Home > Archive > Scheme > June 2004 > Quick and Dirty Scheme Propaganda Available
You are viewing an archived Text-only version of the thread.
To view this thread in it's original format and/or if you want to reply to
this thread please [click here]
| Author |
Quick and Dirty Scheme Propaganda Available
|
|
| Terrence Brannon, Scheme Hacker 2004-05-12, 9:12 pm |
| http://www-lnc.usc.edu/~brannon/sch...each-scheme.pdf
is the URL to a set of slides I plan to use during my interview to teach high
school math. I want to start a programming club or perhaps teach a class using
Scheme.
The slides give what I think is valid motivation for this.
Please respond with any suggestions you might have for the slides.
| |
| Faried Nawaz 2004-05-24, 6:36 am |
| "Terrence Brannon, Scheme Hacker" <metaperl@urth.org> writes:
> http://www-lnc.usc.edu/~brannon/sch...each-scheme.pdf
Why is there so much emphasis on "females"?
slide 3: Proven results with females and beginners
slide 4: Females show 4:1 preference for Scheme
Faried.
--
continuations are hard. let's go []! -- 'barbie
| |
| taltman@noshpam.lbl.government 2004-05-24, 6:36 pm |
| "Faried Nawaz" <faried@gmail.com> writes:
> "Terrence Brannon, Scheme Hacker" <metaperl@urth.org> writes:
>
>
> Why is there so much emphasis on "females"?
Probably because females are underrepresented in computer science
courses, so any system which brings the representation closer to 50%
is preferable (barring other considerations). This is an important
consideration in any introductory course, since you don't want to
inadvertently bias your department by deterring certain groups at the
onset.
~Tomer
>
> slide 3: Proven results with females and beginners
> slide 4: Females show 4:1 preference for Scheme
>
>
> Faried.
> --
> continuations are hard. let's go []! -- 'barbie
--
()
| |
| Ray Dillinger 2004-06-03, 7:27 pm |
| taltman@noshpam.lbl.government wrote:
> "Faried Nawaz" <faried@gmail.com> writes:
>
>
>
> Probably because females are underrepresented in computer science
> courses, so any system which brings the representation closer to 50%
> is preferable (barring other considerations). This is an important
> consideration in any introductory course, since you don't want to
> inadvertently bias your department by deterring certain groups at the
> onset.
I don't think I agree. I think any system which produces a
greater number of more competent graduates and makes the work
accessible to a greater number of students is preferable, and
gender diversity among those students is at most a side issue.
It's important to find good, efficient ways to reach and teach
students that don't benefit from "standard" teaching methods
for whatever reason, but I don't think that gender constitutes
such a reason. I would be talking about personality types,
learning styles, values, study habits, and degree of mathematical
vs. verbal background instead. These are frequently determining
factors in someone's success in learning computer science.
Gender is irrelevant by comparison, however sensitized our
society may have become to it. Considering it in isolation
or as a goal in itself blinds us to more important issues.
Bear
| |
| Alex Shinn 2004-06-03, 7:27 pm |
| At Wed, 26 May 2004 03:32:03 GMT, Ray Dillinger wrote:
>
> Gender is irrelevant by comparison, however sensitized our
> society may have become to it. Considering it in isolation
> or as a goal in itself blinds us to more important issues.
Gender balance is important for a happy and productive environment.
Considering pure teaching efficiency in isolation or as a goal in itself
blinds us to more important issues like physical, mental and social
health.
At least until humans stop thinking for themselves and we switch to
machines teaching machines :)
--
Alex
| |
| Anton van Straaten 2004-06-03, 7:27 pm |
| Ray Dillinger wrote:
> taltman@noshpam.lbl.government wrote:
>
> I don't think I agree. I think any system which produces a
> greater number of more competent graduates and makes the work
> accessible to a greater number of students is preferable, and
> gender diversity among those students is at most a side issue.
>
> It's important to find good, efficient ways to reach and teach
> students that don't benefit from "standard" teaching methods
> for whatever reason, but I don't think that gender constitutes
> such a reason. I would be talking about personality types,
> learning styles, values, study habits, and degree of mathematical
> vs. verbal background instead. These are frequently determining
> factors in someone's success in learning computer science.
>
> Gender is irrelevant by comparison, however sensitized our
> society may have become to it.
Except that if teaching styles, values, etc. have all been geared towards
what works for males - which seems quite likely, and also has quite a bit of
evidence to support it - then that would make females the biggest single
group which is not having its learning styles etc. catered to. That's
assuming there are learning factors common to a majority of females, but
there's evidence for that, too.
> Considering it in isolation or as a goal in
> itself blinds us to more important issues.
I don't think it is considered in isolation. It's just that the entire set
of issues is incredibly large and complex, and the importance of teaching
strategies which work well for females are understood by those familiar with
the issues.
Besides, if "females show 4:1 preference for Scheme", perhaps that's an
indication that the prevailing mainstream languages have actually been
inflicted on us by some strange quirk of the male psyche, like not wanting
to ask for directions. How else would you explain Perl?
Anton
| |
| Neil W. Van Dyke 2004-06-03, 7:27 pm |
| "Faried Nawaz" <faried@gmail.com> writes:
> "Terrence Brannon, Scheme Hacker" <metaperl@urth.org> writes:
>
> Why is there so much emphasis on "females"?
>
> slide 3: Proven results with females and beginners
> slide 4: Females show 4:1 preference for Scheme
I think the emphasis is there because gender imbalance and bias in
science and engineering education have been regarded as a problems for
years.
(My own inexpert suspicion is that gender bias causes do exist, but that
the biggest cause is a gender-independent problem of motivating students
to learn despite difficulty and frustration. I suspect that gender bias
in cultural conditioning, classroom dynamics, preparatory education,
etc., exacerbates that. I'm skeptical of the poorly-supported
assertions I've seen that female and male humans have innate major
differences in how they learn subjects like computer programming.)
Back on the topic of Scheme in general, I'd be surprised to see studies
that showed a strong student preference for language A over language B,
*all other variables controlled*. That's a very expensive study, if not
an impossible one.
What is much easier to say is that, after being through a course in
which students were exposed in some manner to both A and B, a
significant majority of students polled professed a preference for A.
That could mean that the whole of the program involving A left a more
favorable impression than that involving B, which is useful info. I
would guess that the "4:1 preference" factoid was based on a study
similar to that. No doubt, the talk accompanying the slides elucidates
on the bullet-point.
Shriram or someone has a bit of tongue-in-ch marketing puffery that I
think had some truth to it, and reflects a consideration that should be
central in HS and higher education (I forget the exact percentage, and
am taking it out of context): "Scheme makes you 37% smarter!"
--
http://www.neilvandyke.org/
| |
| Bruce Lewis 2004-06-03, 7:27 pm |
| Ray Dillinger <bear@sonic.net> writes:
> I would be talking about personality types,
> learning styles, values, study habits, and degree of mathematical
> vs. verbal background instead. These are frequently determining
> factors in someone's success in learning computer science.
How would you go about measuring these factors? Would you use
self-reporting as the basis for a statistic such as "people with
thus-and-such learning style prefer Scheme 4:1"?
Gender, race, religion, etc. certainly aren't perfect measures of
academic diversity. However, my personal observations are of some
correlation between being male and preferring a linear problem-solving
approach, and betwen being female and preferring a non-linear
problem-solving approach.
| |
| Terrence Brannon, Scheme Hacker 2004-06-03, 7:27 pm |
| Faried Nawaz wrote:
> "Terrence Brannon, Scheme Hacker" <metaperl@urth.org> writes:
>
>
>
>
> Why is there so much emphasis on "females"?
Hmmm, I will be interviewing with a female principle.
Also, females have been said to perform less well in left-brain tasks, though
that is changing.
>
> slide 3: Proven results with females and beginners
> slide 4: Females show 4:1 preference for Scheme
>
>
> Faried.
| |
| Jens Axel Søgaard 2004-06-03, 7:27 pm |
| Terrence Brannon, Scheme Hacker wrote:
> Also, females have been said to perform less well in left-brain tasks,
> though that is changing.
Careful. That sentence is ambigious.
--
Jens Axel Søgaard
| |
| Neil W. Van Dyke 2004-06-03, 7:27 pm |
| Jens Axel Søgaard <usenet@soegaard.net> writes:
> Terrence Brannon, Scheme Hacker wrote:
>
> Careful. That sentence is ambigious.
It is true: historically, there have been a lot of myths about innate
cognitive aptitude, which have been used to justify prejudicial
treatment based on gender and ethnicity -- and what is changing is that
those myths are rightfully being discarded. :)
I think that perpetuating those myths in well-meaning attempts at
inclusiveness is counterproductive. What I think *is* productive is
recognizing the influence of ill conditioning, and working to counteract
it.
| |
| Terrence Brannon, Scheme Hacker 2004-06-03, 7:27 pm |
| Neil W. Van Dyke wrote:
>
> It is true: historically, there have been a lot of myths about innate
> cognitive aptitude, which have been used to justify prejudicial
> treatment based on gender and ethnicity -- and what is changing is that
> those myths are rightfully being discarded. :)
Actually, if you study a basic book on neurobiology, you will note that females
lose less language ability when suffering brain damage to either hemisphere
because their linguistic capabilities are spread over both hemispheres.
So, differences in gender are supported by experimental evidence, not just
conjecture or prejudice.
| |
| Neil W. Van Dyke 2004-06-03, 7:27 pm |
| "Terrence Brannon, Scheme Hacker" <metaperl@urth.org> writes:
> So, differences in gender are supported by experimental evidence, not
> just conjecture or prejudice.
Obviously there are innate differences. I just don't currently believe
those differences have a major effect on net aptitude (or observable
learning method) for things like computer programming.
I believe the differences are in the noise, when we consider the major
influence of gender-independent pedagogic and learning factors (in
addition to genetic, developmental, and societal factors).
I'm willing to be persuaded otherwise, but I don't think the studies
have yet been performed (if they can ever be). My beliefs follow from
anecdotal observations.
Summarizing on a few points of this thread: yes, there has obviously
been gender imbalance in CS; no, I don't think that's driven by innate
gender predisposition; yes, I think that correcting that imbalance is
good, for many reasons; no, I'm not aware of clear evidence that there
is significant gender bias in the Scheme language itself, nor that any
such bias differs significantly from that of other programming
languages.
On the last summary point, I would *not* be surprised to see evidence
otherwise, and those results would inform new language design. I
*would* be surprised were any demonstrated gender bias in the Scheme
language to have significant external effect.
I didn't forsee considering gender issues of Scheme. :)
| |
| Anton van Straaten 2004-06-03, 7:27 pm |
| Neil W. Van Dyke wrote:
> I'm not aware of clear evidence that there
> is significant gender bias in the Scheme language itself
Given that Scheme is a fairly short hop from a pure lambda calculus, any
"significant gender bias in the Scheme language itself" would imply that
there's a significant gender bias in aspects of mathematics and in the
semantic construction of the universe.
Philip Wadler alludes to these sorts of issues when he talks about both
"God" and "space aliens" programming in (typed) lambda calculi. These
connections lead to the obvious conclusion that if Scheme has a gender bias,
it could help answer the all-important question of the gender of God. Both
Judaism and Christianity have struggled with this question [1], [2]. Who
would have thought that the answer could lie in programming language
research?
Anton
[1] http://www.leaderu.com/ftissues/ft9...cles/berke.html
[2]
http://www.religion-online.org/cgi-...icle?item_id=31
4
| |
| Anton van Straaten 2004-06-03, 7:27 pm |
| I wrote:
> [2]
>
http://www.religion-online.org/cgi-...icle?item_id=31
> 4
The URL wrap in this case is particularly egregious since the truncated URL
leads to a completely unrelated article. The URL should, of course, end
with: ...?item_id=314
Anton
|
|
|
|
|