Search and Destroy

Friday, July 18, 2008

Transcript

The ability to search through massive amounts of data, Google-style, is having far-reaching effects. And, according to Wired Magazine's Chris Anderson, one of the most significant casualties may be the venerable scientific method. He explains why in the age of the petabyte, scientific testing is forever changed and why the numbers now speak for themselves.

Comments [8]

Paul G. Conroy from Boston, MA

Regarding: The End of Theory, By Chris Anderson

Chris Anderson makes the statement in “The End of Theory” that “The scientific method is built around testable hypothesis.” While this may be the general perception, observation, not theory, is and always has been the foundation of the scientific method.

The scientific method was originally a generalized description of the essence of natural science, ie., knowledge arrived at by inductive reasoning from observations. This distinguished natural science, earlier known as natural philosophy, from religion and philosophies based on a priori knowledge. With this understanding we can see there is little issue with the recent explosion of data and the use of observation and inductive reasoning to acquire knowledge; for this is after all the essence of the scientific method.

Physical laws are derived from observations without benefit of the ‘hypothesis, experiment, theory’ bit usually added to observation as steps in the scientific method. Theoretical scientists use hypotheses, a form of a priori reasoning, and experiments to arrive at theories. Theoretical scientists have shown real contributions to scientific knowledge, however, the bulk of scientific knowledge is acquired by observation and using inductive reason to formulate and refine physical laws without benefit of hypothesis, experiment and theory.

Paul G. Conroy
Chemist

Jul. 23 2008 09:31 PM
Philip Davis from Ithaca NY

Re: Interview of Chris Anderson of Wired Magazine

Ice cream sales are strongly correlated with drownings, murder, shark attacks, boating and lawnmower accidents. Does that mean that ice cream causes these events? In the absence of theory, we would flippantly believe that these associations are causally based.

Its frankly hard to believe that Google researchers are atheoretical. They hire some of the greatest minds — minds that would be wasted if all they did was look for simple relationships in the terabytes of data they have available.

Without theory, we’d be adrift in a sea of meaningless data.

see comments on the Wired article at:
http://scholarlykitchen.sspnet.org/2008/07/14/will-data-beat-hypotheses/

Jul. 21 2008 01:42 PM
Rick Evans from Massachusetts

Chris Anderson doesn’t seem to understand how science works. Data has always led in science. Science starts when data is observed, classified and correlated. Models are developed to fit data and make predictions. New theories are developed when earlier theories weaken in the face of more precise data.
Craig Venter's genomic survey of the ocean, doesn’t overthrow the traditional view of science. What he is doing is finding the DNA of unknown life forms and predicting their existence.

By analogy astrophysicists have used gravity to infer the existence of unseen masses. The existence of dark matter inferred this way. Using massive quantities of data to infer the existence of indirectly observed entities has been a norm in astronomy for decades. What Chris Anderson confuses is increased processing power with how science is done. The massive data combined with greater processing power is just a new tool like supercomputing or MRI.

All that has happened in Anderson's Google world is that scientists will have a lot more data from which to form hypotheses and formulate theories to test and better tools to process the data. Whether it's experimental science like chemistry or historical science such as archeology a prediction confirming observation is the ultimate test.

Jul. 20 2008 11:46 AM
Steve K from Portland, OR

Once again, Wired Magazine and Chris Anderson stake everything on
[1] their belief in unending technological progress (and energy resources);
[2] their belief in the ability of technology to address and resolve all human (and environmental) problems;
and worst of all,
[3] their belief that our current society and economy represents the end of history, the ultimate and natural human order -- a social order which is free and fair, where every individual is equally free and capable, with equal access to opportunity, and equal access to accurate knowledge, and propelled solely by self-interest.

Accordingly, society & economy are one and the same. And the majority -- as expressed by demand, or Google rankings -- is always correct. Thus, who needs science when you can just use data mining to poll 'the market'?! (And, therefore, why be bothered with things like hypotheses, or theories, or the difference between the two!!)

Fundamentalist religion is not the only ideological threat to science today. Libertarianism -- a fundamentalism combining at least 3 myths: individualism, free-market capitalism & inevitable progress -- also poses a serious threat to science.

Libertarians like Chris Anderson are just trying to deploy their own ideologically-based "rules of nature" to create their own dominant [pseudo-]scientific paradigm through which they will produce "scientific" reports.

And what is the shining example of this 'new science'? Freakonomics, or "behavioral economics".

Jul. 20 2008 11:22 AM
Jeremy Yoder from Moscow, ID

I was glad to hear Chris Anderson recant that ridiculous headline - but it sounds like he still doesn't really understand how science works. Data collection always has preceeded hypothesis generation and testing (which eventually leads to new theories). All the "statistical analysis" that Anderson is excited about is actually either collecting and summarizing data, or automated hypothesis testing. The Petabyte age is truly exciting - but good, old fashioned scientific methods are going to be necessary to make sense of the flood of data.

http://denimandtweed.blogspot.com/2008/06/wired-drinks-science-20-kool-aid.html

Jul. 20 2008 02:29 AM
Jim Anderson from Maryland

I agree. Fight back, don't take any crap. If we had used that tactic four and eight years ago, who knows how many things would be different and how many of our troops would still be alive, not to mention the countless Iraqis and other "miscellaneous casualties."

Jul. 19 2008 06:19 PM
Donald Miller, Ph.D. from Salt Lake

Chris Anderson's thesis was entertaining but specious.

As you raise the sample size of two statistical populations from 0
towards infinity, you raise the absolute value of the correlation
coefficient from 0 towards 1.

Observation leads to art which precedes theory, which is always subject
to challenge by other data and theories. Those who operate differently
believe in magic and/or religion instead of observable fact.

Dogmatic religion is the biggest threat to the human race.

Correlations must be taken with a grain of salt. One of my texts
on statistics noted a correlation coefficient of 0.98 between the
birth rate of one country in one year to the production of pig iron
in a different country and year.

Donald Miller, Ph.D.
Scientist and Agnostic

Jul. 19 2008 06:11 PM
Susan Wood from Rochester MI

Re Brooke's commentary on the New Yorker cover:
Since at least 1988, Democratic candidates have been subjected to a relentless barrage of ridicule and caricature, not only from Lee Atwater and Karl Rove, but from the so-called "liberal" mainstream media, often under the guise of harmless humor. One of the favorite defenses is "We're just ridiculing what the other side is saying," or, in the case of Dowd, "I'm just warning the candidate of what those nasty Republicans will say about him -- why, they'll call him a weakling, a sissy, an elitist, and say that his wife is a termagant!" But somehow, the stories that she and other "humorists" propagate find their way promptly into conservative talking points.

Michael Dukakis, Al Gore and John Kerry were all determined to take the high road, not only by ignoring right wing smears but by demonstrating that they could "take a joke" if it came from a supposedly liberal source. And we know how well that worked out for all three of them. This time, Dowd & co., to their great dismay, are dealing with a tough, street-smart pol who doesn't let any misrepresentation of his actions or beliefs go unchallenged, even at the risk of being told that "you humorless liberals just can't take a joke." And I say, more power to him.

Jul. 19 2008 10:49 AM

Leave a Comment

Register for your own account so you can vote on comments, save your favorites, and more. Learn more.
Please stay on topic, be civil, and be brief.
Email addresses are never displayed, but they are required to confirm your comments. Names are displayed with all comments. We reserve the right to edit any comments posted on this site. Please read the Comment Guidelines before posting. By leaving a comment, you agree to New York Public Radio's Privacy Policy and Terms Of Use.