substitution for a civil religion to me, with a more subtle organizing physics that preserves the freedom to think outside the confines of any monolithic pattern of ideation.
Sunday, February 24, 2008
It was the triumphalist laissez-faire notions that celebrated industrialists under the banner of Social Darwinism that I found the most interesting, since there remains a core of this in contemporary Republican/Conservative reasoning. Indeed, paleoconservatives like Pat Buchanan routinely revive a weakened form of these notions when opposing social programs like affirmative action and justifying positions where individual and racial differences are not correctable by public policy.
And therein is the irony, I suppose, that Social Darwinism informs some of the most anti-Darwinist modern thinkers. Competition is good for you. Still, the most telling development to emerge from the same era is that ideas have a certain prolific vitality, mutating outward from basic scientific observations into new formulations that reengineer the scientific core into a series of quasi-scientific speculations. Then, over time, the speculations are whittled down and even completely discarded as the pseudoscience is ejected and only the core remains.
Tuesday, February 12, 2008
Environmental Factors in Schizophrenia Three epidemiological studies bolster the evidence for infectious and social influences on the development of schizophrenia. Swedish national registers show an association between psychotic illness and childhood viral, but not bacterial, infections of the central nervous system (CNS). Dalman et al. (p. 59) analyzed hospitalizations for CNS infections before age 13 and psychotic illnesses from age 14 onward in children born during 1973–1985. Psychosis risk was almost tripled by childhood mumps exposure and was over 16 times as high after cytomegalovirus exposure. Using blood samples collected routinely by the U.S. military, Niebuhr et al. (p. 99) confirmed a relationship between schizophrenia and toxoplasmosis. IgG antibodies to Toxoplasma gondii were compared in service members medically discharged with schizophrenia between 1992 and 2001 and matched healthy subjects. The antibody level was nearly 25% higher for the subjects with schizophrenia in the 6 months preceding the diagnosis or after it. Dr. Alan Brown examines these two studies in an editorial on p. 7. Veling et al. (p. 66) identified social isolation as a risk factor for psychotic disorders among immigrants in The Hague. City records provided the ethnic backgrounds and locations of residents who received a first diagnosis of psychotic disorder over 7 years. Immigrants in neighborhoods with high densities of immigrants from the same country had a rate of psychotic illness similar to that of native Dutch residents, but those in neighborhoods with low densities of the same ethnic group had a rate more than double that for the Dutch.
Monday, February 11, 2008
Another area that has rapidly advanced and structurally changed the way of doing business is the growth of cheap computing power and even cheaper software to tackle web and enterprise software challenges. Open source software in the form of MySQL and Linux effectively demolished the barriers to building web-grade technology stacks. Ten years ago, expensive Sun server farms and Oracle licenses were a prerequisite to doing business on the web, but now we can do it comparatively cheaply.
Still, there remained until recently one final barrier for large-scale web businesses: scaling from thousands to millions of users with huge bandwidth and content needs. That is changing too, though, with cloud computing efforts like Amazon’s EC2 and related projects. I’ve just embarked on a brand new effort that has the potential to reach huge audiences and am seriously looking at the Amazon model because it means that I no longer have to deal with racks of hardware and complex and expensive colocation plans.
In the cloud computing universe, you buy as much as you need and can buy more as your demand levels rise without having long hardware deployment and software imaging cycles. And thanks to the massive server farms, the capacity comes with certain guarantees concerning reliability. A remarkable feature of cloud computing is the ability to snapshot a given virtual computing instance and create new copies as needed to expand capacity.
The end result, though, is that the combination of cloud computing and open source software has flattened the world of technology to the point where creating world-class web technologies is purely a matter of brainpower and business models.
Thursday, January 31, 2008
Senator DeMint was described as the most conservative member of Congress and delivered on that description in his interview. After a short intro on his endorsement of Mitt Romney and a discussion of the currently pending economic stimulus package, DeMint sketched his thesis: anyone who expresses religious values is oppressed by the government; that the reduction in religiosity in America since the 60s has been accompanied by the erosion of the greatness of our society; that the government promotes immorality through various policies; and that the “Democrat Party” was in opposition to good values (he sort of apologized when someone pointed out that “Democrat Party” was a slur and that the party is the “Democratic Party,” though he said that we are a democratic nation and he didn’t associate that with Democrats, thus compounding his slur while claiming he would try to do better)
Now, I can safely say that Jim DeMint has some serious problems with facts, categories and logical argumentation. But that’s to be expected. For instance, he floats the case about a parent in Massachusetts who was incensed that same-sex relationships were mentioned in his child’s school. He then claimed that the parent was arrested for speaking out about this “immoral” education, when in reality (as Diane Rehm’s stand-in from the BBC noted) the parent was arrested for refusing to leave the school premises. DeMint continued on a strange quest to conflate the notion that free speech was being thwarted when people were unable to change policy, as opposed to the commonsense notion of free speech having to do with speech acts, written or verbal.
Many callers questioned his statistical claims concerning a range of issues, as well. When pressed on why he claimed abstinence education was being blocked by evil secular government forces, for instance, rather than the recent discovery that they didn’t much work, he claimed that the Heritage Foundation had statistics that showed that if you combined abstinence education with a commitment on the part of parents to furthering the abstinence cause, that then you might get a 75% reduction in out-of-wedlock pregnancies. Fair enough, but he is completely redefining the problem. I could achieve the same results with secular humanist education: if you get pregnant before you are ready, it will have negative consequences to you, your partner, your parents and society. Don’t do it. Given good basic values of respect for oneself and others, care for others, and belief that harming others is wrong, you get precisely the same result.
DeMint is easily criticized and I didn’t come away with a deep respect for his intellect or his willingness to stretch intellectually to fathom the complex issues. Indeed, he was so wedded to his theses that I felt queasy. But there was one topic that he mentioned that was somewhat interesting to me: does government, though its policies, serve to direct, encourage or sanction moral behavior? DeMint suggested that our values have changed regarding smoking based on government action. A similar argument applies to seatbelts or talking on cell phones while driving.
The odd thing to me is how an arch-conservative seems to have adopted a nanny-state attitude that also reveals a divide between liberals and conservatives. Jonathan Haidt’s research shows that liberals and conservatives differ on a couple of measures, with conservatives valuing “reverence” more than liberals, and liberals value “fairness” more than conservatives. Now an easy application of this principle shows the divide: DeMint castigates homosexuality as immoral while liberals don’t care what others do with their lives and think they should be treated fairly. But for DeMint, government has been shaping those attitudes by failing to criticize the gay community for (in his analysis) the cost to society from AIDS, STDs and other alleged problems. Gay marriage, oddly, remains a problem for DeMint despite the claim that it might lead to longer-term relationships and reduce exactly the issues that he is concerned over. So DeMint wants a nanny state for personal morality, but also wants a small government that is less nanniful about business regulation, guns and the environment.
A strange collection of virtues and values that shows how the 1800s live-on in the backwaters of America.
Tuesday, January 22, 2008
What amuses me about this is that just how to do that pairing, or how to evaluate wine quality, is not just a mystery, but is a profound black hole of irrationality and subtle psychology. If determining wine quality was rational, we would expect consistency among experts. We would expect predictability in the Wine Spectator scorings. There isn’t. We would not expect to find that people buy wines with animals on the bottle more than other wines, or that you can label the same wine at $40/bottle and at $3/bottle and get remarkably different reviews from tasters (the $40 version is remarkably better!)
In the end, though, no one really knows anything beyond some simple rules about how to combine wine and food flavors together. Sweeter wines with spicier foods. More robust meats with richer and darker reds. A sommelier and a tasting menu makes much better sense than any of us wine spectators.
Sitting at that table, one of my consultants managed to dig out an 1899 Rothchild Pauillac from the tablet computer. Price tag would have been $8999.00. I nixed the idea. I promised him that someday, maybe, we might get it in a fit of irrationality.
Sunday, January 13, 2008
A brief description may help.
Some Israeli researchers following a Kabbalah-like speculation about hidden codes in the Hebrew OT looked for words and word relationships hidden in the characters. The way they thought the words were hidden was as what they called equidistant letter sequences (ELSs). An ELS is where each letter of a word is separated in the text by a certain number of other characters. When they found an ELS, they then looked in the immediate area of the text around the words for other ELS sequences that said something interesting about the original word. They paired these together as questions and answers.
Needless to say, it is pretty easy to take any text, find interesting short words as ELSs and then find interesting words as ELSs around them. With my original code, I used the system to decide what to have for lunch. I would find the word “lunch” as an ELS, then look around and get words like “taco,” “steak” and “fish.” As one can imagine, shorter words tend to have greater representation!
Here's some examples borrowed from Brendan McKay showing ELS patterns predicting the assassination of Martin Luther King, Jr. in the text of Moby Dick:
More are here.
The whole episode demonstrates a surprising vitality to craziness impregnated by religious fervor, considering the start in 1994, the analyses and counter-analyses from various fronts, the publication of a best-selling book, and the availability of commercial systems that help you now do your own bible code analyses.
A slightly less crazy speculation comes from Clifford Pickover (although it's not clear what the original source is) that given a (countably) infinite digit sequence for PI, Shakespeare must be inevitability coded with those numbers given a suitable representation scheme. While that might be, ELSs in PI would be rarer than in human language texts, I think, because the digit probabilities are very uniform for PI after a few thousand digits, unlike most languages that tend to have a more skew distribution of letters. So ELS words would be way down deep in the code, though not quite as far along as The Bard.
Tuesday, January 1, 2008
The moral dimension of decision making ended up pervading my holidays, with a perhaps only nominally healthy degree of worrying over the whys of many otherwise pedestrian considerations. I upset my nine-year-old as we drove by the Shaklee building in
A related concern arose in a commentary I made to the local paper: should we lie to children? Christopher Hitchens and Richard Dawkins argue absolutely no, and for the most part my wife and I have done very well with our own variation on this position. We dropped the ball a few years back, though, because of suburban parental expectations and the fear that our son would be castigated by his friends (as much if not more than he already has been by those who assert Hell is in his future!). What we did was to vaguely allow the Santa Claus myth to continue. The results two years back were perhaps predictable and, to this day, he reminds us that we cruelly deceived him and made him look like a fool. Never again.
So, when the inevitable “Yes,
There are many things that The
Virginia, we want you to believe that we have your best interests at heart, and therefore we can’t lie to you about these things. Opinions differ about many of these ideas. Some people think that if we deceive you it helps you have a more interesting childhood. Some people think there are many gods swirling around in space. Some believe that forests are filled with wood spirits inhabiting the trees. But all evidence points to quite the opposite of all of these ideas.
Still, giving gifts of cheer, celebrating our lives together, and being kind to one another are all beautiful and very human ideas. We won’t lie to you, Virginia, but we do want you to be kind and enjoy the holidays. Be of good cheer, and the spirit of Santa Claus will always be with you.
The New Editorial Staff
But this is an example of a basic moral revulsion (do not lie to children) interfering with social and rational ideas. In any case, my son at least approves.