Tuesday, September 20, 2011

Gossip, CMC, and Tight Knit Communities


Published: September 19, 2011
As more people share gossip over the Internet rather than over coffee and eggs, anonymous, and startlingly negative, posts have provoked fights, divorce and worse.

Sunday, September 11, 2011

Good eye/mind catches logical fallacy in WSJ Analysis

Jean Whit notes that the authors of this piece about Wall Street Journal number crunching about sovereign debt default and bond ratings, WSJ Analysis: Rating Firms Not Effective at Predicting Government Defaults, got their analysis backwards. A classic case of sampling on the dependent variable or percentaging in the wrong direction: how many of the defaults had a given rating rather than how many with a given rating end up defaulting. See Jean's comment at bottom of post.

GPS, Orwell, and the 4th Amendment

The 9/11 anniversary reminds us, among all the other things, of the questions of government surveillance that have arisen in the last decade, some related to terrorism, some reflecting challenges raised by new technologies, and many at the intersection of these.

This fall, the US Supreme Court will consider whether law enforcement should be able to attach a GPS tracking device on a vehicle without a warrant. Adam Liptak reports on the issue in "Court Case Asks if ‘Big Brother’ Is Spelled GPS" in today's New York Times. Lower courts have ruled in different directions on the question.

One way to think about it is in terms of aggregating information and whether there's an emergent property that changes how we would classify obtaining, possessing, or using information. Consider, for example, one's daily round. Leave the house at 7:30, stop for coffee, pick up the dry-cleaning, get stuck in traffic, arrive at work, park in the lot over behind the pine trees, etc. All of these are done in public with no expectation of privacy. And then it all happens again tomorrow, and tomorrow and tomorrow. Except the dry cleaning stop is only made on Mondays and every other Friday there's a stop at a bar on the edge of downtown. If there is a GPS attached to your car, the separate public facts of any given daily round -- the sequence and full set of which perhaps only you know -- are assembled as a unit of information. And, if the GPS is there for a month, both the overall, boring, day-in-day-out pattern and the regular exceptions and the truly unique exceptions are all a part of the information bundle available "out there."

 Even if all of the component information is about mundane, innocent, non-embarrassing activities, indeed has all the properties that would exclude it from your understanding of "private" information, does your willingness to do these things in public view aggregate to willingness for information about them to be aggregated into a tracking record?

See also
UNITED STATES v. GARCIA No. 06-2741.
New York Times. Articles on Surveillance of Citizens by Government
New York Times. Articles on Global Positioning System

Saturday, September 10, 2011

From Musical Consonance to Styles of Thought

An article published in Physical Review Letters, reported on in Science News, describes a mathematical model of how neurons can distinguish consonant sounds (say, a C-major chord) from dissonant ones (say, D-E-F). A simple network of neurons, behaving like neurons behave, produces qualitatively different outputs depending on the quantitative differences in the sound frequencies it receives as inputs.

Very interesting as an example of an information processing system with emergent information processing capacities.

I suspect something, at least metaphorically, similar might go on in the processing/experience of consonant ideas. At first I'm tempted to say "least that part of consonant or resonant ideas that we want to ascribe to consonance in the external world" but I think you could take it further and imagine the development of structures along similar lines for the detection of "constructed" consonance. Eventually, one could arrive at mechanisms for implementing "styles of thought" that would not be limited to algorithmic systems that "crank through a set of data" in the same way every time. Rather, we could talk about styles of thought in terms of the kinds of thoughts, tropes, logics, metaphors that would appeal as consonant with "everything else I believe." Or, the flip side of this would be to move toward mechanisms for cognitive dissonance.

 Just a highly speculative bit of musing, but clearly news of this research did strike a chord with some stuff I've been thinking about for a long time.

Sunday, September 04, 2011

Sociology of Information in the New York Times


Published: September 3, 2011
Why all the sharp swings in the stock market? To Robert J. Shiller, it’s a case of investors trying to guess what other investors are thinking....
Seeking not what is the case, but what others probably think is, or even what others think that others think is...
Published: September 2, 2011
When Rick Perry, the governor of Texas and a presidential hopeful, debates his rivals, his assertions on climate change, Social Security and health care could put him to the test....
Once it's out there, it's out there...
Published: August 29, 2011
The antisecrecy organization WikiLeaks published nearly 134,000 diplomatic cables, including many that name confidential sources....
Developing story -- a leak, a revelation, or just a mistake?  (See also previous posts on Wikileaks.)

Bloomberg Contra Notification Norms

Consider a recent NYT article by Mosi Secret and Michael Barbaro about the controversy over New York mayor Michael Bloomberg's failure to inform the public about the actual reason -- an arrest in Washington, D.C. for domestic violence* -- deputy mayor Stephen Goldsmith resigned this summer.

 According to the article, Bloomberg "rejected the notion that he had an obligation to tell the public of the arrest." He is quote saying, “I always assumed it would come out, but it’s not my responsibility.”

 It's a first rate example of notification in the public sphere and of how overlapping relational circles can suggest contradictory notification rules.

 It turns out that it's not just a notification issue, though. Initially the mayor said the resignation was to pursue other opportunities -- in other words, he was pretty explicitly misleading not just failing to reveal.

 But back to notification. Bloomberg, apparently, takes the line is that his first obligation was instrumental, making sure "he no longer works for the city." And then his next obligation is to treat Goldsmith and his family with respect. His critics suggest that his first obligation is to the public, although the one quoted in the article, Scott M. Stringer, the Manhattan borough president, sticks with the instrumental saying Bloomberg has responsibility to "protect the public, not to protect a staff member” according to the article. But nobody really seems to be saying that there was any instrumental damage done by the non-notification and it's a bit disingenuous to say that getting Goldsmith off the city payroll was facilitated by non-notification.

 The issue, then, is how the various relational imperatives governing who ought to tell whom what when and how interact. New York City law, as it happens, has something to say: "the city’s Department of Investigation must be notified" if an official is arrested in the city (not clear by whom), but this did not come into play here since arrest was in DC. The article reports a debate within the mayor's team about the matter with the mayor saying that Goldsmith should get to decide how much to reveal. And after the fact Goldsmith, who took some heat for not indicating in his resignation announcement what the reason was, has "admitted" that HE had a responsibility to be more forthcoming at the time, though he added that he thought that immediacy of his resignation "mooted the need for further explanation."

So, does the mayor's official role and its informational obligations trump the social obligation to allow another "ownership" of his own announcement?   Does the consequential outcome -- resignation -- obviate the obligation to notify (for the record, Goldsmith says he was wrong on that count). If there is public outrage based only/mainly on the relational expectation of "we should have been told," does it support sanctions? Does Bloomberg's citation of a norm that certain personal situations are one's own to disclose get him off the hook? Does affirmatively suggesting other reasons rather than simply failing to disclose the actual ones cross another line entirely?

 Bottom line: in many locations within the social, institutional, moral orders, the import of information behaviors goes far beyond the instrumental, consequential, substantive realm.

 * The case is not being pursued as Mr. Goldsmith's wife dropped the complaint.