Is "Scorpion" Really a Genius? 391
An anonymous reader writes CBS's upcoming hacker show Scorpion is pitched as based on the real life of Irish 'eccentric genius' Walter O'Brien a.k.a. "Scorpion". Some of the claims made for the real Scorpion are extraordinary. A child prodigy with an IQ of 197, hacking Nasa at age 13, [supplying] Ireland with more Personal Computers than DELL and Gateway together. Searching online I wasn't able to find anything which, for me, clearly backed up any of these (or other) claims. For example, rather than being the sixth fastest programmer in the world in 1993, his team ranked 90th out of 250 teams. Curiously, his degree grade was an ok, but hardly stellar B+ (II-I). Does anyone know anything to back up the genius claims being made about Scorpion?
Its nonsense (Score:5, Interesting)
His website proves itself false. He claims it was founded in 1988; however Whois records [domaintools.com] for the domain only go back to 2000, and the web address doesnt appear in the Wayback Machine [archive.org] until 2003.
Looks like the guy has tried to mix his own marketing material into google results, but you can see where his highly touted ScenGen actually comes from here:
http://www.cgd.ucar.edu/cas/wi... [ucar.edu]
This version of MAGICC/SCENGEN was developed primarily with funding from the U.S. Environmental Protection Agency, but it rests on developments carried out over the past 20 years that were funded by a number of organizations.
So the "ScenGen" you keep seeing in all the results is not the same as the one this O'brien dude keeps blathering about. In fact, hes apparently the only one who cares about it; he did do one talk at IEEE in 2010 (though strangely theres no mention of it anywhere except the bog-standard event page), but there doesnt appear to have been any chatter on the internet about it whatsoever.
So, to the AC who posted this: hopefully this is a useful lesson. Anyone can say anything on the internet, and even make it look passingly believable. But if it sounds "too perfect", its probably rubbish.
Re:Never let the truth (Score:3, Interesting)
This Story Reeks of Advertisement (Score:3, Interesting)
I haven't heard of this show until now. I wonder Anonymous Coward is just a sort of straw man trying to drum up interest.
Genius or not, the story is bogus (Score:5, Interesting)
Leaving aside the fact that an IQ score in the 190s is absurd (no one has curved a test over a large enough population for such an answer to reflect actual score distributions), as far as actual, normed IQ tests conducted by actual psychologists go, it's hard to find a test with a ceiling higher than 160 these days. The Weschler, easily the most popular among these, has a ceiling of 160, and getting a score above the low 140s requires doing very well across most of the individual batteries, some of which aren't especially g-weighted. No, the quiz in Omni is not, as far as most psychometricians are remotely concerned, an IQ test. To define it as such is to destroy most of the meaning of the term.
Occasionally, you see high scores due either to very old versions of the Stanford-Binet that did reach above 160 (it's likely that Ted Kaczynski got such a score) or the use of extensions of the old Stanford-Binet to investigate young people who hit or near ceilings, typically on verbal parts of these tests where raw scores tend to have a little more variance, but extrapolations to actual IQ scores aren't valid today due to the Flynn effect (ie: more young people are properly nourished and in intellectually stimulating environments than were in the early 20th century) and the fact that old versions of the Stanford-Binet weren't necessarily normally distributed along the 15-point sigma most tests are today. Though people have attempted to write on the upper echelons of performance on tests of cognitive ability, there's remarkably little that is peer-reviewed.
The tl;dr of all this is that whenever you hear reports of IQ scores above 160, you can more or less assume someone is talking out of their ass.
Re:Never let the truth (Score:4, Interesting)
Yep AFAIK the tests stop at 165 or around there. Anything above is made up as there is no statistical data that can confirm it.
197 would imply there is someone out there with an IQ of 3 as well.
Re:Grades vs IQ (Score:5, Interesting)
From my experience with internet forums, especially gaming forums, youtube commentaries, twitter and facebook, 98% of the observable internet IQs would barely scratch the three digit threshold. A lot of people appear to be well-read, yet basic logic seems to escape most of them. Non-sequitur, strawmen, false dilemma, practically the whole list of logical fallacies can be found there. Yet a lot of people are easily fooled and mistake a few fancy words for competence, which is probably why politicians get elected despite being dumber than a bag of rocks.
I'd say, that most of the internet has about the same average IQ as the general population. Some of US may be a bit more tech savvy, but that's it.
Re:I worked for Dell back then (Score:5, Interesting)
If you read, he supplied more computers than Dell and Gateway combined....... Before 1993.
While both Dell and Gateway existed since the '80s, neither were international powerhouses until the mid-90s. I'm sure both HP and IBM were blowing this guy out of the water in Ireland.
I mean, I sold more cell phones worldwide in 2006 than Apple and Google combined, for crying out loud! (AKA: I sold one.)
Re:Grades vs IQ (Score:2, Interesting)
It has been that way ever since Eternal September. The internet lets everyone speak - but people in general are terrible at recognising the limitations of their knowledge. They aren't stupid, exactly - they are usually entirely competent in their specialised field. But they don't see how inept they are at everything else. They've read a few opinion columns on economics, so they consider themselves fit to weigh in upon tax policy. They took high-school science, so they act as if they can judge the entire field of climatology - and a lot of the time, everything else in science too.
IQ of 197? (Score:4, Interesting)
The standard deviation of IQ seems to be 15
octave:16> erfc((197-100)/15)
ans = 5.9493e-20
That means only a fraction of 5*10^-20 of total humankind would exceed his intelligence.
Let me make a few remarks:
-That would mean humankind could exists in it current size for another 10^11 years without finding a second one like him
-Normal itelligence tests dont resolve in that region. It's pretty impossible to design a tests which ca resolve between 100 and 140 and at the same time distinct between 180 and 190. i am not sure if designing a test between 190 and 197
-The most likely other option is that the distribution of measured IQs is heavy tailed (instead of normal). In that case, the IQ measurement needs to be corrected for that.
I wish that journalists would turn their brain on and not off at every number they cite
Their web site doesn't say much (Score:5, Interesting)
The web site reads like they're a big consultancy, another McKinsey. Then the testimonals are all about Walter. Oracle manager: "Walter showed a great depth of knowledge in Word, WordBasic Macro programming". He still has recommendations up which mention Turbo Pascal. Not seeing rocket science here. The biggest success reported was translating some large English-only application into multiple languages, which made it valuable in Asia. That's nice, but a routine job. He claims to have written a general-purpose program to help with such jobs.
He also claims to have written ScenGen, a "scenario generator". It looks like that originated at Boeing in the mid-1980s [slashdot.org]. Running on a Compaq PC with 2MB back then. The pitch for the current model sounds like the one from back then, although the graphics are probably better now.
The web site is awful. There are lines of text with excess white space in the middle. I looked at the HTML, expecting to find some overly complex Javascript which was misbehaving. No. The HTML source just has explicit non-breaking spaces in the wrong places.
He seems to speak at a lot of strange conferences, such as the Family Office Association. A "family office", in this context,is a staff which manages the family fortune for a large, wealthy family. The Rockefellers have one.
This is getting weird.
Re:Grades vs IQ (Score:5, Interesting)
The curse of the academically capable: I breezed through school getting very good grades with no effort at all. Never revised - I was just good at the code subjects (Except English Lit). Then went to university and had a breakdown, because it was the first time I'd been seriously challenged.
Re:Never let the truth (Score:2, Interesting)
I was tested back in 1981 and was technically "off the chart", which made it difficult for the test proctors to fill a box on the form. So, they were basically forced to estimate my IQ, probably through extrapolation from my other tests. For what it's worth, *my* estimated score was 189.
I also started my first company at 13 and have started and sold several since then. At a relatively young age, I wised up and distanced myself from my 'black hat' personas and handles. I became an consultant and I currently use my skills to make good money. Anyway, this guy till looks and sounds like a smug, dick-waving narcissist and is probably a complete bullshit artist.
I was a "recognized and respected" hacker and phreaker myself, well before the days of the public Internet. I was a member of a few groups back in the day, and my 'main group' occasionally worked with American groups. The only 'Scorpion' I remember was a member of MOD, the group that got into a pissing match with LOD after I left the scene in the late 80's. All but a couple of my old handles have since been claimed by script kiddies and mouth-breathing punks, so I can only guess this is the case for 'Scorpion'.
It's not a big deal for me - I prefer to use my real name around the office.
Re:He claims this himself (Score:5, Interesting)
How about this then? From http://www.scorpioncomputerservices.com/whoweare.html [scorpionco...rvices.com]:
That's production quality.
Re:Never let the truth (Score:5, Interesting)
Also, from what I've come to understand, IQ tests just aren't ultimately that useful for ranking the intelligence of smart people. At least, according to a few different psychologists that I've talked to, the main purpose of IQ tests, the reason they're used and considered valid, is in detecting developmental problems rather than detecting genius.
So if someone scores a 160 as opposed to a 130, it gives some indication that the person is good at certain kinds of mental activity-- for example, spotting patterns in numbers and geometric shapes. That's about all you can really say, and it's ultimately not that meaningful. Of course, people who are really brilliant math/science types will likely do pretty well on these tests, but doing well on these tests does not make you a brilliant mathematician or scientist.
Where the test is helpful is in seeing problems/deficiencies. If you test a child who gets a score of 70, then it's a pretty good indicator that he should be put into a special program. That's what the test is good for, and that's largely why they administer it. If you're an adult, lording your 150 IQ over someone who scored a 145, then you're an idiot. If you're citing your 197 IQ as some sort of qualification for something, it's that much dumber.
Re:Never let the truth (Score:4, Interesting)
If you estimate the distribution of all test takers for a given age category, then it is entirely possible to show that someone is 6.5 sigma from the mean, without actually sampling 1/10 billion people to be sure that all of them are less intelligent. That, after all, is the point of statistics.
First, note that most of the test norming has been done on samples of a few thousand people. The major IQ tests have been normed rather rigorously a number of times, but certainly not on samples of more than tens of thousands of people. Extrapolating an that outlier is somewhere around 6.5 sigma from the mean with any accuracy from such population samples is pretty difficult to begin with.
But beyond that, one has to ask whether it is even possible to measure "general intelligence" with a level of accuracy that we could pinpoint, literally, the smartest person in the world. I don't think we can. Maybe we could come up with a definition of "general intelligence" that accurately puts someone in the top 1% or top 0.1%, but beyond that, we're probably not able to differentiate with any precision.
So, the correct scientific and statistical conclusion to draw from an outlying score that appears to be 6.5 sigma from the mean in this case is -- "he's pretty far above average." Whether he's actually 4 sigma or 7 sigma is not really something any test can specify with any precision... and stating such a number is in fact meaningless. (The norms for modern IQ tests generally have a ceiling of 160 for that reason -- that's 4 sigma above norm. Beyond that, the tests make no claim, and that's probably appropriate given the ambiguity about what exactly we're measuring.)