Vital Systems Security

An article in today’s New York Time’s announces the release of Google Flu Trends, a tool developed by the philanthropy wing Google.org. Google Flu Trends tracks a specified set of search terms for “ebbs and flows”, broken down geographically by regions and states. Monitored search terms include “flu symptoms” and “muscle aches” (A complete list of monitored terms does not seem to be publicly available, perhaps because of the possibility of manipulation). Read the rest of this entry »

A fishy effort is underway by the LAPD’s counter-terrorism unit to map Muslim communities in the city. Drawing from census data, which does not capture religious affiliation of individuals or household, and other demographic information, the men and women in blue hope to be able to extrapolate the relevant information. It’s a curious VSS issue not least of which because of the interesting tactic police officials have begun taking in understanding and characterizing potential threats at the local level. Quoting from today’s LA Times article on the subject:

In outlining the program last week before a congressional committee, Deputy Police Chief Michael P. Downing, who heads the counter-terrorism operation, said the department’s plan was designed to minimize the radicalization of Muslims in Los Angeles. Instead of relying on experts, he said, the mapping would produce a “richer picture” of the community and guide future strategies.

“While this project will lay out geographic locations of many different Muslim populations around Los Angeles, we also intend to take a deeper look at their history, demographics, language, culture, ethnic breakdown, socioeconomic status and social interactions,” he said. “It is also our hope to identify communities, within the larger Muslim community which may be susceptible to violent ideologically based extremism and then use a full spectrum approach guided by intelligence-led strategy.”

So, in other words, poverty and isolation, coupled with more conservatively oriented mosques and a population under increasing scrutiny, should be able to produce a picture of the prototypical threatening Muslim - no doubt viewed as a kind of “vulnerability” within the larger societal fabric. The chief of police of the LAPD, William J. Bratton, had this to say in response to critics: “This is not . . . targeting or profiling… It is an effort to understand communities.” Hmm. I have absolutely no objection to law enforcement incorporating demographic and census data at the population level into their purview and techniques — I think it’s rather a long time in coming and quite necessary. However, I do object to couching the obvious targeting of one religious community in these terms, utilizing these techniques. Seems to me that a better way to gain a kind of seamless integration with - and entree to - the Muslim community would be to actually reach out to members of those communities, consistently, actively and visibly, along with every other (non-Muslim) community! Talk about a good dual use strategy!!

* * UPDATE - - NOVEMBER 15, 2007 * *

The LAPD just decided to cancel the surveillance program. My analysis was whacky, but the program — now DOA — was odious. Have a look at: http://www.latimes.com/news/local/la-me-muslim15nov15,1,7941397.story?coll=la-headlines-california&ctrack=1&cset=true

Another troubling disease event in China. What is interesting, from an event detection perspective, is the disjuncture between the Chinese government’s official description of the event (165,000 pigs infected with a common pig virus) and international estimates. Because of China’s reluctance to release data, international officials are basing estimates not on health-related information, but on economic data–skyrocketing pork prices. “In part, the skepticism comes from the fact that pork prices have skyrocketed 85 percent in the last year — an increase that, absent other factors, suggests the losses from disease are more widespread than Beijing admits.”

Article below.

Read the rest of this entry »

I ran rather accidentally into two further examples of attempts to create what might be called “immanent surveillance systems” — in other words, systems that are engineered to produce information about themselves as they function, or in which the parts are self-surveilling. Both are Berkeley based. One is a project by a graduate student named R.J. Honicky in the computer science department. His dissertation project, as he describes it, is to build “a societal scale, distributed scientific instrument by integrating evironmental sensors (such as carbon monoxide) into location aware cell phones.” The basic idea is that sensors on cell phones would record environmental data and send it back via SMS (along with a geographic marker) to a centralized database. His thinking lies at an interesting conjuncture of surveillance and data management, on the one hand, and what he calls “participatory urbanism” on the other.

The other morsel was from Kris Pister, a professor of electrical engineering working on dust sensors. Here is a rather old article on the concept of using dust sensors to create energy-aware buildings. Pister also consults with a company called Dust Networks that is a (self-proclaimed, anyway) “leader in the wireless sensor networking market.” The applications include energy efficiency and, as one might imagine, defense and security.

Paul drew my attention to whoissick.org, a site in which individuals are supposed to report their own symptoms and it provides analytical tools to break down their incidence, spatial concentration, and so on. Browsing through different areas, it does not yet seem very densely populated. (My own Lower East Side, one of the more densely settled areas in the United States, only shows a smattering of “reports.”) The best thing: you can receive outbreak alerts by email. Also, check out the symptom tag cloud on the lower right. This is our web 2.0 world.