Home

Sep. 4th, 2006

eyes black and white, eyes

Deft or Daft?

I've long noticed that the focus of a gaze, the a propos of a smirk, can distinguish the aware and open-minded from the fools and the fanatical. Just peek at random people in the street, compare to pictures of random researchers on the Internet, or to pictures of random religious fanatics also on the net. There's a lot you can tell -- or maybe you can't, but then odds are a lot of people could tell you can't.

Several friends recently shared their similar experience with me. And so, in the spirit of HotOrNot, I eventually bought DEFTORDAFT.COM. The idea would be to rate people by intelligence according to published pictures. Is there a business plan? Not quite yet. Maybe I could just be diversification for the hotornot site. Or maybe we could link to a geek-oriented dating site such as OkCupid? We could also provide a captcha system in the spirit of HotCaptcha that recognizes people intelligent enough to differentiate the deft from the daft.

I suspect the data processing will be more complex than for HotOrNot, since a sense for sexual attraction is more uniformly shared than a sense for intellectual discrimination. Although even HotOrNot could probably classify people in clusters according to their aesthetic tastes. Because DeftOrDaft will judge the intelligence of the viewer as well as the viewee, we'll probably have to resort to pre-calibrated samples, such as people with known IQs or SATs, or from colleges with known SAT averages; and we'll probably have to use Bayesian analysis techniques such as used by LikeBetter. So, interesting stuff going on behind the scenes...

Ideas are cheap. This one is free. Wanna join with the expensive part: actually building the site?

Jul. 30th, 2006

eyes black and white, eyes

Risk aversion, or Why smart women are worth more than smart men

Even though the topic is actively censored by the intellectual terrorists who control the intelligentsia with their political correctness, it is a fact confirmed by everyone's experience as well as by repeated scientific experiments that there is a measurable discrepancy between the distribution of general intelligence among men and women. IQ as measured in either sex leads to a normal distribution of intelligence along a bell curve, but the standard deviation of IQ amongst males is noticably larger, and the average of IQ amongst males is higher.

What the statistics mean... )

Now, I have a rational explanation for both statistics that does not depend on any significant difference distribution of genetic abilities at birth, except for one essential trait: that males in general have a tendency to be more reckless than females.

Why Risk-Aversion... )

What does this difference in risk-aversion have to do with the distribution of intelligence, will you ask? Everything! Because more risks means more selection, both internal and external.

Intelligence as the result of selection... )

Ben once wrote that One good Husband is worth two good Wives; for the scarcer things are, the more they're valued. Well, then by the same token, one intelligent woman is worth a lot of intelligent men.

Postscriptum... )

Dec. 3rd, 2005

eyes black and white, eyes

My mouth washed / Pan sur le bec

Ulrich Hobelmann remarks that in my article Public Goods Fallacies (and hence the essay Government is the Rule of Black Magic from which it is excerpted), I had misattributed to Ambrose Bierce's The Devil's Dictionary the following quip:

Majority, n.: That quality that distinguishes a crime from a law.

Well, I had wrongly jumped to conclusion from the proximity and similarity of the quote to other funny definitions by Bierce, and I hadn't deigned to double check the link I had provided myself. The quip is no quote from there, and Google doesn't obviously find any attribution to what I found in the Linux fortune(6) files. Mea Magna Culpa.

Making wild generalizations is a good thing; it is the very process of imagination by which any science progresses; but the necessary counterpart of an idea generator, as good as it may be (and good yields are still infinitesimal), is a sturdy crap detector. (OK, Google finds only my blog for this formula, that I derived from Vernor Vinge's A deepness in the sky, and who knows where is the origin of this traditional AI model.) And the bullshit filter part of my mind seems to be lacking, at times. From now on, I will try to systematically devise tests for the conjectures I make, before I may jump to conclusion.

It isn't enough to say: I was mistaken; one must say how one was mistaken. -- Claude Bernard

 

Ulrich Hobelmann me fait remarquer que dans mon article Public Goods Fallacies (et donc dans l'essai L'État, Règne de la Magie Noire dont il est extrait), j'ai attribué à tort au Dictionnaire du Diable de Ambrose Bierce l'aphorisme suivant:

Majorité, n. f.: Cette qualité qui distingue un crime d'une loi.

Eh bien, j'ai trop hâtivement déduit cette attribution de la proximité et similarité de cette citation avec d'autres définitions amusantes de Bierce, et j'ai manqué aux vérifications les plus élémentaires avec le lien même que j'offrais. Mon attribution était erronée, et Google ne permet pas de trouver facilement une attribution à ce que je trouvai initiallement dans les fichiers du programme fortune(6) de Linux. Mea Magna Culpa.

Pouvoir faire des généralisations rapides est une qualité; c'est par l'imagination de telles généralisations que toute connaissance avance; mais la contrepartie nécessaire d'une génération d'idée, aussi bonne soit-elle (et même les bons rendements sont infinitésimaux), est un solide filtre à conneries. (Bon, Google ne trouve que mon propre blog pour cette formule, que j'ai reprise à A deepness in the sky de Vernor Vinge, et qui vient originellement de dieu sait quel modèle traditionnel d'IA.) Or, trop souvent, mon détecteur de bêtise semble ne pas fonctionner correctement. Dorénavant, j'essaierai d'effectuer systématiquement des tests pour toute conjecture que je ferai, avant de sauter aux conclusions.

Il ne suffit pas de dire: je me suis trompé; il faut dire comment on s'est trompé. -- Claude Bernard

Sep. 20th, 2005

eyes black and white, eyes

Fear of Differences, Fear of Reality

Since I'm processing the backlog of my blog notes, here's one piece about a topic that has been at the center of several scandals this year yet surprisingly little open public discussions: genetic differences amongst human beings.

Read more... )
eyes black and white, eyes

May 2008

S M T W T F S
    123
45678910
11121314151617
18192021222324
25262728293031

Tags

Syndicate

RSS Atom
Powered by LiveJournal.com