Google Groups no longer supports new Usenet posts or subscriptions. Historical content remains viewable.
Dismiss

Soviet discrimination legacy in academia

35 views
Skip to first unread message

Alex 226-5350 LKG2-2/T2 29-Apr-1992 2225

unread,
Apr 29, 1992, 10:52:18 PM4/29/92
to
Being discriminated against is not unusual at the most prestigious Soviet
institutions of higher learning. Much has been written about discrimination
against Jews, a terrible injustice that each year cripples the lives of
thousands of gifted young men and women. Religious believers, the children of
dissidents, and, in Moscow's institutions of higher education, nonresidents of
the city are also subject to discriminatory treatment. Nonresidents actually
are required to obtain higher scores on entrance exams. Other forms of unequal
treatment are more subtle. Exam papers are identified by number, not by name,
but at Moscow University, at least, on key digit of the number alerts the
admissions board to "undesirable" candidates, who are routinely assigned to a
special group for oral exams (where arbitrary grading is easier to justify).
Members of these special groups are supposed to be flunked -- as the story
goes, one young Jew, seeing the other members of the group gathering with him
for the oral exam, exclaimed: "You mean they're taking us straight from here
to Auschwitz?"

Anti-semitic discrimination in university admissions is not just the work of
individual bigots, but part of a deliberate policy of squeezing Jews out of the
country's intellectual establishment.

Anti-semitism is just one aspect of the widespread anti-intellectual and the
caste prejudices of Soviet society. Decent people must be concerned about this
cruelly unjust policy.

Unused guest account

unread,
Apr 30, 1992, 2:06:10 PM4/30/92
to


In article <1992Apr29.1...@apollo.hp.com> goykh...@apollo.hp.com (Alex
Goykhman) writes:
>In article <1992Apr28.0...@cmcl2.nyu.edu> gu...@lab.ultra.nyu.edu
(Unused guest account) writes:
>>Written exams: writen works don't bear neither name of applicant, nor
>>his/her nationality. I don't know how to figure out who's writing
>>that work -- Jew or Uzbek.
> I once talked to a person who wrote software that determined
> if a NON-JEWISH applicant had traces of Jewish blood in his/her
> veins for Moscow Economic Institute. The software employed quite
> a sophisticated algorithm that analyzed first, middle, and last
> names of both the applicant and the applicant's parents. For that
> it had to keep a database of names along with probability of a
> particular name belonging to a Jew, and use some pattern-recognition
> technics for oddly-spelled names (who says the SU is a techologically
> backward country:).

To do such a programming job people must have very serious
reasons and very serious higher ups (nachal'niki). None of this I saw
in Sovok. But maybe, there were hiding.

Seriously, it's very unlikely that such software be ordered by
admission comission and trusted by it. Bureaucrats in Sovok don't trust
computers in general. To trust computers in such sensitive things
as admissions is wild and suicidal for them. What if nachal'nik's
son/daughter has Jewish-sounding name?

Then, is there anybody out there who really believes that by name
alone you can told a Jew from non-Jew? My Moscow experience makes me
to say: "No".


>>Oral exams: I didn't see senior examiners looking for nationality or such
>>things. Even more, the nationality is listed only in the anketa which
>>is not used at exams. More or less, we were filtering out guys that are
>>stupider than they should be for current situation.
> That depends on the institution. My "Alma Mater", Minsk
> Radioengineering Institute, used oral exams to filter the Jews out.
> For example, my oral math exam lasted 7+ hours while the average Aryan
> would get off easy in ~1.5-2 hours. Besides, the extra hours were
> mostly spent discussing college math curriculum. Later my "examiners"
> decided they went too far in their questioning and falsified my
> oral exam notes. As a result, I was admitted, and they fired
> (for their stupidity, not anti-semitism).

So it contradicts the theory that authorities were deadly antisemitic.
They had not to admit you and they had to protect their beloved antisemits at
ANY price. These are the rules.


>>Computers: I was working on computer that kept data on applicants
>>in VMK. I don't know how 1/4 Jew can be traced by those computers.
> By analyzing his/her parents' middle name.

This not the case with MGU and VMK. Vadim Antonov correctly states that
applicant `database' for VMK was developed by minority man (Uzbek). This
data base contained `nationality' field. But it could not and can not now
(it was not updated/corrected from 1985) analyze middle names.
The central data base (for the whole MGU) was much stupider and
could not make complicated searches at all. It was (and is?) running on
BESM-6 computer which was first introduced in Sovok in 1968.
So you can see how technologically advanced MGU (and Russia) is.

Disclaimers: I am not antisemit/zionist. Other disclaimers see at my
previous posting.

Vadim Maslov.

Dima Volodin

unread,
May 2, 1992, 3:49:53 PM5/2/92
to
Here we go again. Is it a coincidence, that "Antisemitism in MGU" is
a hot topic in spring? It looks like it's due to incurable avitaminosis,
which was blagopriobretion (anyone cares to translate it to English?)
by majority of active readers of this newsgroup back in the SU.

BTW, Vadim, kogda ty uspel svintit'? S avg kontachil?
--
Dima

Alex Allister Shvartsman, LKG2-2/T2 02-May-1992 2248

unread,
May 2, 1992, 11:15:01 PM5/2/92
to
Re: Dima Volodin <d...@HQ.DEMOS.SU>, who replies:

> Here we go again. Is it a coincidence, that "Antisemitism in MGU" is
> a hot topic in spring? It looks like it's due to incurable avitaminosis,

> which was [acquired] by majority of active readers of this newsgroup back
> in the SU.
>
> Dima

It seems that avitaminosis may have caused me to forget to initial with A.S.
that post about the Soviet discrimination legacy in academia. In the unlikely
case, Dima, you have also acquired the dreaded avitaminosis "back" in the SU,
I post the properly attributed article to refresh your memory on the subject.

Alex
- - - - - - - - - -

From: Andrei Sakharov, Memoirs, A. Knopf, NY, 1990.

[Being discriminated against] is not unusual at the most prestigious Soviet


institutions of higher learning. Much has been written about discrimination
against Jews, a terrible injustice that each year cripples the lives of
thousands of gifted young men and women. Religious believers, the children of
dissidents, and, in Moscow's institutions of higher education, nonresidents of
the city are also subject to discriminatory treatment. Nonresidents actually
are required to obtain higher scores on entrance exams. Other forms of unequal
treatment are more subtle. Exam papers are identified by number, not by name,
but at Moscow University, at least, on key digit of the number alerts the
admissions board to "undesirable" candidates, who are routinely assigned to a
special group for oral exams (where arbitrary grading is easier to justify).
Members of these special groups are supposed to be flunked -- as the story
goes, one young Jew, seeing the other members of the group gathering with him
for the oral exam, exclaimed: "You mean they're taking us straight from here
to Auschwitz?"

Anti-semitic discrimination in university admissions is not just the work of
individual bigots, but part of a deliberate policy of squeezing Jews out of the

country's intellectual establishment. [...]

Anti-semitism is just one aspect of the widespread anti-intellectual and the
caste prejudices of Soviet society. Decent people must be concerned about this
cruelly unjust policy.

A.S.

Dima Volodin

unread,
May 4, 1992, 1:27:29 PM5/4/92
to
On and on and on. Why the hell I should be more concerned with antisemitism,
than "antifeminism" (wrong word, hope you understand what I mean) and other
forms of discrimination? Still this gro...

Oh fuck, no more. The case is hopeless.
--
Dima

Ivan Covdy

unread,
May 4, 1992, 1:35:36 PM5/4/92
to
It would be interesting to publish a book with the problems which
some Jewish guys were given during their entrance exams.

Here is one a friend of mine was asked to solve in MSU, mekh-mat:

Problem: Two guys, A and B, have their houses separated by the river
(see the drawing). Distance from A to the river is a,
distance from B to the river is b, the river's broad is r,
distance AB is l.
The guys want to build a bridge MM across the river in such
a place, that the path between A and B is minimal.
Question: If they build such a bridge, what will be the length of that path?

Have fun!

A
*

_________________________________________
M
M
M
_________________________________________

*
B

Interesting, that it took me at least couple of hours to solve it...
after I graduated from MSU :-)

--
Ivan Covdy, co...@shemesh.metaphor.com, METAPHOR Inc., CA, USA
*** People are looking for a Mind somewhere in the Universe, ***
*** because on the Earth they've been finding only the Stupidity. ***

Vadim Antonov

unread,
May 4, 1992, 5:26:14 PM5/4/92
to
In article <21...@cronos.metaphor.com> co...@cronos.metaphor.com (Ivan Covdy) writes:

>It would be interesting to publish a book with the problems which
>some Jewish guys were given during their entrance exams.

>Here is one a friend of mine was asked to solve in MSU, mekh-mat:
>
>Problem: Two guys, A and B, have their houses separated by the river
> (see the drawing). Distance from A to the river is a,
> distance from B to the river is b, the river's broad is r,
> distance AB is l.
> The guys want to build a bridge MM across the river in such
> a place, that the path between A and B is minimal.
>Question: If they build such a bridge, what will be the length of that path?
>
>Have fun!
>
> A
> *
>
>
>
>_________________________________________
> M
> M
> M
>_________________________________________
>
> *
> B
>
>Interesting, that it took me at least couple of hours to solve it...
>after I graduated from MSU :-)
>

> Ivan Covdy, co...@shemesh.metaphor.com, METAPHOR Inc., CA, USA

Sorry, it took less than 30 seconds to me to solve this "problem"

I think anybody incapable of solving it in 5 minutes shoudn't be
admitted to Moscow U.

The solution is: Look at the picture -- then move the B's side of the
river close to A's bank:


* A

-------M--------

* B

(r=0). Ok, the solution is the straight line, right? Now, what's
changing if we're making river wider and wider? Nothing! The shortest
paths A-M1 M2-B remains the same (M1 and M2 are the ends of the bridge).
(agh, length = l + r).

/And I got "marginal" sum for Moscow U when i was younger and haven't got
my head garbadged with all that programming stuff/.

C'mon, paranoics, tell me something more believable.

Vadim Antonov

Gregory Landsberg

unread,
May 4, 1992, 9:18:21 PM5/4/92
to
In <21...@cronos.metaphor.com> co...@cronos.metaphor.com writes:

> It would be interesting to publish a book with the problems which
> some Jewish guys were given during their entrance exams.
>
> Here is one a friend of mine was asked to solve in MSU, mekh-mat:
>
> Problem: Two guys, A and B, have their houses separated by the river
> (see the drawing). Distance from A to the river is a,
> distance from B to the river is b, the river's broad is r,
> distance AB is l.
> The guys want to build a bridge MM across the river in such
> a place, that the path between A and B is minimal.
> Question: If they build such a bridge, what will be the length of that path?
>
> Have fun!
>
> A
> *
>
>
>
> _________________________________________
> M
> M
> M
> _________________________________________
>
> *
> B
>
> Interesting, that it took me at least couple of hours to solve it...
> after I graduated from MSU :-)

Come on, this is one of the standard trivial problems. It could be
solved in 5-10 minutes using ``reflection method.'' The main idea is that
the light is always propagates in the shortest way (Ferma principle). Thus
if one consider the river as a media with some refraction factor, the
condition of the minimum is that the angles of the pathes from A to the
bridge and from B to the bridge relative to the normal to the river are
equal (see fig.).

A
*
| /
|^/phi
|/


_________________________________________
M
M
M
_________________________________________

/|
/~|phi
* |
B

After that the problem is pure trigonometric and could be solved in 5-10
minutes as I said. I got a very similar problem on the oral mathematics in
MFTI but never considered it like a ``grob'' - it is really very simple.
The ``grobs'' are much more complicated. An example of the ``grob'' is:
Build the square using 4 points on its different sides (with ruler and
``tsirkul''').

Have fun!

Greg

Gregory Landsberg

unread,
May 4, 1992, 9:41:14 PM5/4/92
to
In <1992May4.2...@uunet.uu.net> a...@uunet.uu.net writes:

> Sorry, it took less than 30 seconds to me to solve this "problem"
>
> I think anybody incapable of solving it in 5 minutes shoudn't be
> admitted to Moscow U.
>
> The solution is: Look at the picture -- then move the B's side of the
> river close to A's bank:
>
>
> * A
>
> -------M--------
>
> * B
>
> (r=0). Ok, the solution is the straight line, right? Now, what's
> changing if we're making river wider and wider? Nothing! The shortest
> paths A-M1 M2-B remains the same (M1 and M2 are the ends of the bridge).
> (agh, length = l + r).

Your solution is incorrect, sorry. The right answer is

x = r + sqrt(l**2 - r**2 -2*(a+b)*r) (see my previous posting for details).

Mind, that l is |AB| and not X - X (here X axis is parallel to river).
A B
You may easy check that your solution is wrong if imagine that both A and B
are exactly on the banks (i.e., a=b=0). Then it doesn't matter where to
build the bridge (if it is not to the left from B and not to the right from
A). And the length of the path is r + X -X = r + sqrt(l**2-r**2) which
A B
------------A--------
is less than l+r, by the way.


-----B--------------


> /And I got "marginal" sum for Moscow U when i was younger and haven't got
> my head garbadged with all that programming stuff/.

Yep, it seems to garbaged now :-) (just kidding).

> Vadim Antonov

Greg Landsberg

Alex Goykhman

unread,
May 5, 1992, 11:02:05 AM5/5/92
to


Nah... It's simpler than that. Here's the function we need to optimize:

distance(x) = r + sqrt(a**2 + x**2) + sqrt((b**2 + (l-x)**2))

Let us assume that x>=0 and (l-x)>=0 (meaning that the bridge is built
"between" A and B, which sounds reasonable).

Now, the constants have no bearing on where the function reaches its
minimum, so let's eliminate them.

f(x) = sqrt(x**2) + sqrt(((l-x)**2)) = x + l - x = l

Another word, it is NOT important where the bridge gets built,
as long as it is" between" A and B.

Of course, had it been a real MSU exam, Vadim would have passed,
and I would have flunked :)



--
-----------------------------------------------------------------------------------
Disclaimer: all opinions are mine.
-----------------------------------------------------------------------------------

Igor Belchinskiy

unread,
May 5, 1992, 12:54:26 PM5/5/92
to
In article <1992May5.0...@dxcern.cern.ch> G...@fnal.fnal.gov (Gregory Landsberg) writes:

>The ``grobs'' are much more complicated. An example of the ``grob'' is:
>Build the square using 4 points on its different sides (with ruler and
>``tsirkul''').
>
>Have fun!

Sorry, but this seems also rather simple. :-)
Let's see:

* A

* B
* C
* D

Let's turn picture to PI/2 with center of turn in any letter (let's in A).
X' is an image of X :-)


* C'


* D'
* A = A'

* B
* C
* D
* B'

Now let's move flat' parallel so that A' -> B

* C'
* C ''

* D'
* A = A'
* D''
* B = A''
* C
* D
* B'

* B''

Now D'' and C belongs to the same line as the side of original square
Two cases possible:
1. D'' not equal to C -
than [CD''] gives direction of side of the original square
we make parallel to [CD''] through B, and 2 perpendiculars through A and D.

2. D'' is the same as C
Exercise for the reader. (what's evident without thinking
- we may do the same 2 step transformations with center in B or C or D.
If in all cases we would have appropriate second image the same as
necessay original point, it's indeed special case - intersection
of AD and BC would be a center of square. So it's is trivial)

>
>Greg


Yes, "groby"s were much more complicated :-)
--
- Igor Belchinskiy b...@bsdi.com Here and now.
- Berkeley Software Design, Inc +1(703)204-8089

Igor Belchinskiy

unread,
May 5, 1992, 1:40:36 PM5/5/92
to
In article <1992May5.1...@apollo.hp.com> goykh...@apollo.hp.com (Alex Goykhman) writes:
>In article <1992May5.0...@dxcern.cern.ch> G...@fnal.fnal.gov (Gregory Landsberg) writes:
>
>
>Nah... It's simpler than that. Here's the function we need to optimize:
>
> distance(x) = r + sqrt(a**2 + x**2) + sqrt((b**2 + (l-x)**2))
>
>Let us assume that x>=0 and (l-x)>=0 (meaning that the bridge is built
>"between" A and B, which sounds reasonable).
>
>Now, the constants have no bearing on where the function reaches its
>minimum, so let's eliminate them.

Are you serious?

> f(x) = sqrt(x**2) + sqrt(((l-x)**2)) = x + l - x = l
>
>Another word, it is NOT important where the bridge gets built,
>as long as it is" between" A and B.

It's simply wrong.

>Of course, had it been a real MSU exam, Vadim would have passed,
>and I would have flunked :)

Yes - because your answer is wrong. Try to use common sense
if you forget math. Are this pictures the same?

* A * A
| /
| /
- /
- /
* B * B

> Alex Goykhman

Vadim Antonov

unread,
May 5, 1992, 2:15:45 PM5/5/92
to
In article <1992May5.0...@dxcern.cern.ch> G...@fnal.fnal.gov
(Gregory Landsberg) writes:

>The ``grobs'' are much more complicated. An example of the ``grob'' is:
>Build the square using 4 points on its different sides (with ruler and
>``tsirkul''').

You call IT a grob? Well, it's a task for 7th year schoolars (Igor
instists it's for 8th year :)

My solution is:

Let's call the given points A,B,C and D.
Draw lines AB, BC, CD, and DA. Build four circles using AB, BC, CD and DA
as diameters (trivial). Obviously, the square's vertexes lie on external
half-circles. Then, square's diagonals dissects internal half-circles
to two equal arcs; so it's trivial to find points of intersection of
diagonals and inner half-circles (simply raise perpendiculars to the
circles' diameters from centers). Lets' call these dots E,F,G,H.
Then, choose a pair of dots EG of FH (better to use one with longer
distance to improve the quality of the picture :) and draw a line.
This line will intersect outer half-circles in the points which are
the square vertexes.

In case if all four dots EFGH are equivalent, the problem has an infinite
number of solutions.

Igor found another solution :)

I wish the problems i got on my exams were the same :)

Cheers!

Vadim Antonov

PS. I agree that my solution for the first puzzle was wrong -- i simply
wrote l instead of l' -- the distance AB on the first picture. Getting l'
from a,b,l,r is trivial. Damned habit of leaving triviaities to debugging :)

Viznyuk

unread,
May 5, 1992, 3:44:15 PM5/5/92
to

This posting of prominent russophobe Alex Allister
is a direct fodgery (no surprise - it's usual practice of
such "scientists"). I read Sakharov's letter and see
no resemblance of what Alex Allister writes to the original
text. Well, it could be the result either of Alex's ill
imagination or that of the author of the book he refers to.
Actually Sakharov spoke not so much about antisemitism
as about general unjustice in Sovok's schools:
discrimination of non-residents of Moscow,
discrimination of relatives/children of those who emigrated, brainwashing, etc.
S.V.

Ivan Covdy

unread,
May 6, 1992, 3:58:27 AM5/6/92
to
in article <1992May4.2...@uunet.uu.net>, a...@uunet.uu.net (Vadim Antonov) says:
# Nntp-Posting-Host: rodan.uu.net
#
# In article <21...@cronos.metaphor.com> co...@cronos.metaphor.com (Ivan Covdy) writes:
#
#>It would be interesting to publish a book with the problems which
#>some Jewish guys were given during their entrance exams.
#
#>Here is one a friend of mine was asked to solve in MSU, mekh-mat:
#>
#>Problem: Two guys, A and B, have their houses separated by the river
#> (see the drawing). Distance from A to the river is a,
#> distance from B to the river is b, the river's broad is r,
#> distance AB is l.
#> The guys want to build a bridge MM across the river in such
#> a place, that the path between A and B is minimal.
#>Question: If they build such a bridge, what will be the length of that path?
#>
#>Have fun!
#>
#> A
#> *
#>
#>
#>
#>_________________________________________
#> M
#> M
#> M
#>_________________________________________
#>
#> *
#> B
#>
#>Interesting, that it took me at least couple of hours to solve it...
#>after I graduated from MSU :-)
#>
#> Ivan Covdy, co...@shemesh.metaphor.com, METAPHOR Inc., CA, USA
#
# Sorry, it took less than 30 seconds to me to solve this "problem"
#
# I think anybody incapable of solving it in 5 minutes shoudn't be
# admitted to Moscow U.
#
# The solution is: Look at the picture -- then move the B's side of the
# river close to A's bank:
#
#
# * A
#
# -------M--------
#
# * B
#
# (r=0). Ok, the solution is the straight line, right? Now, what's
# changing if we're making river wider and wider? Nothing! The shortest
# paths A-M1 M2-B remains the same (M1 and M2 are the ends of the bridge).
# (agh, length = l + r).
#
# /And I got "marginal" sum for Moscow U when i was younger and haven't got
# my head garbadged with all that programming stuff/.
#
# C'mon, paranoics, tell me something more believable.
#
# Vadim Antonov

C'mon, Vadim, don't gimme a shit!

Just consider 2 simple cases:
1) l=r
A
---------*---------
Obviously, that path p=l in this case
(your formula gives p=l+r = 2*l)
---------*---------
B

2) a=b=0 and l>r
A
-------------*-----
Obviously, that you can put the bridge in any
place between C and B (say, in C), the path
----*--------.----- remains AC+BC (sum of the two catets), which is
B C AC+sqr(AB**2 - AC**2) = r + sqr(l**2 - r**2).
Has it anything common with your "formula"?

Okay, you failed and received "2".
So, you'd better try your alma mater VMK instead of mekh-mat :-)


--

Ivan Covdy, co...@shemesh.metaphor.com, METAPHOR Inc., CA, USA

Gregory Landsberg

unread,
May 6, 1992, 8:56:41 AM5/6/92
to
In <1992May5.1...@uunet.uu.net> a...@uunet.uu.net writes:

> In article <1992May5.0...@dxcern.cern.ch> G...@fnal.fnal.gov
> (Gregory Landsberg) writes:
>
> >The ``grobs'' are much more complicated. An example of the ``grob'' is:
> >Build the square using 4 points on its different sides (with ruler and
> >``tsirkul''').
>
> You call IT a grob? Well, it's a task for 7th year schoolars (Igor
> instists it's for 8th year :)
>
> My solution is:

<... solution omitted..>

Yep, you are right - this is a simple problem. But there was a ``grob''
with rather similar conditions. Unfortunately I don't remember the exact
conditions now and obviously I mixed this ``grob'' with this simple
problem. I'm terribly sorry. However, somebody on this Net should remember
this problem, because it was explained as an example of ``grob'' in
different schools and places. May be in the ``grob'' it was some kind of
trapezoid and not a square. I really forgot it. It's pity, because the
solution was very nice - as in the most of ``grobs.'' But finding of this
nice and ``simple'' solution is very complicated task if you don't know
this problem - that's the main principle of ``grobs.''

> Cheers!
>
> Vadim Antonov
>

Cheers,

Greg

Alexander Burshteyn

unread,
May 6, 1992, 4:20:48 PM5/6/92
to
In article <1992May6.1...@dxcern.cern.ch> G...@fnal.fnal.gov (Gregory Landsberg) writes:

>Yep, you are right - this is a simple problem. But there was a ``grob''
>with rather similar conditions. Unfortunately I don't remember the exact
>conditions now and obviously I mixed this ``grob'' with this simple
>problem. I'm terribly sorry. However, somebody on this Net should remember
>this problem, because it was explained as an example of ``grob'' in
>different schools and places. May be in the ``grob'' it was some kind of
>trapezoid and not a square. I really forgot it. It's pity, because the
>solution was very nice - as in the most of ``grobs.'' But finding of this

^^^^^^^^^^^^^^^


>nice and ``simple'' solution is very complicated task if you don't know

^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^


>this problem - that's the main principle of ``grobs.''

^^^^^^^^^^^^ ***************************************

Yep, that's the thing. Now here is a genuine ''grob'': you gotta be a genius
to solve it during the exam. I've never seen its solution, though.

OK, here is the problem.

You are given three points associated with an inscribed quadrilateral no sides
of which are parallel: the two points of intersection of the opposite sides
of the quadrilateral and the point of intersection of its diagonals. Find,
only straightedge and ruler, the center of the circle in which the
quadrilateral is inscribed.
(Warning: the quadrilateral itself is NOT given, only those three points.)

>Cheers,
>
>Greg


Good luck.

Alex Burshteyn

***************************
Ne mir tesen, a sloy tonok.
***************************

Ivan Covdy

unread,
May 6, 1992, 9:18:23 PM5/6/92
to

Damn, it's a typical way physics solve the problems :-)
(On mekh-mat there used to be such an expression: to prove a theorem
by a way a physics would do it).

But, as I said, it took place on a math, not physics exam,
and it's a bad taste to use physical laws to solve math problems,
at least for mehk-mat.
What an examinator would do in such a situation, is, probably, asking
you to give a mathematical definition of the light, reflection, refraction,
propagation, etc., and then proving mathematicaly that for the shortest path
the angles are equal...

As for me, I cannot see another way but to calculate the path:
f(x)= sqrt(x^2 + a^2) + sqrt(y^2 + b^2) + r,
where y = sqrt(l^2 - (a+b+r)^2) - x,
and then find the minimum of the function, that is, f'(x)=0.
It'll take awhile to make all the calculations correctly.

> The ``grobs'' are much more complicated. An example of the ``grob'' is:
> Build the square using 4 points on its different sides (with ruler and
> ``tsirkul''').
>
> Have fun!
>
> Greg

Yes, this problem is not so easy, but I wouldn't say that it's much
more difficulut than with the bridge, if you forget about the
physical methods, of course :-)
A
*
* *D
B
*C

Idea:
1) Connect the 4 points in a way to get a 4-angler;
2) For each side of the 4-angler, find its middle, and make a
semi-circle outside of the 4-angler, with diameter=its side.
You got a "flower". It's obvious, that the square's tops should
be on those semi-circles.
3) Take the smallest semi-circle (say, AD), continue BA and CD till
their second intersections with the semi-circle if they exist,
let those be S1 and S2 (in our case BA does have second intersection,
and CD doesn't, so, we take S2=D). Now, pick any point between
on the semi-cicrle between S1 and S2 (let it be S) and
draw the lines SA and SB till their intersections with )AB and )CD
semi-circles. Then take one of received intersections, say, on )AB,
connect it with B and continue till intersection with )BC.
Okay, you received all 4 dots on the 4 semicolons, so, now
connect them in order to get a rectangle.
4) Since you got a rectangle, find its center; obviously, it's going
to be the center for the square, too.
5) Finaly, go to the center and then draw one line in the way to
get somewhere between S1 and S2, and the other one, again
from the center, should be ortogonal to the first one.
Of course, the points of intersection of those lines and
semi-circles are going to be our square's tops.

Note: The square is not unique, that is, you can build more than one
squares which have the given points on their different sides.

Allon Percus

unread,
May 7, 1992, 4:00:55 PM5/7/92
to
In article <1992May5.1...@physics.ucla.edu> viz...@physics.ucla.edu (Viznyuk) writes:
>>[Quote of Sakharov]

>
> This posting of prominent russophobe Alex Allister
>is a direct fodgery (no surprise - it's usual practice of
>such "scientists"). I read Sakharov's letter and see
>no resemblance of what Alex Allister writes to the original
>text.

I just came into this argument right now so forgive me for a bit of
ignorance.

What makes you so sure it's a forgery? This is a very serious accusation
and so I imagine you must have some evidence on which you base it.
Could you, for example, quote what you claim the original text is?

And I would also like to know precisely what you mean when you say that
forgery is a usual practice of scientists.

Allon Percus

Ivan Covdy

unread,
May 8, 1992, 8:49:04 PM5/8/92
to
in article <1992May5.1...@uunet.uu.net>, b...@uunet.uu.net (Igor Belchinskiy) says:
# Nntp-Posting-Host: rodan.uu.net
#
# In article <1992May5.0...@dxcern.cern.ch> G...@fnal.fnal.gov (Gregory Landsberg) writes:
#
#>The ``grobs'' are much more complicated. An example of the ``grob'' is:
#>Build the square using 4 points on its different sides (with ruler and
#>``tsirkul''').

#>
#>Have fun!
#
# Sorry, but this seems also rather simple. :-)
# Let's see:

#
# * A
#
# * B
# * C
# * D
#
# Let's turn picture to PI/2 with center of turn in any letter (let's in A).
# X' is an image of X :-)
#
#
# * C'
#
#
# * D'
# * A = A'
#
# * B
# * C
# * D
# * B'
#
# Now let's move flat' parallel so that A' -> B
#
# * C'
# * C ''
#
# * D'
# * A = A'
# * D''
# * B = A''
# * C
# * D
# * B'
#
# * B''
#
# Now D'' and C belongs to the same line as the side of original square

I didn't get this point. Maybe, it's obvious, but I still cannot see
why D'' and C will be on the same side of the square.
The problem sounds like the following:

A *D' There is a square PQRS. Point A is on PQ, B on PS,
P *..*......* Q and D on SR. There are D' and D'': )DAD' = pi/2,
. . and AD=AD'. AB and D'D'' are parallel and equal.
. . *D'' Prove, that D'' belongs to the side QR.
B * .
. . Is it really so obvious?
S *....*....* R Ne vypendrivajsja, Igorek!
D

Igor Belchinskiy

unread,
May 11, 1992, 1:27:09 PM5/11/92
to
Sorry for long citations.

^^^^^^^^^^^^^^^^^
1. I don't like that level of familiarity between unintroduced peoples.
2. I never told that D'' belongs to QR. What I used is
that lines defined by QR and S''R'' are the same
(especially for Ivan Covdy: not QR and S''R'' but lines defined be them).
I think that should evident (hint for Ivan Covdy: look at
lines defined by PS and Q''R'').
3. I think all other question connected with with this simple task
should be discussed in e-mail if at all.

> Ivan Covdy, co...@shemesh.metaphor.com, METAPHOR Inc., CA, USA

--

0 new messages