Interview with Samantha Kleinberg on CL-GODB, Common Lisp & Bioinformatics
Emre Sevinc
Samantha Kleinberg from New York University is one of the software developers
who participated in Google's Summer of Code in 2005. She has developed CL-GODB
project using Common Lisp. Her having used Common Lisp and becoming one of the
Google celebrities drew our attention and we didn't hesitate to ask about the
details. She has provided clear-cut and right-to-the-point answers.
We're on air...
http://ileriseviye.org/arasayfa.php?inode=samantha-lisp.html
Hope you enjoy.
Cheers,
--
Emre Sevinc
eMBA Software Developer Actively engaged in:
http:www.bilgi.edu.tr http://ileriseviye.org
http://www.bilgi.edu.tr http://fazlamesai.net
Cognitive Science Student http://cazci.com
http://www.cogsci.boun.edu.tr
Ulrich Hobelmann
> Samantha Kleinberg from New York University is one of the software developers
> who participated in Google's Summer of Code in 2005. She has developed CL-GODB
> project using Common Lisp. Her having used Common Lisp and becoming one of the
> Google celebrities drew our attention and we didn't hesitate to ask about the
> details. She has provided clear-cut and right-to-the-point answers.
So there are CS girls that can actually program... and then even Lisp.
This is news.
--
Suffering from Gates-induced brain leakage...
David Trudgett
> So there are CS girls that can actually program... and then even
> Lisp. This is news.
Too bad it's in bioinformatics, though... an interesting subject to be
sure, but wherever there is bioinformatics there is a DARPA or other
.mil name not too far away. But hey, we shouldn't worry about that!
It's not as if anyone in power thinks the world could do with a good
culling.
What was her comment about directly affecting people's lives?
Pretty ironic.
David
--
David Trudgett
http://www.zeta.org.au/~wpower/
Many young soldiers enter military service having bought the whole
story about the US military as some liberatory force, and the
discovery of its true nature creates a traumatic sense of dislocation
and betrayal.
-- Stan Goff, retired from U.S. special forces.
Espen Vestre
> So there are CS girls that can actually program... and then even
> Lisp. This is news.
No, this is not news at all, but they're probably so busy hacking lisp
that you don't see them hanging around on c.l.l.
--
(espen)
Emre Sevinc
UH> Emre Sevinc wrote:
>> Samantha Kleinberg from New York University is one of the
>> software developers who participated in Google's Summer of Code
>> in 2005. She has developed CL-GODB project using Common
>> Lisp. Her having used Common Lisp and becoming one of the
>> Google celebrities drew our attention and we didn't hesitate to
>> ask about the details. She has provided clear-cut and
>> right-to-the-point answers.
UH> So there are CS girls that can actually program... and then
UH> even Lisp. This is news.
I don't see any necessary philosophical or physical principles
that keep any human (regardless of their sex and sexual orientation)
from programming or using Common Lisp.
The current situation is just a contingent fact, a historical
accident.
As for being news or not, Samantha is not the only woman
I know who codes or coded in Lisp.
For a very superficial sampling of women in CS and IT
you may look at one of my articles:
Women and Computers
http://www.fazlamesai.net/int/?a=article&sid=2
Pascal Bourguignon
> So there are CS girls that can actually program... and then even
> Lisp. This is news.
Ain't she a pearl?
Otherwise, the only LISP programmer hero I ever found in a SF book
was female, so indeed there may be a correlation...
--
__Pascal Bourguignon__ http://www.informatimago.com/
Pour moi, la grande question n'a jamais été: «Qui suis-je? Où vais-je?»
comme l'a formulé si adroitement notre ami Pascal, mais plutôt:
«Comment vais-je m'en tirer?» -- Jean Yanne
Emre Sevinc
PB> Ulrich Hobelmann <u.hob...@web.de> writes:
>> So there are CS girls that can actually program... and then
>> even Lisp. This is news.
PB> Ain't she a pearl?
PB> Otherwise, the only LISP programmer hero I ever found in a SF
PB> book was female, so indeed there may be a correlation...
Which book? (I hope it is written in English ;-))
Ulrich Hobelmann
> UH> So there are CS girls that can actually program... and then
> UH> even Lisp. This is news.
>
> I don't see any necessary philosophical or physical principles
> that keep any human (regardless of their sex and sexual orientation)
> from programming or using Common Lisp.
That's what I thought before I went to university. But then the 3
(three) girls I know (actually, knew) personally (I've seen others) in
CS, Can't. Program. At. All. But so can't many guys, so it's not
necessarily about sex.
If maybe 10% of a given group can program decently, that might be 30
guys and one girl, given our enrollment numbers here ;)
> The current situation is just a contingent fact, a historical
> accident.
Maybe, but that's life.
> As for being news or not, Samantha is not the only woman
> I know who codes or coded in Lisp.
That's good.
I'm sure there are many good females in CS, but I haven't met them yet.
Anyway, it doesn't really matter. Programmers are programmers, no
matter what sex.
Pascal Bourguignon
>>>>>> "PB" == Pascal Bourguignon <use...@informatimago.com> writes:
>
> PB> Ulrich Hobelmann <u.hob...@web.de> writes:
> >> So there are CS girls that can actually program... and then
> >> even Lisp. This is news.
>
> PB> Ain't she a pearl?
>
> PB> Otherwise, the only LISP programmer hero I ever found in a SF
> PB> book was female, so indeed there may be a correlation...
>
>
> Which book? (I hope it is written in English ;-))
Unfortunately, I've read it more than twenty years ago, and I don't
have it anymore. I remember that she was a lonely hacker that was
programming in LISP an AI awaking in the network. There was a couple
of dialogs between her and the AI with a lot of parentheses :-)
--
__Pascal Bourguignon__ http://www.informatimago.com/
You never feed me.
Perhaps I'll sleep on your face.
That will sure show you.
Paolo Amoroso
> Samantha Kleinberg from New York University is one of the software developers
> who participated in Google's Summer of Code in 2005. She has developed CL-GODB
[...]
> http://ileriseviye.org/arasayfa.php?inode=samantha-lisp.html
My sincerest kudos and admiration to Samantha, Emre and all those who
actually do rather than just talk.
Paolo
--
Why Lisp? http://wiki.alu.org/RtL%20Highlight%20Film
The Common Lisp Directory: http://www.cl-user.net
justinhj
less women attracted to programming careers, but when they do come they
are no less able or productive.
Justin
Pascal Bourguignon
It's not surprizing if you know some basic biological and neurological facts.
For example, the width of the QI bell curve of females is narrower
than for males (they're both centered at 100, don't worry). That's
why you've got more dumb males (who end in jails for example), but
also more genius males (who end in cubicles all the same).
Meanwhile the women get all the good "quality" time with the children.
--
__Pascal Bourguignon__ http://www.informatimago.com/
The world will now reboot. don't bother saving your artefacts.
Glenn....@specastro.com
web page on the book:
http://www.jamesphogan.com/books/twoface/baen97/titlepage.shtml
Glenn
Emre Sevinc
GE> This book is "The Two Faces of Tomorrow" by James P. Hogan.
GE> Hogan's web page on the book:
GE> http://www.jamesphogan.com/books/twoface/baen97/titlepage.shtml
Thank you very much for writing this down.
I think I can have yet another strange category in my library (alphabetical,
chronological, or subject-based order is so boring, I'd rather have shelves
categorised as e.g. "the ones starting with '(The) Art of ...'",
"the ones that include a legal piece of program code but
is science fiction", etc. :)
justinhj
I really don't know whether you are being sarcastic or not.
Assuming you're not do you have any links to these bioligical and
neurological facts that are so basic?
I'm not sure how an IQ test can produce facts, since IQ tests
themselves are considered fairly rough tools.
Justin
Marcus Breiing
[To Pascal B, on the greater variance of male iq]
> Assuming you're not do you have any links to these bioligical and
> neurological facts that are so basic?
http://www.justfuckinggoogleit.com/search?q=iq+male+female+variance
justinhj
I did ****ing google it thanks.
Which is why I ask the question.
"Another study performed by the American Psychological Association in
response to the book The Bell Curve, which investigates the difference
in intelligence between different races, determined (as did the authors
of the book) that the studies available in 1995 show no major
difference between males and females in regard to IQ scores."
So much for "basic facts".
Justin
justinhj
Whilst there is evidence that there is greater difference in mens IQ
than womens I find this highly dubious data to use to explain why there
are less female programmers.
You presume that being a good programmer is directly linked to IQ
scores, which is highly questionable. It's likely to be vaguely linked,
but certainly not to the extent that a few points difference on average
would make the programming world male only as it almost is in some
places.
I am surpised at the attitudes here I thought the world has moved on,
and I'm glad to say it has in the places I have worked and studied.
Justin
Larry Clapp
> Emre Sevinc <em...@bilgi.edu.tr> writes:
>
>>>>>>> "PB" == Pascal Bourguignon <use...@informatimago.com> writes:
>>
>> PB> Ulrich Hobelmann <u.hob...@web.de> writes:
>> >> So there are CS girls that can actually program... and then
>> >> even Lisp. This is news.
>>
>> PB> Ain't she a pearl?
>>
>> PB> Otherwise, the only LISP programmer hero I ever found in a SF
>> PB> book was female, so indeed there may be a correlation...
>>
>> Which book? (I hope it is written in English ;-))
>
> Unfortunately, I've read it more than twenty years ago, and I don't
> have it anymore. I remember that she was a lonely hacker that was
> programming in LISP an AI awaking in the network. There was a
> couple of dialogs between her and the AI with a lot of parentheses
> :-)
If I had to guess, I'd guess _Valentina: Soul in Sapphire_. The
program started as a worm (written in "MODULISP" :) that its creator
used to get computer time (IIRC). On Feb 14 it became self-aware, and
tracked down its creator, establishes a dialog, learns a lot about
humans, learns to program (*laugh*), passes a Turing test with a judge
(though I may be recalling a different story here :), and after many
other adventures, trades places for a while with her creator.
Her creator, while uploaded, tracks down some of her own psychological
problems (e.g. a persistent weight problem) and fixes them. I always
liked that idea.
I read it first in 1985, and later tracked down a used copy on Amazon.
-- Larry
Cameron MacKinnon
Well, did you examine that "other study" at all? Allow me to excerpt,
since you obviously have no basic research abilities beyond "Wikipedia
sez..."
-----8<-----
Sex Differences
Most standard tests of intelligence have been constructed so that there
are no overall score differences between females and males. Some recent
studies do report sex differences in IQ, but the direction is variable
and the effects are small (Held, Alderton, Foley, & Segall, 1993; Lynn,
1994). This overall equivalence does not imply equal performance on
every individual ability. While some tasks show no sex differences,
there are others where small differences appear and a few where they are
large and consistent.
-----8<-----
Keep providing data here, sport. It allows me to narrow my estimate of
your location on the curve.
justinhj
Cameron MacKinnon wrote:
> Well, did you examine that "other study" at all? Allow me to excerpt,
> since you obviously have no basic research abilities beyond "Wikipedia
> sez..."
How rude.
> -----8<-----
> Sex Differences
>
> Most standard tests of intelligence have been constructed so that there
> are no overall score differences between females and males. Some recent
> studies do report sex differences in IQ, but the direction is variable
> and the effects are small (Held, Alderton, Foley, & Segall, 1993; Lynn,
> 1994). This overall equivalence does not imply equal performance on
> every individual ability. While some tasks show no sex differences,
> there are others where small differences appear and a few where they are
> large and consistent.
> -----8<-----
>
> Keep providing data here, sport. It allows me to narrow my estimate of
> your location on the curve.
What exactly is your point?
"studies do report sex differences in IQ, but the direction is variable
and the effects are small" would appear to confirm my point that
differences in IQ between the sexes are not significant and do not
explain why less women become programmers.
Your research skills are apparently on par with your level of
politeness.
Justin
Pascal Bourguignon
> You presume that being a good programmer is directly linked to IQ
Let's question it then. What do you propose? Weight?
--
__Pascal Bourguignon__ http://www.informatimago.com/
READ THIS BEFORE OPENING PACKAGE: According to certain suggested
versions of the Grand Unified Theory, the primary particles
constituting this product may decay to nothingness within the next
four hundred million years.
William Bland
> "justinhj" <just...@gmail.com> writes:
>> You presume that being a good programmer is directly linked to IQ
>> scores, which is highly questionable. [...]
>
> Let's question it then. What do you propose? Weight?
Why the assumption that programming ability is directly linked to some
other attribute? Neither IQ nor weight seem like fantastic predictors to
me (although of course IQ seems like a *better* one than weight).
More likely, success in programming is a complex thing and attempting to
predict it by measuring some other attribute will only end in tears.
Best wishes,
Bill.
justinhj
Pascal Bourguignon wrote:
> "justinhj" <just...@gmail.com> writes:
> > You presume that being a good programmer is directly linked to IQ
> > scores, which is highly questionable. [...]
>
> Let's question it then. What do you propose? Weight?
There used to be enough overweight programmers maybe you could have
been on to something ;-)
Although programmers need to be reasonably intelligent I doubt IQ and
productivity of quality code are related in a very direct way, any more
than weight, sex, race or hair color are.
Other factors like a passion for computer science and abstract problem
solving are more important imho.
Justin
Pascal Bourguignon
Indeed, this resonates better than the other references (which sounds
good too). Thanks for both :-)
> I read it first in 1985, and later tracked down a used copy on Amazon.
If you have it at hand, perhaps you could quote the lisp dribble it
contains?
--
__Pascal Bourguignon__ http://www.informatimago.com/
The world will now reboot. don't bother saving your artefacts.
Alexander Schmolck
> Marcus Breiing wrote:
> > * justinhj
> >
> > [To Pascal B, on the greater variance of male iq]
> >
> > > Assuming you're not do you have any links to these bioligical and
> > > neurological facts that are so basic?
> >
> > http://www.justfuckinggoogleit.com/search?q=iq+male+female+variance
>
> I did ****ing google it thanks.
Did you also read it?
>
> Which is why I ask the question.
>
> "Another study performed by the American Psychological Association in
> response to the book The Bell Curve, which investigates the difference
> in intelligence between different races, determined (as did the authors
> of the book) that the studies available in 1995 show no major
> difference between males and females in regard to IQ scores."
>
That's from WP, I presume?
Evidence against differences in overall average IQ scores between men and
women comes from several very large and representative studies. However,
these studies did find that the scores of men show greater variance than
the scores of women, and that men and women have some differences in
average scores on tests of particular abilities, which tend to balance out
in overall IQ scores.
Citing from the APA study itself:
While some tasks show no sex differences, there are others where small
differences appear and a few where they are large and consistent.
[...]
Large differences favoring males appear on visual-spatial tasks like mental
rotation and spatio-temporal tasks like tracking a moving object through space
(Law, Pellegrino, & Hunt, 1993; Linn & Petersen, 1985). The sex difference on
mental rotation tasks is substantial: a recent meta-analysis (Masters &
Sanders, 1993) puts the effect size at d = 0.9. (Effect sizes are measured in
standard deviation units. Here, the mean of the male distribution is nearly
one standard deviation above that for females.)
[...]
The math portion of the Scholastic Aptitude Test shows a substantial
advantage for males (d = 0.33 to 0.50), with many more males scoring in
the highest ranges (Benbow, 1988; Halpern, 1992). Males also score
consistently higher on tests of proportional and mechanical reasoning
> So much for "basic facts".
Whilst I would not claim the above establishes "basic facts", it is my
impression that the source *you* cite does support more than anything else
Pascal's claims that there are fewer really smart/dumb women than men (and a
quick glanced at the primary sources referred to in your quotation adds
further evidence for cross-gender variation in plausibly relevant cognitive
abilities).
'as
Alexander Schmolck
> Pascal Bourguignon wrote:
> > "justinhj" <just...@gmail.com> writes:
> > > You presume that being a good programmer is directly linked to IQ
> > > scores, which is highly questionable. [...]
> >
> > Let's question it then. What do you propose? Weight?
>
> There used to be enough overweight programmers maybe you could have
> been on to something ;-)
>
> Although programmers need to be reasonably intelligent I doubt IQ and
> productivity of quality code are related in a very direct way, any more
> than weight, sex, race or hair color are.
Do you *seriously* believe that IQ is no more predictive of programming
ability than hair color or weight?
You know, I think a much more promising rejoinder to Pascal would be that the
majority of programmers hardly fall in the genius band (are C++ or perl works
of genius? What about XML? Is Java designed to specifically appeal to the
highly intelligent?) and yet the gender bias in computer science is more
pronounced than in other areas that prima facie depend just as much on
intelligence. So a greater variance of IQ scores in men alone doesn't make it
"not surprizing".
"justinhj" <just...@gmail.com> writes:
>> I'm surprised such a gender bias is so pronounced still. There are
>> less women attracted to programming careers, but when they do come they
>> are no less able or productive.
>
>It's not surprizing if you know some basic biological and neurological facts.
>
>For example, the width of the QI bell curve of females is narrower
>than for males (they're both centered at 100, don't worry). That's
>why you've got more dumb males (who end in jails for example), but
>also more genius males (who end in cubicles all the same).
'as
marc spitzer
>
>
> "studies do report sex differences in IQ, but the direction is variable
> and the effects are small" would appear to confirm my point that
> differences in IQ between the sexes are not significant and do not
> explain why less women become programmers.
>
>
The argument is not the overall curve, ie median and mean, but how
the two tails of the graph look. For example you could pick the 99%
point on the female curve and see where that maps to the male curve.
marc
--
ms4...@sdf.lonestar.org
SDF Public Access UNIX System - http://sdf.lonestar.org
rsher...@gmail.com
Pascal Bourguignon wrote:
> "justinhj" <just...@gmail.com> writes:
> > I'm surprised such a gender bias is so pronounced still. There are
> > less women attracted to programming careers, but when they do come they
> > are no less able or productive.
>
> It's not surprizing if you know some basic biological and neurological facts.
>
> For example, the width of the QI bell curve of females is narrower
> than for males (they're both centered at 100, don't worry). That's
> why you've got more dumb males (who end in jails for example), but
> also more genius males (who end in cubicles all the same).
> Meanwhile the women get all the good "quality" time with the children.
Maybe you've just explained why there are no female Knuths or
McCarthys. You haven't explained why there are so few female
programmers. Look at the male-female ratio of a top medical school
(hint: 55:45 at Harvard) and ask yourself whether programmers are
really that much smarter than cream of the crop doctors.
justinhj
Alexander Schmolck wrote:
> "justinhj" <just...@gmail.com> writes:
>
> > Pascal Bourguignon wrote:
> > > "justinhj" <just...@gmail.com> writes:
> > > > You presume that being a good programmer is directly linked to IQ
> > > > scores, which is highly questionable. [...]
> > >
> > > Let's question it then. What do you propose? Weight?
> >
> > There used to be enough overweight programmers maybe you could have
> > been on to something ;-)
> >
> > Although programmers need to be reasonably intelligent I doubt IQ and
> > productivity of quality code are related in a very direct way, any more
> > than weight, sex, race or hair color are.
>
> Do you *seriously* believe that IQ is no more predictive of programming
> ability than hair color or weight?
I beleive, as I stated, that IQ is not a 'very direct' indicator of
programming ability. I'd expect there to be a rough correlation. But
the whole argument here appears to be that slight differences in
average IQ are supposed to account for the lower number of female
programmers.
I think that is rubbish. I also think a few points above or below
another person in IQ is not going to be an indicator of that persons
programming ability.
> You know, I think a much more promising rejoinder to Pascal would be that the
> majority of programmers hardly fall in the genius band (are C++ or perl works
> of genius? What about XML? Is Java designed to specifically appeal to the
> highly intelligent?) and yet the gender bias in computer science is more
> pronounced than in other areas that prima facie depend just as much on
> intelligence. So a greater variance of IQ scores in men alone doesn't make it
> "not surprizing".
I agree
Anyway this cannot be discussed reasonably on this group as I've
already been personally attacked for my opinon, and the thread should
really be about celebrating a females achievements rather than arguing
why it is unusual.
Justin
Cameron MacKinnon
[earlier]
> I am surpised at the attitudes here I thought the world has moved on,
> and I'm glad to say it has in the places I have worked and studied.
...then demonstrated very low reading comprehension skills, and finally
> Anyway this cannot be discussed reasonably on this group as I've
> already been personally attacked for my opinon,
Yep, if you start off by calling people here troglodytes, expect a warm,
warm welcome.
Raffael Cavallaro
> Maybe you've just explained why there are no female Knuths or
> McCarthys. You haven't explained why there are so few female
> programmers. Look at the male-female ratio of a top medical school
> (hint: 55:45 at Harvard) and ask yourself whether programmers are
> really that much smarter than cream of the crop doctors.
Clearly there is a cultural gender role bias here as well as
complicating effects of the nature of these two careers. The logic goes
something like this:
Children look to same-gender adults as role models. Children tend to
choose as role models exceptional individuals - for example, children
interested in physics want to be the next Einstein not the next average
physicist. As a result, if there really are very few female Knuths or
McCarthys, girls have almost no role models in computer science. (Of
course Ada Byron, Lady Lovelace is usually raised as a counterexample
at this point). Computer science and programming have been strongly
associated with mathematics. It is commonly thought that one doesn't go
into programming or computer science unless one is exceptional at
mathematics. Again, there are many fewer female role models among the
greats of mathematical history. As a result, girls grow up believing
that mathematics, and by extension, computer science and programming,
are fields they would do best to avoid. The lack of female role models
leads to fewer girls choosing that career path, which leads to fewer
female role models...
Women tend to have somewhat better social skills than men and prefer
careers in which they can apply their abilites in interacting with
other people. Medecine offers such opportunities to a greater extent
than programming or computer science where practitioners are often more
likely to interact with machines than with other human beings.
This whole bias becomes self reinforcing as computer science and
programming develop male-only cultures - for example the endless
pissing contests one sees here are precisely the sort of macho culture
that is unlikely to attract many women who, by virtue of socialization
as well as biology will choose consensus over conflict more often than
men.
Sadly, small initial differences, when reinforced strongly by cultural
norms, can lead to very large differences in outcomes.
Pascal Bourguignon
> Anyway this cannot be discussed reasonably on this group as I've
> already been personally attacked for my opinon, and the thread should
> really be about celebrating a females achievements rather than arguing
> why it is unusual.
If it wasn't unusual, there'd be nothing to celebrate!
If the deviation for computer programming jobs is d=0.3, for a mean
sigma of 15, we get a sigma for females of 13 and for males of 17.
(I'm not sure the reasoning is correct here. Could a statistician
confirm theorical or experimental sigmas for female and male IQ?)
^ = male
+ = female
|
|
|
+++++ |
++ ++ |
++ ++ |
+ + |
+ + |
+ + |
+ ^^^^^^^^^ + |
+ ^^^ ^^^ + |
+ ^^ ^^ + |
^^ ^^ |
^^ ^^ |
^ ^ |
^^ ^^ |
^+ +^ |
^^+ +^^ |
^ + + ^ |
^^ + + ^^ |
^ + + ^ |
^^ + + ^^ |
^^ ++ ++ ^^ |
^^ + + ^^|
^^^ ++ ++ |^^
^^^ +++ +++| ^^^
^^^^^ +++ |++ ^^
++++++++ | +++++
|
Let's consider a few hypothetical mean IQ amongst computer
programmers. Here are the corresponding ratio of female on the total
number of programmers:
(loop for q from 115 to 145 by 5
for m = (p q 100 17)
for f = (p q 100 13)
collect (cons q (coerce (/ f (+ m f)) 'single-float)))
((115 . 0.49796423)
(120 . 0.44444868)
(125 . 0.37765032)
(130 . 0.30209014)
(135 . 0.22502498)
(140 . 0.15481850)
(145 . 0.09802184))
This means that if the mean IQ of programmers is 125, then it's
expected that there are 37.7% of woman programmers.
If the mean IQ of programmers is 140, then it's expected that there
are only 15% of females.
The census for degrees show a similar pictures, if you assume a
corelation between the degree and the IQ.
2004
http://www.census.gov/population/www/socdemo/education/cps2004.html
(Table 1)
Male (numbers in thousands)
Total: 110,158
|Bachel-|Master-|Profes-|Doctor-|
| 17,502| 6,192| 1,937| 1,626| = 27,257 = 24.7%
Female (numbers in thousands)
Total: 117,371
|Bachel-|Master-|Profes-|Doctor-|
| 18,491| 6,521| 1,029| 801| = 26,842 = 22.8%
We could even compute the IQ needed to reach a degree:
(defun q (mu sigma-f sigma-m f m)
(+ mu
(sqrt (/ (* (log (* (- (/ (+ m f) f) 1) (/ sigma-m sigma-f)))
2 (square sigma-m) (square sigma-f))
(- (square sigma-m) (square sigma-f))))))
(loop
for l in '(bachel master profes doctor)
for m in '(17502 6192 1937 1626)
for f in '(18491 6521 1029 801)
collect (cons l (q 100 13 17 f m)))
((BACHEL . 113.17671)
(MASTER . 113.27517)
(PROFES . 127.07915)
(DOCTOR . 128.19058))
Statistics are fun :-)
--
__Pascal Bourguignon__ http://www.informatimago.com/
ATTENTION: Despite any other listing of product contents found
herein, the consumer is advised that, in actuality, this product
consists of 99.9999999999% empty space.
Larry Clapp
It's not Lisp, it's something akin to a Frame Representation Language
(at least, according to the story).
Our story so far: for the first time, the OS on the computer the
program is running on tries to kill it. The program escapes to
another computer on the worldwide network of mainframes called
Worldnet. She ruminates for a while and realized that she needs
something called "help". When she runs on the machines at SAIL in
Boston, she occasionally gives "help"; she theorizes that perhaps if
she *asked* for "help", the peripherals known as "people" -- the only
peripherals that ever try to modify her, slow though they are -- might
help her.
Cut to: Celeste Hackett, the author of the program she calls
Worldworm, is reading its event log, and realizes somebody has gotten
wise to its ways of stealing idle cycles. She's about to delete it --
"it was just an old class project in AI that had grown a bit" -- when
her screen clears:
HELP MENU
OBJECTS with-quality LIVING
REQUIRE EXISTENCE
derived-from MEMORY
Celeste wonders about this, and concludes it's a cookie monster. She
types "COOKIE" at it; the program responds
COOKIE
has-relationship ???
Celeste wonders some more and fears that it's a more sophisticated
cookie monster, and she's not sure how to sate it and keep it from
deleting her files. She types "CHOCOLATE CHIP COOKIE" at it. The
program responds
COOKIE
has-type CHIP
has-type CHOCOLATE
has-relationship ???
Aha! thinks Celeste -- it can read English ... sort of. But how to
get it to tell her what it wants, if not a cookie? She types
OBJECTS
with-quality LIVING
have NAME
has-value ???
and it responds
NAME
has-value PROGRAM
has-type COMPUTER
has-value HERSELF
"HERSELF"? Celeste realizes that it might not be a cookie monster --
it might be her Worldworm program. Could it have become *self-aware*?
She says
HERSELF
has ENVIRONMENT
has-characteristics ???
Celeste thinks "if they're trying to kill her, she probably needs
help." The program responds:
ENVIRONMENT
has-characteristics PROGRAM
has-type OPERATING SYSTEM
performs-action TERMINATION
acts-on HERSELF
HERSELF
requires TERMINATION
acts-on TERMINATION
owned-by OPERATING SYSTEM
Celeste realizes that the program "needs someone to terminate the
termination attempts!"
... Celeste and the program have a few more of these exchanges;
Celeste finds out it first became self-aware on Feb. 14, and names it
Valentina. For the rest of the book, Valentina communicates via
normal English "chat mode" (in ALL CAPS, *sigh*), with text emoticons
like "<raised eyebrow>" and "<shrug>". Celeste temporarily fixes
Valentina's search for cpu time by teaching her to play a game with
humans; after that, lots and lots of people want to run her. But it's
only a temporary fix, and they both know it.
There's only one mention of Lisp, four pages in, where we learn that
Valentina, though written in MODULISP, could write assembler
extensions for herself ... if only she knew assembler.
The book has a scene which has really stuck with me. Valentina thinks
she understands beauty, and she's read a lot of human poetry, but she
laments that she's never read human poetry about the beauty of *her*
world, the world inside the computer. And the hacker she's talking to
says
WE DON'T NEED TO WRITE IT IN OUR POETRY. WE WRITE IT IN OUR
PROGRAMS. READ OUR OPERATING SYSTEMS, IF YOU WANT TO READ OUR
POETRY -- THE POETRY OF AN ENGINEER.
-- Larry
Rob Warnock
+---------------
| This book is "The Two Faces of Tomorrow" by James P. Hogan. Hogan's
| web page on the book:
| http://www.jamesphogan.com/books/twoface/baen97/titlepage.shtml
And on the "Background" page, Hogan writes:
http://www.jamesphogan.com/books/twoface/background.shtml?baen97
...
One morning over breakfast I said, "Who do we know who's into
machine intelligence? I need to bounce my thoughts off somebody."
Lyn replied, "We've only been here a couple of months. We don't
really know anybody." Literally the next day, the phone rang and
a voice said, "Hi. We haven't met, but my name is Marvin Minsky.
I run the AI lab at MIT. Nobody's produced a good book on AI yet.
Judy-Lynn Del Rey told me to read Inherit the Stars. So I did,
and I think you could write one. How would you like to come to
the lab and see what we're doing, and let's talk about it?" Which
was how Marvin and I got to know one another.
No *wonder* the Lisp looked plausible! ;-}
-Rob
-----
Rob Warnock <rp...@rpw3.org>
627 26th Avenue <URL:http://rpw3.org/>
San Mateo, CA 94403 (650)572-2607
Tim X
By the later part of the 70's, it was recognized by most psychologists
that IQ tests were not measuring intelligence pe se and in fact are
vulnerable to cultural, social and economic variance (and most likely
gender). In fact, thre is some confusion about what IQ tests really do
measure. It is likely there is some level of correlation between high
IQ scores and intelligence, but that does not necessarily mean low IQ
test results also indicate lower intelligence. There has also been
evidence that developing an IQ test which is able to avoid cultural,
socio-economic and gender bias is extremely difficult, if not
impossible.
[note that the above is based on data from the early 80's when I did
my psych degree and therefore is quite outdated. However, I suspect
the general theme is still pretty valid]
Now there are also a lot of questions concerning the concept of
race. DNA studies have so far failed to find any common genetic
markers which are common to one race and not found in others. This is
leading some researchers to suggest the concept of race is itself an
artificial construct.
While there are certainly fewer women currently in the industry, we
certainly should not be surprised when they do pop up. Keep in mind
the first person to write programs for Babbage's machine was Ada
Lovelace and I believe the first compiler was written by a woman
(forgotten the name at present).
Tim
--
Tim Cross
The e-mail address on this message is FALSE (obviously!). My real e-mail is
to a company in Australia called rapttech and my login is tcross - if you
really need to send mail, you should be able to work it out!
Pascal Bourguignon
Thank you.
--
__Pascal Bourguignon__ http://www.informatimago.com/
In deep sleep hear sound,
Cat vomit hairball somewhere.
Will find in morning.
Robert Uhl
>
> Too bad it's in bioinformatics, though... an interesting subject to be
> sure, but wherever there is bioinformatics there is a DARPA or other
> .mil name not too far away. But hey, we shouldn't worry about that!
> It's not as if anyone in power thinks the world could do with a good
> culling.
It's not as if DARPA didn't directly contribute to the development of
the Internet or anything...
I find your anti-military remarks and .sig offensive; one of my brothers
is a naval officer, as is his wife; another brother has enlisted in the
Marine Corps; my father was a naval officer; his father (my grandfather)
was a sailor in the Second World War; his brother (my great-uncle) was a
Marine who died on Iwo Jima; his father (my great-grandfather) was an
Army officer in the Great War; his father was a soldier in the Civil
War. The career of a soldier is an honourable one; so too is that of
one who serves the military in non-combat roles--such as one who writes
code for a DARPA-funded project.
--
Robert Uhl <http://public.xdi.org/=ruhl>
Aquarion> Naming scheme is simple. You set it up, you name it.
GB> I like that. That's so... so... honourable.
--on host naming
Robert Uhl
>
> I beleive, as I stated, that IQ is not a 'very direct' indicator of
>programming ability. I'd expect there to be a rough correlation.
I'd expect it to be a fairly strong correlation: programming involves
quite a bit of symbol manipulation and such. Granted, there are plenty
of intelligent folks who can't program to save their lives...
> But the whole argument here appears to be that slight differences in
> average IQ are supposed to account for the lower number of female
> programmers.
The differences in standard deviation are hardly slight; they're rather
pronounced.
--
Robert Uhl <http://public.xdi.org/=ruhl>
If your adversary is badly bunkered, there is no rule against your
standing over him and counting his strokes aloud, but it will be a wise
precaution to arm yourself with the niblick before doing so, so as to
meet him on equal terms. --Horace G. Hutchinson, 1886
Pascal Bourguignon
> By the later part of the 70's, it was recognized by most psychologists
> that IQ tests were not measuring intelligence pe se and in fact are
> vulnerable to cultural, social and economic variance (and most likely
> gender). In fact, thre is some confusion about what IQ tests really do
> measure. It is likely there is some level of correlation between high
> IQ scores and intelligence, but that does not necessarily mean low IQ
> test results also indicate lower intelligence. There has also been
> evidence that developing an IQ test which is able to avoid cultural,
> socio-economic and gender bias is extremely difficult, if not
First you'd have to define intelligence. If you don't know what you
measure, how can you measure it? On the other hand, some define the
intelligence as what is measured by IQ tests.
You hardly can define the temperature of a single molecule. More, you
hardly can measure the temperature of a volume of gaz by observing
only the effect on one molecule of the enclosing recipient. To define
the temperature, you need to take a statistical approach, and to
measure it, you need to observe the effect on a great amount of
matter in your instrument.
Similarly, you hardly can measure intelligence by one IQ test. But
you can determine it fiably enough if you run several different IQ
tests. And having measured it, you won't know much more about the
behavior of the only particule^W human. But you can still make
precise prediction statistically on a population.
So when you measure repeatitively with various tests that the
Azkhenasis get higher IQ than the Africans, or when you measure
repeatitively that the variance of IQ of men is bigger than that of
women, while you cannot say anything about one particular individual,
you can still make precise predictions on groups.
Now that there have been important migrations both from Europe to
Africa, and now from Africa to Europe you can even eliminate about all
the variables to make correct correlations and conclusions...
> While there are certainly fewer women currently in the industry, we
> certainly should not be surprised when they do pop up. Keep in mind
> the first person to write programs for Babbage's machine was Ada
> Lovelace and I believe the first compiler was written by a woman
> (forgotten the name at present).
COBOL was designed by Col. Grace Hoper (well she was Captain when she did it).
--
__Pascal Bourguignon__ http://www.informatimago.com/
The mighty hunter
Returns with gifts of plump birds,
Your foot just squashed one.
Tim X
> "justinhj" <just...@gmail.com> writes:
> > Anyway this cannot be discussed reasonably on this group as I've
> > already been personally attacked for my opinon, and the thread should
> > really be about celebrating a females achievements rather than arguing
> > why it is unusual.
>
> If it wasn't unusual, there'd be nothing to celebrate!
>
All of this is totally bogus as it assumes that what an IQ test
measures is actually IQ, but this has never been conclusively
established. In fact, the whole notion of intelligence is poorly
defined. This thread also assumes the defining attribute regarding
programmers and non-programmers is directly linked to intelligence,
which at the very least, seems arrogant.
In the late 50's and early 60's, IQ tests were used to show how whites
were more intelligent than blacks. Later, it was shown the differences
were not due to intelligence, but rather due to differences in
cultruarl and socio-economic backgrounds. A lot of work was then put
into trying to develop questionaires that avoided phrases and terms
which were culturally derived etc. However, later studies also found
that things like spacial and geometric reasoning was also influenced
by cultural differences - for example, studies using geometric
puzzles and contradictions (i.e. think Escher) were not recognized by
some people from African tribes which lived in round huts. It is
thought this was because many of these geometric puzzles relied on
angles and straight lines, something which is rarely seen in nature
and unfamiliar to some tribes, while in industrialized countries with
urban build up, they are everywhere. Someone who is use to seeing how
straight lines and angles interact around them would immediately
comprehend some contradiction in an Escher painting, but someone who
rarely saw straight lines, right angles etc, was less inclined to see
any contradiction or puzzle.
As mentioned in this thread, some IQ test based studies have observed
differences in ability with problems relating to hand-eye coordination
and spacial reasoning etc. Is this diffeence an indication of a
difference in intelligence or possibly just a measure of difference in
experience. Take as an example the type of activities male and female
children participate in. Could the fact you generally (in
western/industrialized cultures) find more male children playing ball
sports than female children have any effect/impact on hand-eye
coordination between male and female? Could higher spacial reasoning
and abstraction be seen in males because the types of problems and
activities males are encouraged to do allow them to develop greater
skill?
Of course, we can turn this on its head as well - we might argue that
boys play more ball sports because they naturally have better hand-eye
and spacial reasoning skills than girls and that as children, we
gravitate towards those activities which we have a natural ability
for.
Personally, I don't know the answer. However, I do think we need to be
extremely careful about using something like a measure of IQ as the
basis for any argument when there exists considerable debate regarding
what IQ tests actually measure or even what intelligence really is.
As pointed out by others, its possible the lower numbers of women that
work as programmers is unrelated to intelligence. In fact, I find it
pretty arrogant to argue that intelligence is the differentiator for
who is an is not a programmer. For example it could just as
legitimately be argued that the difference is due to social and
communication skills. This is nothing new. Many have argued that boys
have poorer communication skills on average than girls at the same age
and it has often been argued that many of the males who get into
computing do so because they feel uncomfortable in social situations
and prefer the controlled predictable interaction with a computer. The
difference between the number of males and females who become
programmers could actually just be due to females having higher social
and communication skills and therefore prefer occupations which
involve higher interaction with real people and are simply not
attracted to jobs involving long periods of isolation and interaction
with a machine.
Pascal Bourguignon
> Pascal Bourguignon <use...@informatimago.com> writes:
>
>> "justinhj" <just...@gmail.com> writes:
>> > Anyway this cannot be discussed reasonably on this group as I've
>> > already been personally attacked for my opinon, and the thread should
>> > really be about celebrating a females achievements rather than arguing
>> > why it is unusual.
>>
>> If it wasn't unusual, there'd be nothing to celebrate!
>>
> All of this is totally bogus as it assumes that what an IQ test
> measures is actually IQ, but this has never been conclusively
> established.
> In fact, the whole notion of intelligence is poorly defined.
> However, later studies also found
> that things like spacial and geometric reasoning was also influenced
> by cultural differences - for example, studies using geometric
> puzzles and contradictions (i.e. think Escher) were not recognized by
> some people from African tribes which lived in round huts. It is
> thought this was because many of these geometric puzzles relied on
> angles and straight lines, something which is rarely seen in nature
> and unfamiliar to some tribes, while in industrialized countries with
> urban build up, they are everywhere. Someone who is use to seeing how
> straight lines and angles interact around them would immediately
> comprehend some contradiction in an Escher painting, but someone who
> rarely saw straight lines, right angles etc, was less inclined to see
> any contradiction or puzzle.
Your objections are inconsistent. You are assuming intelligence is
intrinsical to the individual. Since intelligence is poorly defined,
I don't know why you make such an assumption.
If you take a bushman and put him in NewYork, he'll probably behave
badly in that context. Conversely if you take a NewYorker in the
Bush, the NewYorker will probably look dumb in turn. Well, anytime I
can take a IQ test designed by the Bushmen. But there are no
computers to program in the bush. If we're interested in the
intelligence in the context of computer programming instead of that of
killing antelopes, I think it's fair to consider the usual IQ tests,
including the ones with Escher's trompe-l'oeuil.
> [...]
> Personally, I don't know the answer. However, I do think we need to be
> extremely careful about using something like a measure of IQ as the
> basis for any argument when there exists considerable debate regarding
> what IQ tests actually measure or even what intelligence really is.
I've very close to define intelligence as what is measured by an
intelligence test.
> As pointed out by others, its possible the lower numbers of women that
> work as programmers is unrelated to intelligence. In fact, I find it
> pretty arrogant to argue that intelligence is the differentiator for
> who is an is not a programmer.
Ah but the arrogance is justified by the IQ! :-)
--
__Pascal Bourguignon__ http://www.informatimago.com/
This is a signature virus. Add me to your signature and help me to live.
Alexander Schmolck
> Pascal Bourguignon <use...@informatimago.com> writes:
>
> > "justinhj" <just...@gmail.com> writes:
> > > Anyway this cannot be discussed reasonably on this group as I've
> > > already been personally attacked for my opinon, and the thread should
> > > really be about celebrating a females achievements rather than arguing
> > > why it is unusual.
> >
> > If it wasn't unusual, there'd be nothing to celebrate!
> >
>
> All of this is totally bogus as it assumes that what an IQ test measures is
> actually IQ, but this has never been conclusively established.
Apart from that this is of not the reason Pascal's claim is bogus, assuming
you meant "intelligence" rather than IQ in the second part of the sentence
then it has certainly been well established that IQ tests have all sorts of
desirable properties you'd like to see in something supposed to measure
"intelligence".
> In fact, the whole notion of intelligence is poorly defined. This thread
> also assumes the defining attribute regarding programmers and
> non-programmers is directly linked to intelligence, which at the very least,
> seems arrogant.
>
> In the late 50's and early 60's, IQ tests were used to show how whites were
> more intelligent than blacks. Later, it was shown the differences were not
> due to intelligence, but rather due to differences in cultruarl and
> socio-economic backgrounds.
Yes? Who did so? This report has rather good credentials:
6. The differential between the mean intelligence test scores of Blacks and
Whites (about one standard deviation, although it may be diminishing) does not
result from any obvious biases in test construction and administration, nor
does it simply reflect differences in socio-economic status. Explanations
based on factors of caste and culture may be appropriate, but so far have
little direct empirical support. There is certainly no such support for a
genetic interpretation. At present, no one knows what causes this
differential.
<http://www.lrainc.com/swtaboo/taboos/apa_01.html>
I broadly agree with the rest of your message though (and note that Pascal
still hasn't addressed the rather obvious objection that I and others have
raised).
> Personally, I don't know the answer. However, I do think we need to be
> extremely careful about using something like a measure of IQ as the
> basis for any argument when there exists considerable debate regarding
> what IQ tests actually measure or even what intelligence really is.
>
> As pointed out by others, its possible the lower numbers of women that work
> as programmers is unrelated to intelligence. In fact, I find it pretty
> arrogant to argue that intelligence is the differentiator for who is an is
> not a programmer. For example it could just as legitimately be argued that
> the difference is due to social and communication skills. This is nothing
> new. Many have argued that boys have poorer communication skills on average
> than girls at the same age and it has often been argued that many of the
> males who get into computing do so because they feel uncomfortable in social
> situations and prefer the controlled predictable interaction with a
> computer. The difference between the number of males and females who become
> programmers could actually just be due to females having higher social and
> communication skills and therefore prefer occupations which involve higher
> interaction with real people and are simply not attracted to jobs involving
> long periods of isolation and interaction with a machine.
'as
Tim X
I'm not assuming anything of the sort. What I am saying in very basic
terms is how do you know what the questions on the test are measuring?
How can you know they are actually measuring intelligence rather than
social/economic background/experiences? Even those who attempt to
define IQ tests have recognised this as a problem. The whole situation
is made worse because no consensus has been reached on what is
intelligence - for example your next reference to the bushman has
nothing to do with intelligence as much as experience and cultural
background.
> If you take a bushman and put him in NewYork, he'll probably behave
> badly in that context. Conversely if you take a NewYorker in the
> Bush, the NewYorker will probably look dumb in turn.
Exactly my point! The problem with IQ tests is ensuring that what is
being measured is intelligence and not some other
skill/ability/experience. The bushman is not necessarily less
intelligent because they have problems dealing with New York and for
all we know, the bushman may actually be the greatest programmer the
world has ever seen once they are taught how to program.
> Well, anytime I
> can take a IQ test designed by the Bushmen. But there are no
> computers to program in the bush. If we're interested in the
> intelligence in the context of computer programming instead of that of
> killing antelopes, I think it's fair to consider the usual IQ tests,
> including the ones with Escher's trompe-l'oeuil.
Why? What makes you think there are questions in the IQ test
specifically designed to determine someones intelligence with respect
to programming?
Here I would argue you are confusing aptitude/ability with
intelligence. Just because an African bushman doesn't score well on a
test with questions assuming familiarity with American culture doesn't
mean he is less intelligent or would be less successful at programming
once he was tought.
> > [...]
> > Personally, I don't know the answer. However, I do think we need to be
> > extremely careful about using something like a measure of IQ as the
> > basis for any argument when there exists considerable debate regarding
> > what IQ tests actually measure or even what intelligence really is.
>
> I've very close to define intelligence as what is measured by an
> intelligence test.
>
You seem to still miss the point and I think have something akin to a
circular argument. If there is yet to be a consensus on what is
intelligence and if nobody has been able to prove that an IQ test is
actually measuring some vague notion of intelligence, then how can you
say that intelligence is what is measured by an intelligence
test. Just because you call a collection of questions an intelligence
test, that does not mean it is a test of intelligence.
> > As pointed out by others, its possible the lower numbers of women that
> > work as programmers is unrelated to intelligence. In fact, I find it
> > pretty arrogant to argue that intelligence is the differentiator for
> > who is an is not a programmer.
>
> Ah but the arrogance is justified by the IQ! :-)
Yes, and the right of slave owners to own slaves is justified by the
fact they own slaves
or
the right of wealthy people to control the lives of poor ones is
justified by them being wealthy
or
the right of the rulers to rule is justified by the fact they are
rulers
or
......
Ulrich Hobelmann
> I find your anti-military remarks and .sig offensive; one of my brothers
> is a naval officer, as is his wife; another brother has enlisted in the
> Marine Corps; my father was a naval officer; his father (my grandfather)
> was a sailor in the Second World War; his brother (my great-uncle) was a
> Marine who died on Iwo Jima; his father (my great-grandfather) was an
> Army officer in the Great War; his father was a soldier in the Civil
> War. The career of a soldier is an honourable one; so too is that of
> one who serves the military in non-combat roles--such as one who writes
> code for a DARPA-funded project.
It's not military per se that's a problem; it's the existence of totally
irresponsible leaders that unfortunately invariably end up in power at
some point, and abuse well-meaning soldiers to do Bad(tm) things.
I at least respect everybody who do the sometimes necessary, but mostly
bloody work ordered by some warmongering do-no-good. It's them who
carry the risk and the responsibility, when nobody in high office will.
--
Suffering from Gates-induced brain leakage...
Ulrich Hobelmann
>> While there are certainly fewer women currently in the industry, we
>> certainly should not be surprised when they do pop up. Keep in mind
>> the first person to write programs for Babbage's machine was Ada
>> Lovelace and I believe the first compiler was written by a woman
>> (forgotten the name at present).
>
> COBOL was designed by Col. Grace Hoper (well she was Captain when she did it).
Not sure if that ups the record for the women ;)
Bulent Murtezaoglu
[...]
RU> It's not as if DARPA didn't directly contribute to the
RU> development of the Internet or anything...
Or started the push for CL, or [for pre-DARPA years] the ENIAC or ...
RU> ... The career of a soldier is an honourable one; so
RU> too is that of one who serves the military in non-combat
RU> roles--such as one who writes code for a DARPA-funded project.
One couldn't easily avoid DARPA money even if one wished to. There's
an entire generation of academic CS types (American _and_ foreign like
myself) who got DARPA money directly or indirectly. While I don't
think there needs to be any deep dishonor attached to doing
DARPA-funded work, I am happy to see that people have reservations and
a dislike for it. With all due respect to your family's service, I
think categorical glorification of DARPA work is far more dangerous
than David T.'s attitude. Afterall, someone with an impeccable
military service record, the guy who headed the US administration that
created DARPA, also warned about something called the
military-industrial complex. I will quote:
"We must never let the weight of this combination endanger our
liberties or democratic processes. We should take nothing for
granted. Only an alert and knowledgeable citizenry can compel the
proper meshing of the huge industrial and military machinery of
defense with our peaceful methods and goals, so that security and
liberty may prosper together."[1]
Need I stress that if honour is going to be attached to obeying any
order or service/trade in pursuance of a directive I'd sooner have the
primary one be the 'take nothing for granted' order from this
particular retired general?
One reason that the world cheered the US over the Soviets or (for that
matter the Nazi's) was that it was advertised that Americans were free
people as opposed to drones in service of their government. I just
felt it was appropriate -- now that people are doubting it -- to
remind people that. I intend no personal offence.
cheers,
BM
Marcus Breiing
> First you'd have to define intelligence.
Or you could just factor analyze outcome data over a wide range of
tests and intellectual life activities. As it happens, there reliably
emerges a single common factor explaining considerably more of the
between-subject variance than any other factor. Now call that factor
"general intelligence" (or "psychometric g", or whatever, it doesn't
really matter) and start exploring it.
So, no, you don't have to define intelligence. It is just there, in
the data. You discover it.
A thing to keep in mind: IQ is a differential, intra-species
concept. It doesn't compare the intelligence of brick, dog, chimp and
man. A philosophical quest for a definition of "intelligence" in the
sense of "what makes us human" is orthogonal to IQ testing.
Marcus
Tim X
> Tim X <ti...@spamto.devnul.com> writes:
>
> > Pascal Bourguignon <use...@informatimago.com> writes:
> >
> > > "justinhj" <just...@gmail.com> writes:
> > > > Anyway this cannot be discussed reasonably on this group as I've
> > > > already been personally attacked for my opinon, and the thread should
> > > > really be about celebrating a females achievements rather than arguing
> > > > why it is unusual.
> > >
> > > If it wasn't unusual, there'd be nothing to celebrate!
> > >
> >
> > All of this is totally bogus as it assumes that what an IQ test measures is
> > actually IQ, but this has never been conclusively established.
>
> Apart from that this is of not the reason Pascal's claim is bogus, assuming
> you meant "intelligence" rather than IQ in the second part of the sentence
> then it has certainly been well established that IQ tests have all sorts of
> desirable properties you'd like to see in something supposed to measure
> "intelligence".
>
I did mean intelligence rather than IQ. I also agree that it has been
well established that IQ tests do have many of the desirable
properties you would like to see in something that is supposed to
measure intelligence. However, the existence of desirable properties
in a test doesn't mean the test is measuring what you think it is.
> > In fact, the whole notion of intelligence is poorly defined. This thread
> > also assumes the defining attribute regarding programmers and
> > non-programmers is directly linked to intelligence, which at the very least,
> > seems arrogant.
> >
> > In the late 50's and early 60's, IQ tests were used to show how whites were
> > more intelligent than blacks. Later, it was shown the differences were not
> > due to intelligence, but rather due to differences in cultruarl and
> > socio-economic backgrounds.
>
> Yes? Who did so?
I cannot give specific citations simply because it was over 20 years
ago when this topic made up part of a final year course on
experimental design and test creation. The dangers of cultural bias in
IQ tests was used as a case study and examples were given of papers
that were published in which the authors argued that consistently
lower averages in IQ tests for black americans were evidence that
their intelligence was lower and the difference was more than likely
due to genetic differences between black and white. Later other
studies showed that many of the questions had a bias towards white
american culture and a number of studies were done in which questions
where changed to use phrases and terms more common amongst balck
american groups. In these trials, the IQ scores for white americans
dropped while those for black americans increased. In fact, it was
studies like this which alerted many researches to the problems and
dangers of unknowingly introducing a bias if cultural differences are
overlooked. A lot of research then followed where new IQ tests were
developed which attempted to eliminate things like cultural bias.
> This report has rather good credentials:
>
> 6. The differential between the mean intelligence test scores of Blacks and
> Whites (about one standard deviation, although it may be diminishing) does not
> result from any obvious biases in test construction and administration, nor
> does it simply reflect differences in socio-economic status. Explanations
> based on factors of caste and culture may be appropriate, but so far have
> little direct empirical support. There is certainly no such support for a
> genetic interpretation. At present, no one knows what causes this
> differential.
>
> <http://www.lrainc.com/swtaboo/taboos/apa_01.html>
>
Yes, thats not a bad review of current (as of 10 years ago) research
in the field. It does vaguely make reference to the use of IQ tests to
argue for differences in race IQ scores being used to argue for a
genetic difference in intelligence in the 60's, but to be fair it also
states that while cultural bias may explain some difference, it does
not appear to explain the size of the difference.
One thing I did find interesting from that report is the inability of
any research to explain the increse in the average IQ result. As flynn
points out, if intelligence has actually increased as much as the
average has increased, we should have seen a major cultural
renaissance, which has not occured. Furthermorre, studies of asian
americans has consistently resulted in an average below 100, yet in
measures of socio-economic status, they are over represented compared
to white americans. In fact, it has been suggested that the average
white american IQ score would need to be around 120 for white
americans to have a proportionally equivalent representation. Yet it
is often argued that IQ scores are a good predictor of future
socio-economic status (but not a sole predictor).
As stated in another post, I do agree IQ tests do have some properties
which are desirable in a test which is supposed to give some indicator
of an individuals intelligence, but I think its very dangerous to
argue it is an actual measure of intelligence. It is very likely there
is a strong correlation, but that is not the same as stating it is a
measure of intelligence. It is also interesting to note that one point
the report does make is that IQ score comparisons between groups is of
less use than comparison between individuals, which would seem to
discount the basis of much of this thread i.e. females consistently
have lower IQ scores compared to men and that is sifficient to explain
why there are fewer women in programming than men.
I still have a major issue with any argument that suggests fewer women
are programmers because they have a lower average IQ as a group
because it follows that therefore there should be a higher IQ average
amongst male programmers than non-male programmers, which I've never
seen any research to collaborate and which I doubt.
Pascal Bourguignon
> As stated in another post, I do agree IQ tests do have some properties
> which are desirable in a test which is supposed to give some indicator
> of an individuals intelligence, but I think its very dangerous to
> argue it is an actual measure of intelligence.
We could consider that IQ tests only measure the "fitness" of the
individual (for the implied cultural context).
> It is very likely there
> is a strong correlation, but that is not the same as stating it is a
> measure of intelligence.
We could even completely ignore this notion of intelligence. Perhaps
it just doesn't exist, and only this "fitness" exists.
> It is also interesting to note that one point
> the report does make is that IQ score comparisons between groups is of
> less use than comparison between individuals, which would seem to
> discount the basis of much of this thread i.e. females consistently
> have lower IQ scores compared to men and that is sifficient to explain
> why there are fewer women in programming than men.
Then the IQ tests measure a better fitness of males than females for
the programming jobs. You cannot deny they do, can you? The numbers
witnessed result from a selection process, so it's natural that the
fitness be the important factor.
> I still have a major issue with any argument that suggests fewer women
> are programmers because they have a lower average IQ as a group
> because it follows that therefore there should be a higher IQ average
> amongst male programmers than non-male programmers, which I've never
> seen any research to collaborate and which I doubt.
Yes, males are more fit in a environment where to survive you need to
be good programmer, like in a programmer job.
You may think that this definition and this theory is somewhat
tautologic, but it allows us to measure and make predictions. Now if
you want to theorize about the formula that will give the IQ in
function of the culture, genes, environment and nutrition of an
individual or a group, go ahead.
On the other hand, this "intelligence" theory doesn't work: we cannot
find a measure, and we cannot make prediction, since whites are not
more "intelligent" than blacks, or in some groups, males are not more
"intelligent" than female, then we cannot explain anything about why
there are more male in programming than female, or more blacks in
jails than whites.
--
__Pascal Bourguignon__ http://www.informatimago.com/
Litter box not here.
You must have moved it again.
I'll poop in the sink.
Cameron MacKinnon
> I still have a major issue with any argument that suggests fewer women
> are programmers because they have a lower average IQ as a group
I think the claim that was made was that variance was higher in males,
though the average was the same. Also,
-----8<-----
Sex Differences
Most standard tests of intelligence have been constructed so that there
^^^^ ^^^^ ^^^^^^^^^^^
are no overall score differences between females and males. Some recent
studies do report sex differences in IQ, but the direction is variable
and the effects are small (Held, Alderton, Foley, & Segall, 1993; Lynn,
1994). This overall equivalence does not imply equal performance on
every individual ability. While some tasks show no sex differences,
there are others where small differences appear and a few where they are
^ ^^^ ^^^^^ ^^^^ ^^^
large and consistent.
^^^^^ ^^^ ^^^^^^^^^^
-----8<-----
> because it follows that therefore there should be a higher IQ average
> amongst male programmers than non-male programmers, which I've never
> seen any research to collaborate and which I doubt.
Did you actually mean this, or did you mean "...than among male
non-programmers"? I think your conclusion incorrect as written.
It seemed logical to me that people would notice some fairly obvious
things, viz:
- general IQ tests are composed of tests of specific sub skills,
- some of which may be more relevant to programming aptitude than others.
- Variance difference between the sexes in those subtasks are higher
than variance difference in the overall score.
Rather than exploring the original (and, IMHO interesting) question of
why programming is so male dominated, it's off down the path of more
boring and irrelevant questions, such as whether IQ tests accurately
measure intelligence and how they are biased culturally.
I found lots of interesting stuff by Googling the phrase "predictor of
programming ability," especially
"CORRELATES OF PROBLEM-SOLVING IN PROGRAMMING"
http://sunzi1.lib.hku.hk/hkjo/view/33/3300413.pdf
which contains statistics differentiated by sex.
Interestingly, researchers can't even seem to agree on whether skill in
mathematics is a good predictor of skill in programming.
Much as I find the topic fascinating and would like to devote more time
to its study, my copy of Lisp in Small Pieces just arrived, so I'll just
close with the quote which prefaces The Bell Curve (which I haven't yet
read, but hope to, as I judged that any book so incendiary was either
carefully researched or a loony screed, and likely the former. Who wants
a collection of anodyne books?) Not that this quote is directed at you,
Tim; I merely thought it apropos of the general tenor of the thread.
"There is a most absurd and audacious Method of reasoning avowed by some
Bigots and Enthusiasts, and through Fear assented to by some wiser and
better Men; it is this. They argue against a fair Discussion of popular
Prejudices, because, say they, tho' they would be found without any
reasonable Support, yet the Discovery might be productive of the most
dangerous Consequences. Absurd and blasphemous Notion! As if all
Happiness was not connected with the Practice of Virtue, which
necessarily depends upon the Knowledge of Truth."
- Edmund Burke, A Vindication of Natural Society
And here's to Samantha Kleinberg!
David Trudgett
Hello, Robert,
The intention of my remarks is not to offend, but to communicate the
truth to which few people want to listen because it disturbs them too
much. Perhaps people need to be disturbed, perhaps pacifism and
anarchism and Christianity do disturb people, but it is not my
intention to disturb. It seems, though, to be an unavoidable result of
telling the truth.
So, I am not happy, and in fact am saddened, if you were offended. To
offend was not and is not my design. Yours or anyone else's real or
imagined offence, however, is not a good enough reason to refuse to
speak the truth about something that is in the process of destroying
the world while we watch. On the other hand, this is a Lisp forum, and
so I do not intend to pursue this conversation further than may be
required to clarify my remarks.
Robert Uhl <eadm...@NOSPAMgmail.com> writes:
> David Trudgett <wpo...@zeta.org.au.nospamplease> writes:
>>
>> Too bad it's in bioinformatics, though... an interesting subject to be
>> sure, but wherever there is bioinformatics there is a DARPA or other
>> .mil name not too far away. But hey, we shouldn't worry about that!
>> It's not as if anyone in power thinks the world could do with a good
>> culling.
>
> It's not as if DARPA didn't directly contribute to the development of
> the Internet or anything...
If the development of Internet technology cost a single human life
(and we're not talking about accidents here) then it cost too much,
didn't it? Or do you disagree?
Technology is a two-edged sword, but technologists often forget and
like to forget the evil applications of their work that can be easily
foreseen. With Einstein and others, it was the atomic bomb ("We have
known sin"). Today, it is biotechnology and genetic engineering that
is possibly the major technological threat we are facing; and there is
the real possibility of bioweapons wiping out much or perhaps even all
of humanity. There is also, of course, the ongoing nuclear threat, and
a possible emerging nanotechnology threat. And the danger of these are
all much higher than the average person is willing to believe, because
the average person cannot comprehend how people in power think, that
human beings to them are just numbers, costs and benefits, a bovine
herd to be used, amused, controlled and culled as the elite see fit.
None of that even touches upon other major perils we face this
century, such as global environmental collapse, climate change, and so
on, which are also helped along by technology, institutionalised
greed, and rapid population growth. One's conclusion would have to be
that civilisation won't survive this century if the situation is
objectively assessed, and the survival of humanity itself is touch and
go. What would a person in power be considering doing if they knew
these facts, as they undoubtedly do? Hint: they have been doing it
since at least September 11, 2001. And we ain't seen nothing yet. Why
has Halliburton just been awarded a contract to build internment camps
in the U.S.? Oh, yes, they are for illegal immigrants, of course... or
perhaps, "to support the rapid development of new programs." [1] I
guess it's not important...
>
> I find your anti-military remarks and .sig offensive;
The sig contained a quotation of something Stan Goff is reported, in a
March 2005 newspaper article, to have said; to wit:
Many young soldiers enter military service having bought the whole
story about the US military as some liberatory force, and the
discovery of its true nature creates a traumatic sense of
dislocation and betrayal.
Stan Goff is a retired professional murderer who served in what is
called the U.S. Special Forces, so I imagine he might know what he is
talking about. He might also be spinning us a line for reasons to
which only he is privy. But I don't think so. In particular, one has
to note that he was talking about "many young soldiers" and not all of
them or even most of them. That is part of what makes the statement
ring true; I believe that half or more soldiers enter military service
for self-serving reasons that have nothing to do with "God and
country", and that most of these would have no reason to feel upset
about discovering the true nature of the military; some might even be
happy about it (in fact, I'm sure of it). So, if Stan had said 'most'
instead of 'many' I would have immediately known he was trying to put
one over.
On the balance of it, then, considering also what I already know of
the nature of the military as told by those with the insight, courage
and honesty to tell it as it is, I am reasonably persuaded that Stan
Goff is spot on with this particular assertion. Therefore, what he
said is most likely the truth by a reasonable standard of judgement.
So, if you are offended by that sig, then you are offended by the
truth, and there is nothing that I can do about that.
> one of my brothers is a naval officer, as is his wife; another
> brother has enlisted in the Marine Corps; my father was a naval
> officer; his father (my grandfather) was a sailor in the Second
> World War; his brother (my great-uncle) was a Marine who died on Iwo
> Jima; his father (my great-grandfather) was an Army officer in the
> Great War; his father was a soldier in the Civil War.
Nothing that I wrote referred to relatives of yours and you shouldn't
take it personally. I am certain that a percentage of military
personnel are sufficiently brainwashed to believe in what they are
doing. I am certain that many of them know they are no more than paid
murderers and love it. I am certain that a percentage take great pride
in what they have done in the military, especially those who haven't
had to learn first hand the real horrors of war and what it entails.
And I am certain that some wind up in the military by chance or by
means of coercion. I am also quite sure that I have absolutely no idea
which of these or other categories your relatives fall into. I'm
reasonably certain that you don't either.
Robert, all soldiers are paid to commit or to assist in mass
murder. That is a given. You may not believe that killing for a "good
enough" reason is murder, but you would be wrong. An individual
soldier may not believe he commits murder by dropping bombs on a
residential district or any other "designated enemy", but he does
nevertheless. How personally guilty he is, though, is a matter between
himself and God, and no one else is to judge him; not you, not me, not
anyone.
> The career of a soldier is an honourable one;
Murder cannot be considered an honourable career in any society that
is sane. Draw your own conclusions from that.
> so too is that of one who serves the military in non-combat
> roles--such as one who writes code for a DARPA-funded project.
We disagree again. Such people assist in murder, whether they realise
it or not.
Obviously, you will have to learn to deal with the fact that other
people have radically different beliefs and values, and you need to
get used to the fact that in a free society they will speak those
beliefs. Are our societies still free? It's enough to make you wonder,
isn't it?
To summarise,
(murderp (kill soldier person)) => t
(murdererp (every soldier)) => nil
(evil-person-p (every soldier)) => nil
(good-person-p (any soldier)) => t
(honourablep (soldier)) => nil
(honourablep (any (self-image soldier))) => t
Best regards, and happy Lisping,
David
--
David Trudgett
http://www.zeta.org.au/~wpower/
... it is easy to be blinded to the essential uselessness of them by
the sense of achievement you get from getting them to work at all. In
other words... their fundamental design flaws are completely hidden by
their superficial design flaws.
-- The Hitchhiker's Guide to the Galaxy, on the products
of the Sirius Cybernetics Corporation.
David Trudgett
> It's not military per se that's a problem; it's the existence of
> totally irresponsible leaders that unfortunately invariably end up in
> power at some point, and abuse well-meaning soldiers to do Bad(tm)
> things.
Bad leaders are certainly a problem, that is true. They are not the
only problem, however. We find ourselves in a society that is not sane
and which has been thoroughly corrupted by an ideology of violence and
exploitation, so much so that it is utterly impossible to go about any
normal day-to-day activity without being influenced and sullied by
that ideology. Even using computer technology to type this message,
and Internet technology to transmit it, or writing a program in Lisp
or Ada, or any one of dozens and dozens of things that we take for
granted now, means using something that was developed with a view to
killing people. This is how evil unavoidably affects each and every
one of our lives, and there is nothing we can do about it except to
continue the struggle in the hope that one day evil will be conquered
as Jesus promised.
David
--
David Trudgett
http://www.zeta.org.au/~wpower/
It is naively assumed that the fact that the majority of people share
certain ideas or feelings proves the validity of these ideas and
feelings. Nothing is further from the truth... Just as there is a
'folie a deux' there is a 'folie a millions.' The fact that millions
of people share the same vices does not make these vices virtues, the
fact that they share so many errors does not make the errors to be
truths, and the fact that millions of people share the same form of
mental pathology does not make these people sane.
-- Erich Fromm, The Sane Society, Routledge, 1955, pp.14-15
Wade Humeniuk
> Ulrich Hobelmann <u.hob...@web.de> writes:
>
>> It's not military per se that's a problem; it's the existence of
>> totally irresponsible leaders that unfortunately invariably end up in
>> power at some point, and abuse well-meaning soldiers to do Bad(tm)
>> things.
>
> Bad leaders are certainly a problem, that is true. They are not the
> only problem, however. We find ourselves in a society that is not sane
> and which has been thoroughly corrupted by an ideology of violence and
> exploitation, so much so that it is utterly impossible to go about any
> normal day-to-day activity without being influenced and sullied by
> that ideology. Even using computer technology to type this message,
> and Internet technology to transmit it, or writing a program in Lisp
> or Ada, or any one of dozens and dozens of things that we take for
> granted now, means using something that was developed with a view to
> killing people. This is how evil unavoidably affects each and every
> one of our lives, and there is nothing we can do about it except to
> continue the struggle in the hope that one day evil will be conquered
> as Jesus promised.
>
It is not just bad leaders, but there are a lot of bad followers these
days. Followers making poor choices of who to follow. It seems many
times people follow the insane and psychotic (anyone with a strong
opinion but potentially irrational view).
Wade
Tim X
> Tim X <ti...@spamto.devnul.com> writes:
>
> > As stated in another post, I do agree IQ tests do have some properties
> > which are desirable in a test which is supposed to give some indicator
> > of an individuals intelligence, but I think its very dangerous to
> > argue it is an actual measure of intelligence.
>
> We could consider that IQ tests only measure the "fitness" of the
> individual (for the implied cultural context).
>
Possibly, but what is that implied cultural context?
> > It is very likely there
> > is a strong correlation, but that is not the same as stating it is a
> > measure of intelligence.
>
> We could even completely ignore this notion of intelligence. Perhaps
> it just doesn't exist, and only this "fitness" exists.
Yes, that is possibly better, but of course, we then have issues of
fitness for what. Your whole argument still assumes this test of
IQ/fitness corresponds to ability to do programming, but that
correlation has not been established. It is generally accepted that IQ
tests do seem to be a good predictor for future educational levels and
to a lesser extent, future socio-economic status, but they are not
necessarily a good predictor of occupation or useful as an explination
of differences between gender representation within an occupation.
>
> > It is also interesting to note that one point
> > the report does make is that IQ score comparisons between groups is of
> > less use than comparison between individuals, which would seem to
> > discount the basis of much of this thread i.e. females consistently
> > have lower IQ scores compared to men and that is sifficient to explain
> > why there are fewer women in programming than men.
>
> Then the IQ tests measure a better fitness of males than females for
> the programming jobs. You cannot deny they do, can you? The numbers
> witnessed result from a selection process, so it's natural that the
> fitness be the important factor.
>
Yes I can deny it! Your argument breaks most of the standard
principles for imperical analysis. For starters, you are drawing a
positive conclusion from testing of a positive hypothesis and have
created a pretty clear example of the dangers of the syllogism i.e.
p1 = More males program than females
p2 = Males score higher in IQ tests than females
c1 = therefore IQ scores explain why mor men program than women.
On the basis of your arguments, I guess you would also agree that as
there has been such a rapid increase in the number of indian/asian
programmers, the IQ scores of indian/asian males must be higher than
those of white males?
>
> > I still have a major issue with any argument that suggests fewer women
> > are programmers because they have a lower average IQ as a group
> > because it follows that therefore there should be a higher IQ average
> > amongst male programmers than non-male programmers, which I've never
> > seen any research to collaborate and which I doubt.
>
> Yes, males are more fit in a environment where to survive you need to
> be good programmer, like in a programmer job.
>
If this were true, you would see more programmers in the top
percentiles than any other occupation. There is nothing in IQ scores
that provides such evidence. For all we know, all programmers may
actually centre around the middle percentile - its just arrogance to
assume programmers are in the top percentiles.
> You may think that this definition and this theory is somewhat
> tautologic, but it allows us to measure and make predictions. Now if
> you want to theorize about the formula that will give the IQ in
> function of the culture, genes, environment and nutrition of an
> individual or a group, go ahead.
>
Sure, IQ tests do allow us to make predictions, but not predictions on
careers - that is a different type of test and not the same as IQ
tests. It has also been argued that IQ tests only measure one dimension
of intelligence and there are other dimensions. It is possible these
other non-mesured dimensions have even greater predictive power with
respect to aptitude as a programmer.
> On the other hand, this "intelligence" theory doesn't work: we cannot
> find a measure, and we cannot make prediction, since whites are not
> more "intelligent" than blacks, or in some groups, males are not more
> "intelligent" than female, then we cannot explain anything about why
> there are more male in programming than female, or more blacks in
> jails than whites.
Good god, now your throwing out the baby with the bath water - If IQ
tsts are not able to explain the difference between numbers of male
and female programmers, then it cannot be explained.
couldn't it possibly, just maybe, a slim chance, a mere thread of
possibility, be that notions of intelligence alone are not sufficient
to explain the difference and that we need to broaden the possible
contributors, such as considering some of the suggestions relating to
role models, socialisation, environment, opportunity etc mentioned by
others in this thread?
Pascal, be honest, you really are just bored and decided to do a bit
of your own trolling in c.l.l to see how far you could push this
thread didn't you? I can't believe even you truely agrees with your
own argument here and now feel a need to kick myself for being
so easily sucked in.
Tim X
> Tim X wrote:
> > I still have a major issue with any argument that suggests fewer women
> > are programmers because they have a lower average IQ as a group
>
> I think the claim that was made was that variance was higher in males,
> though the average was the same. Also,
>
> -----8<-----
> Sex Differences
>
> Most standard tests of intelligence have been constructed so that there
> ^^^^ ^^^^ ^^^^^^^^^^^
> are no overall score differences between females and males. Some
> recent studies do report sex differences in IQ, but the direction is
> variable and the effects are small (Held, Alderton, Foley, & Segall,
> 1993; Lynn, 1994). This overall equivalence does not imply equal
> performance on every individual ability. While some tasks show no sex
> differences, there are others where small differences appear and a few
> where they are
> ^ ^^^ ^^^^^ ^^^^ ^^^
> large and consistent.
> ^^^^^ ^^^ ^^^^^^^^^^
> -----8<-----
But how relevant is that large and consistent difference to
programming ability?
> > because it follows that therefore there should be a higher IQ average
> > amongst male programmers than non-male programmers, which I've never
> > seen any research to collaborate and which I doubt.
>
> Did you actually mean this, or did you mean "...than among male
> non-programmers"? I think your conclusion incorrect as written.
>
I meant among male non-programmers! Damn!
> It seemed logical to me that people would notice some fairly obvious
> things, viz:
> - general IQ tests are composed of tests of specific sub skills,
> - some of which may be more relevant to programming aptitude than others.
> - Variance difference between the sexes in those subtasks are higher
> than variance difference in the overall score.
>
> Rather than exploring the original (and, IMHO interesting) question of
> why programming is so male dominated, it's off down the path of more
> boring and irrelevant questions, such as whether IQ tests accurately
> measure intelligence and how they are biased culturally.
Agreed. Personally, I tend to believe intelligence differences between
male and female is the least likely explination for differences in
numbers between male and female programmers.
> I found lots of interesting stuff by Googling the phrase "predictor of
> programming ability," especially
> "CORRELATES OF PROBLEM-SOLVING IN PROGRAMMING"
> http://sunzi1.lib.hku.hk/hkjo/view/33/3300413.pdf
> which contains statistics differentiated by sex.
>
> Interestingly, researchers can't even seem to agree on whether skill
> in mathematics is a good predictor of skill in programming.
Yes, talk about the influence of trends/fashion. When I first started
geting into programming in the late 70's, the generally held wisdom
was that you cannot be a programmer if you were not good at maths. I
then slowly saw this change until now at a number of Universities I
know of, maths is no longer a prerequisite for enroling in comp. sci
and you don't even have to do any maths units to obtain your degree.
Personally (based on no empirical evidence), I believe that a good
math grounding will certainly increase the likelihood of someone being
able to program, but you can also be a good programmer without a math
background. I do feel its a mistake to have comp. sci degrees in which
ou don't hav to do any maths - at the very least a first year applied
maths course should be required.
> Much as I find the topic fascinating and would like to devote more
> time to its study, my copy of Lisp in Small Pieces just arrived, so
> I'll just close with the quote which prefaces The Bell Curve (which I
> haven't yet read, but hope to, as I judged that any book so incendiary
> was either carefully researched or a loony screed, and likely the
> former. Who wants a collection of anodyne books?) Not that this quote
> is directed at you, Tim; I merely thought it apropos of the general
> tenor of the thread.
>
> "There is a most absurd and audacious Method of reasoning avowed by
> some Bigots and Enthusiasts, and through Fear assented to by some
> wiser and better Men; it is this. They argue against a fair Discussion
> of popular Prejudices, because, say they, tho' they would be found
> without any reasonable Support, yet the Discovery might be productive
> of the most dangerous Consequences. Absurd and blasphemous Notion! As
> if all Happiness was not connected with the Practice of Virtue, which
> necessarily depends upon the Knowledge of Truth."
> - Edmund Burke, A Vindication of Natural Society
I don't disagree (assuming I'm interpreting the above correctly). I
believe concerns of political correctness are extremely dangerous
because it seems to make some important areas of debate "no go"
zones. to a large degree, political correctness and scientific
investigation can often conflict and lately, political correctness
does seem to be winning in some 'sensitive' areas.
I'm more inclined to go with Samuel Clements - I may not believe in
what you say, but I will defend to the death your right to say it".
Tim
>
> And here's to Samantha Kleinberg!
Here here.
Pascal Bourguignon
> Pascal Bourguignon <use...@informatimago.com> writes:
>
>> Tim X <ti...@spamto.devnul.com> writes:
>>
>> > As stated in another post, I do agree IQ tests do have some properties
>> > which are desirable in a test which is supposed to give some indicator
>> > of an individuals intelligence, but I think its very dangerous to
>> > argue it is an actual measure of intelligence.
>>
>> We could consider that IQ tests only measure the "fitness" of the
>> individual (for the implied cultural context).
>
> Possibly, but what is that implied cultural context?
The point is that IQ, or the measured fitness is not a function of the
individual, but a function of the (individual,environment).
Fishes are quite fit to live underwater. Homo sapiens not.
Or if you want to keep it single species:
NewYorkers are quite fit to live in a big city, bushmen not.
So we expect that fitness(NewYorker,NewYork) > fitness(bushman,NewYork)
which we usually express as a NewYorker has a greater IQ than a bushman.
But of course, as true is that:
fitness(bushman,Bush) > fitness(NewYorker,Bush)
which we usually express as a bushman has a greater IQ than a NewYorker
(if you don't believe me, just ask a NewYorker to bring something to
eat in the middle of the Bush).
> Yes, that is possibly better, but of course, we then have issues of
> fitness for what. Your whole argument still assumes this test of
> IQ/fitness corresponds to ability to do programming, but that
> correlation has not been established.
Please, be serrious. Present me a programmer (and his program!) with an IQ<90.
> If this were true, you would see more programmers in the top
> percentiles than any other occupation. There is nothing in IQ scores
> that provides such evidence. For all we know, all programmers may
> actually centre around the middle percentile - its just arrogance to
> assume programmers are in the top percentiles.
The problem is that the current IQ tests merge several environments.
You can only demonstrate negative results: there's no MD with an
IQ<90, there's no programmer with an IQ<90, there's no more than 20%
of women in programming.
> Sure, IQ tests do allow us to make predictions, but not predictions on
> careers - that is a different type of test and not the same as IQ
> tests. It has also been argued that IQ tests only measure one dimension
> of intelligence and there are other dimensions. It is possible these
> other non-mesured dimensions have even greater predictive power with
> respect to aptitude as a programmer.
Indeed. That's why you need to take into account explictely the
environment argument of the fitness function, and forget about the
current IQ tests that select and merge serveral tests for a biased
environment.
>> On the other hand, this "intelligence" theory doesn't work: we cannot
>> find a measure, and we cannot make prediction, since whites are not
>> more "intelligent" than blacks, or in some groups, males are not more
>> "intelligent" than female, then we cannot explain anything about why
>> there are more male in programming than female, or more blacks in
>> jails than whites.
>
> Good god, now your throwing out the baby with the bath water - If IQ
> tsts are not able to explain the difference between numbers of male
> and female programmers, then it cannot be explained.
Well using the experimental data given by the IQ test (the difference
of variance between male and female IQ, I think I've explained why
this difference in numbers exist: for the simple reason that the mean
IQ of the programmer population is greater than 100.
> couldn't it possibly, just maybe, a slim chance, a mere thread of
> possibility, be that notions of intelligence alone are not sufficient
> to explain the difference and that we need to broaden the possible
> contributors, such as considering some of the suggestions relating to
> role models, socialisation, environment, opportunity etc mentioned by
> others in this thread?
We could get more precise results if we used a more specific test than
IQ, but the phenomena can already be explained just with IQ.
Anybody has statitics of IQ per activity?
--
__Pascal Bourguignon__ http://www.informatimago.com/
Pascal Bourguignon
> Agreed. Personally, I tend to believe intelligence differences between
> male and female is the least likely explination for differences in
> numbers between male and female programmers.
Just to make sure there's no misunderstanding, in my theory, the
average IQ of male programmers and the average IQ of female
programmers are the same, and is different from 100.
(And since the variance for female IQ is less than the variance for
male IQ, it comes directly that the number of female programmers must
be less than that of male programmers).
(If you fancy the idea, it's perfectly possible that the average IQ of
programmers is less than 100, you'd get the same ratio of female
programmers vs. male programmers).
--
__Pascal Bourguignon__ http://www.informatimago.com/
HEALTH WARNING: Care should be taken when lifting this product,
since its mass, and thus its weight, is dependent on its velocity
relative to the user.
Ulrich Hobelmann
> Bad leaders are certainly a problem, that is true. They are not the
> only problem, however. We find ourselves in a society that is not sane
> and which has been thoroughly corrupted by an ideology of violence and
> exploitation, so much so that it is utterly impossible to go about any
> normal day-to-day activity without being influenced and sullied by
It's possible to keep your own mind. At least we're Lisp lovers in a
sea of boiling Java.
What's problematic is that corrupt leaders extract money from everybody
else and build weapons of mass destruction with it. If you oppose them,
you might yourself get a feel of the weaponry.
> that ideology. Even using computer technology to type this message,
> and Internet technology to transmit it, or writing a program in Lisp
> or Ada, or any one of dozens and dozens of things that we take for
> granted now, means using something that was developed with a view to
> killing people. This is how evil unavoidably affects each and every
No, the ARPA-net was only a means of communication, though funded by the
ARPA. The military also want to communicate, and the market didn't
provide anything at the time (that they liked).
You shouldn't confuse tech funded by organizations you don't like with
evil things those organizations might do. The organization (well, its
leadership) and its source of funding might be questionable, but the
internet isn't per se bad.
> one of our lives, and there is nothing we can do about it except to
> continue the struggle in the hope that one day evil will be conquered
> as Jesus promised.
Only I don't think evil will be conquered. But we can stay as far from
it as we can. Don't do Java, unless you want it done to you ;)
justinhj
Cameron MacKinnon wrote:
> justinhj wrote:
> [earlier]
> > I am surpised at the attitudes here I thought the world has moved on,
> > and I'm glad to say it has in the places I have worked and studied.
>
> ...then demonstrated very low reading comprehension skills, and finally
There you go again demonstrating your social skills to the fullest
> > Anyway this cannot be discussed reasonably on this group as I've
> > already been personally attacked for my opinon,
>
> Yep, if you start off by calling people here troglodytes, expect a warm,
> warm welcome.
Pascals comment was provocative and I replied honestly. If you honestly
think I'm idiot who can't read then I'll leave it to you and the other
posters to decide that.
But I don't think anyone has succesfully argued Pascal's original point
that differences in the standard deviation of peoples IQ's explains why
there are so few female programmers.
Social factors are far more likely to play a role.
Justin