A problem in matchpoint scoring
mara...@prodigy.net
rounds of 3 boards each. It was a Mitchell movement. My partner and I
were playing North-South. We ended up with about a 45% and just missed
the money in the "C" strat. Oh, well.
The problem was this. The only East-West pair that we didn't get to
play was giving gifts all night and managed to score only 31%. If we
had been able to play against them, we would (one assumes) have been
able to score a bit better. It is also likely (ironically) that these
opponents would have improved their score as well, if they had had the
chance to play against us.
Of course, the best method would been to have played the 9th round and
found out, but, with an 11-table (or, worse yet, a 17-table) game this
would get unrulely. Failing that, it would seem that this problem
could be solved mathematically.
So, my question for the group is this: Is there a reasonable way to
account for large disparities in performance when a movement is
truncated?
This isn't an idle concern or sour grapes. I am certain that I have
benefitted in the past from not playing the bullies just as often as I
have been demeritted by not playing against the Santa Clauses. One
reason behind the question is that the logical extension of solving
this problem is the ability to create a rating system to rate pairs
(or even individual players) that have never played each other (and
actually never even played common opponents).
axman
> So, my question for the group is this: Is there a reasonable way to
> account for large disparities in performance when a movement is
> truncated?
The 'easiest' thing to do is to balance and seed the field. Balancing
is having about the same number of strong, middle, weak pairs sit NS
as EW. THen assign pair numbers to ensure that the strongest meet the
strongest [as much as possible] and the weakest meet the weakest [as
much as possible].
axman
Frances Hinden
> rounds of 3 boards each. It was a Mitchell movement. My partner and I
> were playing North-South. We ended up with about a 45% and just missed
> the money in the "C" strat. Oh, well.
>
> The problem was this. The only East-West pair that we didn't get to
> play was giving gifts all night and managed to score only 31%. If we
> had been able to play against them, we would (one assumes) have been
> able to score a bit better. It is also likely (ironically) that these
> opponents would have improved their score as well, if they had had the
> chance to play against us.
>
> Of course, the best method would been to have played the 9th round and
> found out, but, with an 11-table (or, worse yet, a 17-table) game this
> would get unrulely. Failing that, it would seem that this problem
> could be solved mathematically.
>
movement around...
> So, my question for the group is this: Is there a reasonable way to
> account for large disparities in performance when a movement is
> truncated?
>
Various things could be suggested, but there is no standard. In
'proper' events, truncated movements are avoided if at all possible.
> This isn't an idle concern or sour grapes. I am certain that I have
> benefitted in the past from not playing the bullies just as often as I
> have been demeritted by not playing against the Santa Clauses. One
> reason behind the question is that the logical extension of solving
> this problem is the ability to create a rating system to rate pairs
> (or even individual players) that have never played each other (and
> actually never even played common opponents).
I once spent a long time creating an individual ranking system with
Jonathan Mestel for a club based on a long history of club results.
It involved inverting quite a few matrices. Funnily enough, Jonathan
turned out to be the best player!
Steve Willner
news:d30df517.03072...@posting.google.com...
> I once spent a long time creating an individual ranking system with
> Jonathan Mestel for a club based on a long history of club results.
> It involved inverting quite a few matrices. Funnily enough, Jonathan
> turned out to be the best player!
Nick Straguzzi has also done some work on rating systems. He suggested an
iterative method, but nowadays it's just as easy to invert a matrix. If you
have N players, you will need to invert an NxN matrix, but it is nearly
diagonal, so there are tricks you can play. I expect it would be feasible
for modern computers even for an organization as large as the ACBL, should
there be a desire to create a rating system -- which there is not! I don't
know for sure, but I suspect the OKbridge rating system uses iteration.
It would be easy to use the same approach to score a single game. (No
computing tricks needed; just invert the matrix by brute force.) The method
automatically corrects for relative strength of opponents met, and the
covariance matrix tells how the ranking of each pair depends on all the
others. The method's main disadvantage is that it is completely opaque to
the players.
Jurgen R.
<swil...@cfa.harvard.edu> wrote:
>"Frances Hinden" <fhi...@orange.net> wrote in message
>news:d30df517.03072...@posting.google.com...
>> I once spent a long time creating an individual ranking system with
>> Jonathan Mestel for a club based on a long history of club results.
>> It involved inverting quite a few matrices. Funnily enough, Jonathan
>> turned out to be the best player!
>
>Nick Straguzzi has also done some work on rating systems. He suggested an
>iterative method, but nowadays it's just as easy to invert a matrix.
Rating is not a linear problem - if you are satisfied with a
linearization you can have it very cheaply. - Practical methods for
inverting nearly diagonal matrices are iterative, I think without
exception.
> If you
>have N players, you will need to invert an NxN matrix, but it is nearly
>diagonal, so there are tricks you can play. I expect it would be feasible
>for modern computers even for an organization as large as the ACBL, should
>there be a desire to create a rating system -- which there is not! I don't
>know for sure, but I suspect the OKbridge rating system uses iteration.
No it doesn't; it consists of a simple, logically incorrect,
arithmetic formula.
>
>It would be easy to use the same approach to score a single game. (No
>computing tricks needed; just invert the matrix by brute force.) The method
>automatically corrects for relative strength of opponents met, and the
>covariance matrix tells how the ranking of each pair depends on all the
>others. The method's main disadvantage is that it is completely opaque to
>the players.
The theory behind Elo's system is also opaque to practically all
users. The player uses an approximation formula that's close enough
for the purpose.
>
Frances Hinden
> "Frances Hinden" <fhi...@orange.net> wrote in message
> news:d30df517.03072...@posting.google.com...
> > I once spent a long time creating an individual ranking system with
> > Jonathan Mestel for a club based on a long history of club results.
> > It involved inverting quite a few matrices. Funnily enough, Jonathan
> > turned out to be the best player!
>
> Nick Straguzzi has also done some work on rating systems. He suggested an
> iterative method, but nowadays it's just as easy to invert a matrix. If you
> have N players, you will need to invert an NxN matrix, but it is nearly
> diagonal, so there are tricks you can play. I expect it would be feasible
> for modern computers even for an organization as large as the ACBL, should
> there be a desire to create a rating system -- which there is not! I don't
> know for sure, but I suspect the OKbridge rating system uses iteration.
That's pretty much what we did. In those days we were both in the
same maths dept (Jonathan was a lecturer, I was a research student) so
we used the NAG routines with FORTRAN. Seems a long time ago now!
Reef Fish
> "Steve Willner" <swil...@cfa.harvard.edu> wrote in message news:<bjqs94$ctl$1...@bob.news.rcn.net>...
> > "Frances Hinden" <fhi...@orange.net> wrote in message
> > news:d30df517.03072...@posting.google.com...
> > > I once spent a long time creating an individual ranking system with
> > > Jonathan Mestel for a club based on a long history of club results.
> > > It involved inverting quite a few matrices.
Frances, would you care to elaborate on what these matrices are?
In particular, apart from the fact that one may wish to know whether
inverting such matrices have any relevance to the "correct" or "real
life sensible ranking" of the players, there is the MATHEMATICAL
question of whether these matrices are necessarily non-singular so
as to make them invertible.
> > Nick Straguzzi has also done some work on rating systems. He suggested an
> > iterative method, but nowadays it's just as easy to invert a matrix.
The relevant question about iterative methods is whether it guarantees
convengence to a solution; then whether it's a sensible solution.
The relevant question about inverting a matrix is whether the matrix is
invertible; and also whether the solution is a sensible one.
In either case, while it's just "as easy" to do one as the other, it's
not at all clear that either is "feasible", or makes sense from the
point of view of realistic ranking.
> > If you
> > have N players, you will need to invert an NxN matrix, but it is nearly
> > diagonal, so there are tricks you can play.
There's much known literature about the inversion of "sparse" matrices,
and those that are nearly diagonal or have certain structural simplicities.
On the other hand ...
> > I expect it would be feasible for modern computers
Based you loose use of mathematical ideas and seemingly unconcerned with
problems of invertibility or other mathematical necessities, I rather
suspect that your "expectation" of feasibility may be mere idle
speculation, and the real and/or potential results being discussed
are no more than the well-known GIGO (Garbage In, Garbage Out) principle
so frequently applied by those insufficiently trained in computing OR
mathematics.
> > even for an organization as large as the ACBL, should
> > there be a desire to create a rating system -- which there is not! I don't
> > know for sure, but I suspect the OKbridge rating system uses iteration.
>
> That's pretty much what we did. In those days we were both in the
> same maths dept (Jonathan was a lecturer, I was a research student) so
> we used the NAG routines with FORTRAN. Seems a long time ago now!
If it's that long ago, then in view of the limited precision of arithmetic
in FORTRAN those days, NAG subroutines or other routines for that matter
may be UNABLE to distinguish a singular matrix from a non-singular one,
let alone deciding whether a matrix is positive definite, positive
semi-definite, or having some other MATHEMATICAL properties that are
prerequisites to discussing "invertibility" and "covariance" matrices.
It was not even a decade ago that a Ph.D. student under my direction
discovered that some papers published on the use of Principal Components
for Regression Analysis was based on data of "truncated precision" which
(unbeknownst to them) rendered the covariance matrix an IMPOSSIBLE one,
i.e., non-positive definite or one with negative eigenvalues.
That's an example of GIGO for those who couldn't tell if a matrix had
the prerequisite properties for it to be considered a covariance matrix!
>
> >
> > It would be easy to use the same approach to score a single game. (No
> > computing tricks needed; just invert the matrix by brute force.)
Just exactly what is meant by "brute force" if a matrix is singular or
non-invertible?
> > The method
> > automatically corrects for relative strength of opponents met, and the
> > covariance matrix tells how the ranking of each pair depends on all the
> > others. The method's main disadvantage is that it is completely opaque to
> > the players.
Often it is complete opague to those who TALK about such methods too,
without having understood what's at stake mathematically, let alone the
need to validate the efficacy of the feasible methods.
-- Bob.
John Crinnion
> fhi...@orange.net (Frances Hinden) wrote in message news:<d30df517.03091...@posting.google.com>...
> > "Steve Willner" <swil...@cfa.harvard.edu> wrote in message news:<bjqs94$ctl$1...@bob.news.rcn.net>...
> > > "Frances Hinden" <fhi...@orange.net> wrote in message
> > > news:d30df517.03072...@posting.google.com...
<massive SNIP>
Why don't you stick to basics, such as learning the definition of average?
Reef Fish
John,
Why don't you look at the SUBJECT of the thread or the content of the
subject being discussed?
I was commenting on, and asking questions about, Frances Hinden's post
about ranking of players by the inversion of matrices.
I was asking some VERY BASIC questions about those matrices, which I
am sure make no more sense to YOU than discussing the basics of
statistical terms, since you apparently know nothing about either.
Save the bandwidth to post something of SUBSTANCE in which you may have
some understanding. You have already adequately displayed your lack of
knowledge in the OTHER thread.
-- Bob.
John Crinnion
OK, then, learn the meaning of 'matrix' - HINT: it is a technical term
in a fairly advanced branch of algebra, and as such has nothing to do
with the movie of that name, or a type of printer.
Reef Fish
> Large_Nass...@Yahoo.com (Reef Fish) wrote in message news:<8fb7380b.03091...@posting.google.com>...
JC >
JC>> Why don't you stick to basics, such as learning the definition of average?
RF > John,
RF >
RF > Why don't you look at the SUBJECT of the thread or the content of the
RF > subject being discussed?
RF >
RF > I was commenting on, and asking questions about, Frances Hinden's post
RF > about ranking of players by the inversion of matrices.
and note particularly this:
RF> I was asking some VERY BASIC questions about those matrices, which I
RF> am sure make no more sense to YOU than discussing the basics of
RF> statistical terms, since you apparently know nothing about either.
RF>
RF> Save the bandwidth to post something of SUBSTANCE in which you may have
RF> some understanding. You have already adequately displayed your lack of
RF> knowledge in the OTHER thread.
John Crinnion,
You have just given prove to the correctly of my assertions above, about
your IGNORANCE in the subjects of math and statistics in question:
JC> OK, then, learn the meaning of 'matrix' - HINT: it is a technical term
JC> in a fairly advanced branch of algebra, and as such has nothing to do
JC> with the movie of that name, or a type of printer.
A 'matrix' is one of the most ELEMENTARY terms in algebra that even
high school and college freshmen students in math subjects are expected
to know these days.
The paragraph you wrote above couldn't have been a better self-portrait
and self-indictment by you. As the saying go, before you opened your
mouth, there is some doubt as to whether you were COMPLETELY ignorant
about the subject. As soon as you opened your mouth, you removed ALL
such DOUBTS!
-- Bob.
Jurgen R.
am curious to know what your rating procedure was. What were these
matrices that you inverted?
Regards, Jürgen
On 12 Sep 2003 07:17:37 -0700, fhi...@orange.net (Frances Hinden)
wrote:
John Crinnion
Of course, but I put your level at grade school.
Reef Fish
> > RF > John,
> > RF >
> > RF > Why don't you look at the SUBJECT of the thread or the content of the
> > RF > subject being discussed?
> > RF >
> > RF > I was commenting on, and asking questions about, Frances Hinden's post
> > RF > about ranking of players by the inversion of matrices.
JC> OK, then, learn the meaning of 'matrix' - HINT: it is a technical term
JC> in a fairly advanced branch of algebra, and as such has nothing to do
JC> with the movie of that name, or a type of printer.
I do enjoy occasionally watching a Village IDIOT with foot in mouth
while shooting his other foot. So I said,
> > A 'matrix' is one of the most ELEMENTARY terms in algebra that even
> > high school and college freshmen students in math subjects are expected
> > to know these days.
The sign of a true IDIOT is that he keeps returning with foot in mouth,
with this lame retort, after having called a 'matrix' "a term in a
fairly advanced branch of algebra" and was told that even a high school
math student should have known what it is:
Crinnon the IDIOT> Of course, but I put your level at grade school.
Well, I actually knew it when I was in grade school too, and never
thought it a term in "advanced branch of algebra", as you do NOW.
ROTFLMAO.
Furthermore, the first school I attanded was when I passed an entrance
written exam to enter the THIRD grade at the age of six, when John
Crinnon probably hadn't even been weaned. :-))))
Get lost, John Crinnon.
I can see what comes NEXT. Some of JOhn's Village Idiot pals are
going to stick their vacuous heads out and whined that I had BRAGGED
about being in the 3rd grade at the age of 6 -- even though it was
just a statement ofact to stuff down John Crinnon's vacuous cranium,
after his dribbling drivil::
> Of course, but I put your level at grade school.
Remember how the morons came out of the Village in droves when I
responded to Ed Reppert's (analogous lame insult) about what
I posted about a statistical "mode",
Mr. Ed> You're the one who doesn't get it. And apparently never will.
and I told him I had a Ph.D. degree in the subject, etc. and asked
RF> But what have you got to show for your self-proclaimed expertise
RF> about statistics or the usage of statistical terms, Mr. Ed?
The rest was history. Morons then start complaining that I showed them
my credentials, in the above context which they never knew or had
forgotten!
That's the way it is, in SOME internet ngs. Morons and Idjits abound.
Some wise up and kept quiet; Others such as John Crinnon, never
learned anything and never will.
John, you should have listened to your Mama when she told you it was
a bad idea for you to quit school after the 4th grade, even though
you were already 24. <BWG>
What I had said about John Crinnon the Village Idiot remains true:
Reef Fish
news:<8fb7380b.03091...@posting.google.com>...
I referred to John Crinnion several times as John Crinnon.
I checked the archives and found that there was not a "John Crinnon"
in the entire history of newsgroup postings, since 1981, so the
misspelling should have been clear that I meant "John Crinnion" --
but I do want to make this correction, and apologize to Mr. Crinnion
for having misspelled his name those several times in the post:
-- Bob.
John Crinnion
You just don't get it, do you? I was *patronising* you.
Martin Ambuhl
[An 89-line posting, all quotation except:]
> You just don't get it, do you? I was *patronising* you.
Please try to trim back the quoted material. There is no more excuse for
this huge quote just to append one line when you are twitting the other
person than there is when you are adding "me, too!".
--
Martin Ambuhl
Reef Fish
> > JC> OK, then, learn the meaning of 'matrix' - HINT: it is a technical term
> > JC> in a fairly advanced branch of algebra, and as such has nothing to do
> > JC> with the movie of that name, or a type of printer.
After having been told that a 'matrix' is an elementary term, even to
a high school student in math,
Crinnion the IDIOT> Of course, but I put your level at grade school.
> You just don't get it, do you? I was *patronising* you.
Of course I got it. It was your act of condescension, when you should
have stick to your humanities and classics area and not show your
complete ignorance in terms of math that even high school kids know.
That's why I "patroized" you back, quid pro quo:
> > Well, I actually knew it when I was in grade school too, and never
> > thought it a term in "advanced branch of algebra", as you do NOW.
> > ROTFLMAO.
> >
> > Furthermore, the first school I attanded was when I passed an entrance
> > written exam to enter the THIRD grade at the age of six, when John
> > Crinnon probably hadn't even been weaned. :-))))
> >
> > Get lost, John Crinnion.
You just don't get it, do you? I was *patronising* you BACK.
The only difference between our condescension is that yours was entirely
gratuitous while mine was completely justified by your action AND your
absence of knowledge.
Go whine to your Mama. Not here.
-- Bob.
Henry Raymond
"Reef Fish" <Large_Nass...@Yahoo.com> wrote in message
news:8fb7380b.03091...@posting.google.com...
nothing worth reading.
Yawn.
Reef Fish
A Clueless Newbie <tm> of vintage April 2003 with a lifetime
history of 10 posts in ngs.
For a brief moment, I thought I get to read something of substance
about BRIDGE in this ng.
:-( ) <--- That's a BIG YAWN for your type of morons.
-- Bob.
P.S. Whatever happened to the pseudo-mathematicians who started
the "matrix" inversion and slinging to impress their moronic friends
in the ng who know absolutely NOTHING about mathematics or statistics?
What are those matrices that are inverted to rank bridge players??
Enquiring minds want to know.
John Crinnion
>
> > > JC> OK, then, learn the meaning of 'matrix' - HINT: it is a technical term
> > > JC> in a fairly advanced branch of algebra, and as such has nothing to do
> > > JC> with the movie of that name, or a type of printer.
>
> After having been told that a 'matrix' is an elementary term, even to
> a high school student in math,
>
> Crinnion the IDIOT> Of course, but I put your level at grade school.
>
>
> > You just don't get it, do you? I was *patronising* you.
>
> Of course I got it. It was your act of condescension, when you should
> have stick to your humanities and classics area and not show your
> complete ignorance in terms of math that even high school kids know.
>
>
> That's why I "patroized" you back, quid pro quo:
>
> > > Well, I actually knew it when I was in grade school too, and never
> > > thought it a term in "advanced branch of algebra", as you do NOW.
> > > ROTFLMAO.
> > >
> > > Furthermore, the first school I attanded was when I passed an entrance
> > > written exam to enter the THIRD grade at the age of six, when John
> > > Crinnon probably hadn't even been weaned. :-))))
> > >
> > > Get lost, John Crinnion.
>
>
> You just don't get it, do you? I was *patronising* you BACK.
A redneck such as you (and your momma and pappa) cannot patronise
anybody - hence your strange whining note of grievance when in the
midst of your intellectual superiors (ie, everyone else on this
board).
John Crinnion
Sorry, I allowed myself to be distracted by the insolence of a flea.
Frances Hinden
> > "Steve Willner" <swil...@cfa.harvard.edu> wrote in message news:<bjqs94$ctl$1...@bob.news.rcn.net>...
> > > "Frances Hinden" <fhi...@orange.net> wrote in message
> > > news:d30df517.03072...@posting.google.com...
> > > > I once spent a long time creating an individual ranking system with
> > > > Jonathan Mestel for a club based on a long history of club results.
> > > > It involved inverting quite a few matrices.
>
> Frances, would you care to elaborate on what these matrices are?
> In particular, apart from the fact that one may wish to know whether
> inverting such matrices have any relevance to the "correct" or "real
> life sensible ranking" of the players, there is the MATHEMATICAL
> question of whether these matrices are necessarily non-singular so
> as to make them invertible.
>
>
the stuff we did (too many computer & house moves).
To answer the singularity question, we were not exactly inverting but
(if I remember correctly) searching for eigenvalues of a symmetric
matrix. We used the practical check of seeing if any were zero;
though I think we also proved that our matrix was positive definite
(either that or we tried to & didn't manage it).
As this topic has more interest than I expected, I may try and
reconstruct/improve on it, but not for a few weeks as life is a bit
frantic at the moment.
Reef Fish
in response to "Steve Willner" <swil...@cfa.harvard.edu>
FH> I once spent a long time creating an individual ranking system with
FH> Jonathan Mestel for a club based on a long history of club results.
FH> It involved inverting quite a few matrices.
RF> Frances, would you care to elaborate on what these matrices are?
RF> In particular, apart from the fact that one may wish to know whether
RF> inverting such matrices have any relevance to the "correct" or "real
RF> life sensible ranking" of the players, there is the MATHEMATICAL
RF> question of whether these matrices are necessarily non-singular so
RF> as to make them invertible.
> er... not quite now. I had a quick look back, and can't find any of
> the stuff we did (too many computer & house moves).
>
> To answer the singularity question, we were not exactly inverting but
> (if I remember correctly) searching for eigenvalues of a symmetric
> matrix. We used the practical check of seeing if any were zero;
> though I think we also proved that our matrix was positive definite
> (either that or we tried to & didn't manage it).
To check for singularity, the use of the determinant would have been
much simpler, wouldn't it? But a positive definite matrix would
answer more than just the question of invertibility.
> As this topic has more interest than I expected, I may try and
> reconstruct/improve on it, but not for a few weeks as life is a bit
> frantic at the moment.
Look forward to seeing what kind of matrix it is and (more importantly
for my own curiosity) why those matrices were useful for the purposes
of ranking players.
-- Bob.
Reef Fish
I am changing the "Subject" to accommodate Mr. Crinnion's incessant
NOISE in the serious thread so that those of us interested in Frances
Hinden's mathematical formulation of the problem can discuss it when
she returns to that subject, under the thread which John Crinnion the
Village Idiot had posted SIX times, with nothing but NOISE!
JC> OK, then, learn the meaning of 'matrix' - HINT: it is a technical term
JC> in a fairly advanced branch of algebra, and as such has nothing to do
JC> with the movie of that name, or a type of printer.
> >
> > After having been told that a 'matrix' is an elementary term, even to
> > a high school student in math,
Crinnion the IDIOT> Of course, but I put your level at grade school.
You'll have a chance to see if you understand the grade school level
math talk then. :O) But you had better get out your English
dictionary and look up the meaning of 'eigenvalues' first, why it
is relevant to the discussion of a 'matrix'. :-)))
Crinnion > You just don't get it, do you? I was *patronising* you.
> >
> > Of course I got it. It was your act of condescension,
> >
> > That's why I "patroized" you back, quid pro quo:
> >
RF> Well, I actually knew it when I was in grade school too, and never
RF> thought it a term in "advanced branch of algebra", as you do NOW.
RF> ROTFLMAO.
RF>
RF> Furthermore, the first school I attanded was when I passed an entrance
RF> written exam to enter the THIRD grade at the age of six, when John
RF> Crinnion probably hadn't even been weaned. :-))))
RF>
> > You just don't get it, do you? I was *patronising* you BACK.
> A redneck such as you (and your momma and pappa) cannot patronise
> anybody - hence your strange whining note of grievance when in the
> midst of your intellectual superiors (ie, everyone else on this
> board).
A BOARD? That explains in part your total cluelessness and ignorance!
This is a USENET Newsgroup! Not a "board", you Block Head! But you
can make boards by chopping your block head though. :-)))))))
-- Bob.
Thomas Dehn
"Reef Fish" <Large_Nass...@Yahoo.com> wrote:
> fhi...@orange.net (Frances Hinden) wrote in message
news:<d30df517.0309...@posting.google.com>...
> in response to "Steve Willner" <swil...@cfa.harvard.edu>
> > the stuff we did (too many computer & house moves).
> >
> > To answer the singularity question, we were not exactly inverting but
> > (if I remember correctly) searching for eigenvalues of a symmetric
> > matrix. We used the practical check of seeing if any were zero;
> > though I think we also proved that our matrix was positive definite
> > (either that or we tried to & didn't manage it).
>
> To check for singularity, the use of the determinant would have been
> much simpler, wouldn't it?
The determinant is, for all practical purposes, completely
useless to check whether a matrix is singular, as its
computation is vulnerable to round-off errors.
A singular value decomposition approach would be much more
reliable than trying to compute the determinant.
You might want to have a look at Golub - van Loan's
"Matrix computations".
Thomas
Reef Fish
> "Reef Fish" <Large_Nass...@Yahoo.com> wrote:
> > fhi...@orange.net (Frances Hinden) wrote in message
> news:<d30df517.0309...@posting.google.com>...
> > in response to "Steve Willner" <swil...@cfa.harvard.edu>
> > > er... not quite now. I had a quick look back, and can't find any of
> > > the stuff we did (too many computer & house moves).
> > >
> > > To answer the singularity question, we were not exactly inverting but
> > > (if I remember correctly) searching for eigenvalues of a symmetric
> > > matrix. We used the practical check of seeing if any were zero;
> > > though I think we also proved that our matrix was positive definite
> > > (either that or we tried to & didn't manage it).
> >
> > To check for singularity, the use of the determinant would have been
> > much simpler, wouldn't it?
>
> The determinant is, for all practical purposes, completely
> useless to check whether a matrix is singular, as its
> computation is vulnerable to round-off errors.
REALLY? Surely you don't mean to say that other numerical
linear algebra computational methods are not vulnerable to
roundoff errors!
The vulnerability to roundoff errors depends PRIMARILY on:
1. The precision of the computational software, and 2. How
ill-conditioned the matrix is.
Your statement about the "completely useless" aspect of the
determinant showed a sign of naivete in theory and practice.
For one thing, the DEFINITION of a singular matrix is one
whose determinant is ZERO.
> A singular value decomposition approach would be much more
> reliable than trying to compute the determinant.
This broad stroke blanket statement is patently false!
> You might want to have a look at Golub - van Loan's
> "Matrix computations".
No need. I have read dozens of books and journal articles on
the subject. I also have published books, and written scientific
papers in mathematical, statistical, and computational journals on
statistical subjects in which "matrix computations" and especially
the NUMERICAL precision of matrix computations are of fundamental
importance. Singular value decomposition and eigenvalue analysis
are basic in these areas.
>
> Thomas
Just a matter of curiosity -- what field (and what areas within
that field) is your specialty that involves NUMERICAL linear
algebra and matrix computations?
-- Bob.
John Crinnion
<SNIP>
> > A redneck such as you (and your momma and pappa) cannot patronise
> > anybody - hence your strange whining note of grievance when in the
> > midst of your intellectual superiors (ie, everyone else on this
> > board).
>
> A BOARD? That explains in part your total cluelessness and ignorance!
> This is a USENET Newsgroup! Not a "board", you Block Head!
I've got news for you - Google bought Usenet up and took it mainstream
into the brave new world of bulletin boards. I would no more buy
special NG software nowadays than I would take up radio hamming as a
hobby by way of substitute for the cell phone.
Gordon Rainsford
> I've got news for you - Google bought Usenet up and took it mainstream
> into the brave new world of bulletin boards.
Err... I think you're digging yourself into a large hole here, John. All
Google did was buy DejaNews' archive. Usenet is not owned by anyone, not
there to be bought or sold. Rec.games.bridge is certainly a usenet
newsgroup - that does not stop being the reality just because you can
access it through a website.
> I would no more buy
> special NG software nowadays than I would take up radio hamming as a
> hobby by way of substitute for the cell phone.
You don't need special software to access newsgroups, though you'd
probably benefit from having it.
--
Gordon Rainsford
London UK
John Crinnion
> John Crinnion <jcri...@yahoo.com> wrote:
>
> > I've got news for you - Google bought Usenet up and took it mainstream
> > into the brave new world of bulletin boards.
>
> Err... I think you're digging yourself into a large hole here, John. All
> Google did was buy DejaNews' archive. Usenet is not owned by anyone, not
> there to be bought or sold. Rec.games.bridge is certainly a usenet
> newsgroup - that does not stop being the reality just because you can
> access it through a website.
If Usenet is not owned by anyone, then it must be owned by *everyone*.
(Like common grazing lands before they were appropriated by the
State.)
Clicking on the 'Groups' link on Google's homepage enables me to
access any Usenet NG I choose. The merest formality of an application
procedure makes me - or anyone who so chooses - a member, with a
user-name and password. So, from where I am standing Usenet, a vast
resource indeed, is a vast gateway to innumerable BULLETIN BOARDS.
Reef Fish
> > John Crinnion <jcri...@yahoo.com> wrote:
> >
> > > I've got news for you - Google bought Usenet up and took it mainstream
> > > into the brave new world of bulletin boards.
> >
> > Err... I think you're digging yourself into a large hole here, John. All
> > Google did was buy DejaNews' archive. Usenet is not owned by anyone, not
> > there to be bought or sold. Rec.games.bridge is certainly a usenet
> > newsgroup - that does not stop being the reality just because you can
> > access it through a website.
Gordon is absolutely correct in his paragraph above, and also
completely correct in the rest of his post you snipped which showed
your ignorance even more.
Alas, Village Idiot Crinnion has been digging holes deeper and deeper
every time he opens his mouth. And he continues ...
>
> If Usenet is not owned by anyone, then it must be owned by *everyone*.
> (Like common grazing lands before they were appropriated by the
> State.)
That's a good enough analogy for the present discussion. That's why
Usenet is a PUBLIC, OPEN forum for discussion that is SELF-REGULATED
-- regulated by its participating members who pay no membership dues,
rather than by a "moderator" or "owner" in some other kind of
discussion forums.
That's why while John Crinnion is an IDIOT making idiotic nuisance
posts, John Crinnion is entitled to make a fool of himself publicly
and no one has the right to expel him, though everyone can tell him
that he is an idiot. Sooner or later, the idiots disappear from
the group, and that's how a USENET ng is self-regulated.
> So, from where I am standing Usenet, a vast
> resource indeed, is a vast gateway to innumerable BULLETIN BOARDS.
No!
Most bulletin boards are PRIVATELY owned, and subject to membership
use only. Many of them are also under the watchful eyes of a
moderator who can censor what you post or expel you! NOBODY does
that in a ng.
That's why IDIOTs like yourself are seen everyday in hundreds of ngs.
You cannot access the ISP Bulletin Boards such as AOL,
MSN, etc. without subscription to the ISPs. What are posted there
are also NOT archived by groups.google.com, which archives ONLY
newsgroups and certain LISTS.
Keep digging, John Crinnion. ;-)
-- Bob.
John (MadDog) Probst
<Large_Nass...@Yahoo.com> writes
>jcri...@yahoo.com (John Crinnion) wrote in message news:<2f389bab.0309191419.4
>ca6...@posting.google.com>...
>> r...@gordonrainsford.co.uk (Gordon Rainsford) wrote in message news:<1g1jolq.1d
>xock8tdjkwkN%r...@gordonrainsford.co.uk>...
>> > John Crinnion <jcri...@yahoo.com> wrote:
>> >
>> If Usenet is not owned by anyone, then it must be owned by *everyone*.
>> (Like common grazing lands before they were appropriated by the
>> State.)
>
>That's a good enough analogy for the present discussion. That's why
>Usenet is a PUBLIC, OPEN forum for discussion that is SELF-REGULATED
>-- regulated by its participating members who pay no membership dues,
>rather than by a "moderator" or "owner" in some other kind of
>discussion forums.
>
>That's why while John Crinnion is an IDIOT making idiotic nuisance
>posts, John Crinnion is entitled to make a fool of himself publicly
>and no one has the right to expel him, though everyone can tell him
>that he is an idiot. Sooner or later, the idiots disappear from
>the group, and that's how a USENET ng is self-regulated.
So all we have to do is wait for you to stop posting. ?
cheers john
>Keep digging, John Crinnion. ;-)
>
>-- Bob.
--
John (MadDog) Probst| . ! -^- |icq 10810798
451 Mile End Road | /|__. \:/ |OKb ChienFou
London E3 4PA | / @ __) -|- |jo...@asimere.com
+44-(0)20 8983 5818 | /\ --^ | |www.asimere.com/~john
Reef Fish
> In article <8fb7380b.0309...@posting.google.com>, Reef Fish
> <Large_Nass...@Yahoo.com> writes
> >jcri...@yahoo.com (John Crinnion) wrote in message news:<2f389bab.0309191419.4
> >ca6...@posting.google.com>...
> >> r...@gordonrainsford.co.uk (Gordon Rainsford) wrote in message news:<1g1jolq.1d
> xock8tdjkwkN%r...@gordonrainsford.co.uk>...
> >> > John Crinnion <jcri...@yahoo.com> wrote:
> >> >
> snip
> >> If Usenet is not owned by anyone, then it must be owned by *everyone*.
> >> (Like common grazing lands before they were appropriated by the
> >> State.)
> >
> >That's a good enough analogy for the present discussion. That's why
> >Usenet is a PUBLIC, OPEN forum for discussion that is SELF-REGULATED
> >-- regulated by its participating members who pay no membership dues,
> >rather than by a "moderator" or "owner" in some other kind of
> >discussion forums.
> >
> >That's why while John Crinnion is an IDIOT making idiotic nuisance
> >posts, John Crinnion is entitled to make a fool of himself publicly
> >and no one has the right to expel him, though everyone can tell him
> >that he is an idiot. Sooner or later, the idiots disappear from
> >the group, and that's how a USENET ng is self-regulated.
>
> So all we have to do is wait for you to stop posting. ?
>
> cheers john
Just don't hold your breath while waiting.
Some ng readers have been waiting for 15 years, without success ...
others died holding their breath while waiting :-) and the majority
of the idiots left for having been exposed to be idiots, or learned
to keep their feet away from their mouths ... in time.
As for THIS newsgroups, you may not have to wait too long. If I
stop reading/posting, it would only mean that I am tired of reading
a ng dominated by MadDogs, morons, and Village idiots and too little
signal (vs noise) in the threads of bridge substance.
On second thought -- MadDog, why don't you hold your breath while
you're waiting. :-))))
-- Bob.
John D'Errico
Large_Nass...@Yahoo.com (Reef Fish) wrote:
> fhi...@orange.net (Frances Hinden) wrote in message
> news:<d30df517.0309...@posting.google.com>...
> in response to "Steve Willner" <swil...@cfa.harvard.edu>
> > er... not quite now. I had a quick look back, and can't find any of
> > the stuff we did (too many computer & house moves).
> >
> > To answer the singularity question, we were not exactly inverting but
> > (if I remember correctly) searching for eigenvalues of a symmetric
> > matrix. We used the practical check of seeing if any were zero;
> > though I think we also proved that our matrix was positive definite
> > (either that or we tried to & didn't manage it).
>
> To check for singularity, the use of the determinant would have been
> much simpler, wouldn't it? But a positive definite matrix would
> answer more than just the question of invertibility.
Huh!? Never try to pass yourself off as knowing something
about mathematics and then suggest that a determinant
is a good numerical check for singularity. It just makes
someone who does know something about mathematics and
numerical analysis laugh.
John D'Errico
(Numerical analyst and applied mathematician)
--
Follow the directions to get my true e-mail address.
I'm sorry for the inconvenience.
John D'Errico
Large_Nass...@Yahoo.com (Reef Fish) wrote:
> "Thomas Dehn" <thomas...@arcor.de> wrote in message
> news:<bkdb3p$hg3a$1...@ID-57266.news.uni-berlin.de>...
> > You might want to have a look at Golub - van Loan's
> > "Matrix computations".
>
> No need. I have read dozens of books and journal articles on
> the subject. I also have published books, and written scientific
> papers in mathematical, statistical, and computational journals on
> statistical subjects in which "matrix computations" and especially
> the NUMERICAL precision of matrix computations are of fundamental
> importance. Singular value decomposition and eigenvalue analysis
> are basic in these areas.
Sorry, you gave yourself away as a mathematical poser
by simply suggesting the determinant.
I'll admit that you may have taken a class as an
undergrad, maybe even used some linear algebra in
the dark past. Its a wee bit funny though when someone
makes it so obvious that they are not as good as they
loudly proclaim. Merely using the words "singular value
decompsition" and "sparse matrix" in moderately coherant
sentences do not a mathematician make.
John
(I make no claims to be a great mathematician. Just
a decent one. I do claim the ability to know one when
I see one though.)
Reef Fish
> In article <8fb7380b.0309...@posting.google.com>,
> Large_Nass...@Yahoo.com (Reef Fish) wrote:
>
> > "Thomas Dehn" <thomas...@arcor.de> wrote in message
> > news:<bkdb3p$hg3a$1...@ID-57266.news.uni-berlin.de>...
>
> > > You might want to have a look at Golub - van Loan's
> > > "Matrix computations".
> >
> > No need. I have read dozens of books and journal articles on
> > the subject. I also have published books, and written scientific
> > papers in mathematical, statistical, and computational journals on
> > statistical subjects in which "matrix computations" and especially
> > the NUMERICAL precision of matrix computations are of fundamental
> > importance. Singular value decomposition and eigenvalue analysis
> > are basic in these areas.
>
> Sorry, you gave yourself away as a mathematical poser
> by simply suggesting the determinant.
Geez, this ng is definitely not in short supply of Village Idiots
in mathematics related topics, isn't it?
> I'll admit that you may have taken a class as an
> undergrad, maybe even used some linear algebra in
> the dark past.
Can't you read the paragraph you cited above? I taught my first
GRADUATE course in Numerical Analysis probably before you were born!
> Its a wee bit funny though when someone
> makes it so obvious that they are not as good as they
> loudly proclaim. Merely using the words "singular value
> decompsition" and "sparse matrix" in moderately coherant
> sentences do not a mathematician make.
That is correct, and certainly applies to YOU. So, let me pose
a trivial little matrix problem for you.
Let M be the 3 x 3 (symmetric) matrix with elements
( 7 3 5 )
M = ( 3 1 2 )
( 5 2 1 )
I can do BOTH the determinant and the inverse of M in my head
(without pencil, paper, or any calculator) -- with no roundoff
error of any kind.
Now it's your turn to see the size of the foot in your mouth --
you can use whatever computer program and whatever numerical
method you choose:
1. What are the eigenvalues (singular values) of M?
2. I assume you know the determinant of M is the product of
these eigenvalues. What value do you get for the determinant
computed this way (using the product of the singular values)?
3. WHY do you NEED the eigenvalues of M to find the INVERSE of
M which I can do in my head?
4. What is the inverse of M?
> John
> (I make no claims to be a great mathematician.
You could have fool me if you did. :-))))))))
Johnny Boy, you wouldn't by chance trying to compete against the
other John (Crinnion) for the title of "Village Idiot of math
in rgb", would you?
> Just a decent one. I do claim the ability to know one when
> I see one though.)
Let see what kind of "mathematician" you are in your answers to
the 4 questions above. They are REALLY trivial questions you
know -- just a clumsy (stupid) way of using (1) to answer the
question of whether M is "singular".
I got the answer in my head by mentally calculating the det(M)
even before you can type the elements of the matrix into your
computer.
-- Bob.
Ian Payn
Reef Fish wrote:
...
<SNIPPED BOASTING, LONG MATHS QUESTION AND SOME MORE BOASTING>
> I can do BOTH the determinant and the inverse of M in my head
> (without pencil, paper, or any calculator) -- with no roundoff
> error of any kind.
++++Bully for you. No, I mean it. I am second to none in my admiration
for those who are able mathematicians, my own abilities in that field
being no better than average (I think I mean "average", but I'm
certainly not going to get into an argument about it) at best. What I
don't understand is why, given your presumed abilities in one sphere
(although just tossing out names of theorems/practices etc almost at
random is an odd way of showing it) you need to go to such lengths as
you do to "prove" it. Hectoring, bullying, denigration and insult are
the tools of the foolish and insecure, not the wise.
Another reason that you aren't accorded the credibility that you think
you deserve (as if your tone wasn't reason enough) is that your English
is bad, or, to give you the benefit of the doubt, careless. There are
those, believe it or not, who attach as much importance to an ability to
channel thought into lucid and grammatic prose as they do to higher
mathematics, if not (dare I say it) more so.
> Johnny Boy, you wouldn't by chance trying to compete against the
> other John (Crinnion) for the title of "Village Idiot of math
> in rgb", would you?
++++I don't think John Crinnion was "Village Idiot of math", I think he
was just "Village Idiot". Whether he is directly answerable to "Chief
Moron" Gordon Rainsford is unclear. I don't think I'll bother with
another "Confusing morons with Mormons" joke. The last one fell
completely flat.
> I got the answer in my head by mentally calculating the det(M)
> even before you can type the elements of the matrix into your
> computer.
++++At the risk of sounding like a teenager, so what? Someone challenged
your credibility as a mathematician and you type out some endless screed
that means nothing to rather a lot of r.g.b. readers, the point of which
is merely to brag about how clever you are. And don't turn round and
tell me that you're not bragging, you're stating facts. That didn't wash
before, and it won't wash now. Some of us can tell the difference.
This is a newsgroup for bridge discussion. This discussion may often
stray off into other topics. Mathematics, philology, literature -
whatever. These digressions are often a cause of interest and delight.
Why not drop this whole ego-trip and look at another thread or two? Give
your opinion on an auction. Deride an opening lead. Agree with David
Stevenson about something (that last one takes a bit of doing, I'll
grant you). You might find, after a while, that you get some people
agreeing with you, or having amicable discussions with you at least.
That would be a much more constructive use of everyone's time, wouldn't it?
IP.
Sid Ismail
wrote:
: I can do BOTH the determinant and the inverse of M in my head
: (without pencil, paper, or any calculator) -- with no roundoff
: error of any kind.
Amazing - I am so impressed.
Sid
Douglas Newlands
:
:Let M be the 3 x 3 (symmetric) matrix with elements
:
: ( 7 3 5 )
: M = ( 3 1 2 )
: ( 5 2 1 )
:
:I can do BOTH the determinant and the inverse of M in my head
:(without pencil, paper, or any calculator) -- with no roundoff
:error of any kind.
Is this skill worth learning?
Does it help one score (bridge or women)?
Dougie
1/2*n*(n+1) < n*n
--
Douglas Newlands Ph.D., School of IT, Tel +61 3 5227 1165, Fax +61 3 5227 2028
Deakin University, Geelong, Victoria 3217, Australia, www.deakin.edu.au
Reef Fish
> In article <8fb7380b.03092...@posting.google.com> Large_Nass...@Yahoo.com (Reef Fish) writes:
> :
> :Let M be the 3 x 3 (symmetric) matrix with elements
> :
> : ( 7 3 5 )
> : M = ( 3 1 2 )
> : ( 5 2 1 )
> :
> :I can do BOTH the determinant and the inverse of M in my head
> :(without pencil, paper, or any calculator) -- with no roundoff
> :error of any kind.
>
> Is this skill worth learning?
Only to those who work with, and use, matrices.
> Does it help one score (bridge or women)?
>
> Dougie
Only in the Aussie bars you frequent, where just being able to
crawl out of the bar would do it.
-- Bob.
Reef Fish
Ian, I've taken your "discussion" out of the serious thread in which
I was waiting for Frances Hinden to return, for the same reason I have
taken John Crinnion's NOISE out of the same thread. You Ian, have
made too much noise in this ng already. But I'll give you a straight
and detailed reply to your gratuitous post, so that I can point to
this post if and when other Village Idiots came late to the scene
(as you did) and felt the urge to put foot in their mouth (as you
did) without understanding what went on before the immediate post
in which I merely gave my quid pro quo reply on the subject matter!
I believe you DO know how to read English -- you're just too obtuse
to follow any thread or anything of mathematical substance, and
you are too officious to think that YOU have the right to tell
others what they should or shouldn't discuss.
So, read my reply CAREFULLY, Ian -- and I hope this is the last
time I see you wagging your tongue in display of your ignorance
and unadulterated NOISE.
> Reef Fish wrote:
> ...
>
> <SNIPPED BOASTING, LONG MATHS QUESTION AND SOME MORE BOASTING>
>
> > I can do BOTH the determinant and the inverse of M in my head
> > (without pencil, paper, or any calculator) -- with no roundoff
> > error of any kind.
>
> ++++Bully for you. No, I mean it. I am second to none in my admiration
> for those who are able mathematicians, my own abilities in that field
> being no better than average (I think I mean "average", but I'm
> certainly not going to get into an argument about it) at best. What I
> don't understand is why, given your presumed abilities in one sphere
> (although just tossing out names of theorems/practices etc almost at
> random is an odd way of showing it) you need to go to such lengths as
> you do to "prove" it. Hectoring, bullying, denigration and insult are
> the tools of the foolish and insecure, not the wise.
You should read the ENTIRE thread to understand the role of a MATRIX
played in this subthread, and the number of posts in which posters
who know next to NOTHING about the subject kept making NOISE about
it, just as you do now.
In the subthread in question, Frances Hinden was talking about the
ranking of bridge players, with a method that involved the INVERSION
of several MATRICES.
I asked for more information about those matrices (which she said
she would provide, perhaps in a couple of weeks) and the "singularity"
question, and mentioned (to another poster) that it can easily be
answered by computing the determinant, WITHOUT knowing the singular
values (which are not necessarily less prone to roundoff) but are
not as easily computed. Then came a parade of NOISE MAKERS:
First, morons say I don't even know what a matrix is. Village
Idiot John Crinnion led the parade with his 6 (SIX) posts
of vacuous substance and insults.
Next, Thomas Dehn, who apparently do know some math, wrote:
TD> The determinant is, for all practical purposes, completely
TD> useless to check whether a matrix is singular, as its
TD> computation is vulnerable to round-off errors.
TD> A singular value decomposition approach would be much more
TD> reliable than trying to compute the determinant.
I have already posted my full reply to Dehn's erroneous assertions.
Next, John D'Errico came, a week after Dehn, with his insults and
blissful ignorance, even after seeing my reply to Dehn:
JD> Sorry, you gave yourself away as a mathematical poser
JD> by simply suggesting the determinant.
JD>
JD> I'll admit that you may have taken a class as an
JD> undergrad, maybe even used some linear algebra in
JD> the dark past.
John D'Errico is nothing but a MatLab jock, and it's doubtful
that he is even a mathematician, for all the insults he tossed.
So, my reply to him was exactly what he deserved.
Now comes Ian Payne, another rgb Village Idiot with his foot in mouth:
>
> Another reason that you aren't accorded the credibility that you think
> you deserve (as if your tone wasn't reason enough) is that your English
> is bad, or, to give you the benefit of the doubt, careless.
But you understood my careless English, didn't you, Ian? It's the
REST (the thread and SUBSTANCE of the discussion) to which you are
blissfully oblivious and of which you're blissfully ignorant!
> > Johnny Boy, you wouldn't by chance trying to compete against the
> > other John (Crinnion) for the title of "Village Idiot of math
> > in rgb", would you?
>
> ++++I don't think John Crinnion was "Village Idiot of math", I think he
> was just "Village Idiot". Whether he is directly answerable to "Chief
> Moron" Gordon Rainsford is unclear. I don't think I'll bother with
> another "Confusing morons with Mormons" joke. The last one fell
> completely flat.
I think you've just thrown you hat into the ring to compete for the
title of "Chief Village Idiot" amongst all the Village Idiots in rgb.
> > I got the answer in my head by mentally calculating the det(M)
> > even before you can type the elements of the matrix into your
> > computer.
>
> ++++At the risk of sounding like a teenager, so what?
No, you sound worse than a teenager. What you cited was completely
relevant in the CONTEXT of John Dehn's and John D'Errico's
erroneous assertions about the use of the determinant and that
the computation of the det is "vulnerable to roundoff errors",
as if other numerical methods aren't.
When YOU (Ian) read something OUT OF CONTEXT, and is too LAZY
and/or too IGNORANT to find out the context, yet plunge into
dishing out YOUR insults (without provocation, cause, or
substance), THAT's what make you a Village Idiot, Ian Payn!
> Someone challenged
> your credibility as a mathematician and you type out some endless
> screed that means nothing to rather a lot of r.g.b. readers,
Then those rgb readers don't have to read my replies, and they
certainly should have kept their feet on the ground instead of
inserting them in their mouths, as Ian Payn and D'Errico did,
after John Crinnion has finally caught on and ceased to do so.
> This is a newsgroup for bridge discussion.
So it is. So is the thread in which the subthread of MATRIX
INVERSION came into play in Frances Hinden's post which I am not
the only one interested in the mathematical aspects of her
formulation and solution.
> Why not drop this whole ego-trip and look at another thread or two? Give
> your opinion on an auction. Deride an opening lead. Agree with David
> Stevenson about something (that last one takes a bit of doing, I'll
> grant you).
Ian, who appointed you to be the rgb Cop to tell anyone what they
should or shouldn't post or discuss? Read my reply to Village Idiot
Crinnion about Usenet and ngs.
Why don't YOU drop the foot from your mouth, and simply try to
contribute something (ANYTHING) of Bridge substance to this ng,
rather than being an officious arse that you are?
> You might find, after a while, that you get some people
> agreeing with you, or having amicable discussions with you at least.
Why should I care if those who DISAGREE with what I wrote was solely
caused by THEIR ignorance about the subject matter?
What is amicable or non-amicable discussions is a matter of quid
pro quo. It takes TWO to have mutual respect. It takes only ONE
Idiot to earn disdain from his/her idiotic posts.
> That would be a much more constructive use of everyone's time, wouldn't it?
>
> IP.
So, what constructive use of YOUR time have you accomplished by
your gratuitous rant, Ian?
Take your FEET and put them where they belong, out of your mouth!
-- Bob.
Thomas Dehn
"Reef Fish" <Large_Nass...@Yahoo.com> wrote:
> "Thomas Dehn" <thomas...@arcor.de> wrote in message
news:<bkdb3p$hg3a$1...@ID-57266.news.uni-berlin.de>...
> > "Reef Fish" <Large_Nass...@Yahoo.com> wrote:
> > > fhi...@orange.net (Frances Hinden) wrote in message
> > news:<d30df517.0309...@posting.google.com>...
> > > in response to "Steve Willner" <swil...@cfa.harvard.edu>
> > > > er... not quite now. I had a quick look back, and can't find any of
> > > > the stuff we did (too many computer & house moves).
> > > >
> > > > To answer the singularity question, we were not exactly inverting but
> > > > (if I remember correctly) searching for eigenvalues of a symmetric
> > > > matrix. We used the practical check of seeing if any were zero;
> > > > though I think we also proved that our matrix was positive definite
> > > > (either that or we tried to & didn't manage it).
> > >
> > > To check for singularity, the use of the determinant would have been
> > > much simpler, wouldn't it?
> >
> > The determinant is, for all practical purposes, completely
> > useless to check whether a matrix is singular, as its
> > computation is vulnerable to round-off errors.
>
> REALLY? Surely you don't mean to say that other numerical
> linear algebra computational methods are not vulnerable to
> roundoff errors!
There exist methods for deciding whether a matrix
is singular which are significantly less vulnerable
to roundoff errors than computing the determinant.
> The vulnerability to roundoff errors depends PRIMARILY on:
> 1. The precision of the computational software, and 2. How
> ill-conditioned the matrix is.
Yep, but
- upping the precision of your software to the number of
digits needed to avoid round off errors can well mean
that an algorithm needs ten years to compute what
a smarter algorithm can compute in ten minutes.
An algorithm with performs O(n^4) or even slower
- n being the size of the matrix - is
utterly useless except for pet examples.
- the condition number ultimately measures the distance
to the next singular matrix. Hence, those matrices which are
ill-conditioned are those where it is necessary to decide
whether they are non-singular.
> Your statement about the "completely useless" aspect of the
> determinant showed a sign of naivete in theory and practice.
I think that everything you write on rgb shows that you have,
at best, undergraduate knowledge of mathematics applied to
the real world. There are quite a few real mathematicians
on rgb, and we get a good laugh reading your superficial
knowledge.
> > A singular value decomposition approach would be much more
> > reliable than trying to compute the determinant.
>
> This broad stroke blanket statement is patently false!
Write a program which uses a general-purpose algorithm
to compute the determinant of any matrix. Then run that
program against a 1000x1000 Hilbert matrix. Then
compare the result computed by your program to the
actual determinant of the 1000x1000 Hilbert matrix,
by looking up the latter in a table. (1000x1000 is
still very small, compared to the real life size of models,
usually millions of elements in an FE model).
If you have time to waste, do two variations of the program,
one which takes double precision numbers as input, and
one which somehow is able to process the exact 1000x1000
Hilbert matrix.
Oh, and I have a question. I have a 2x2 matrix whose
determinant is 10^-10. Is the matrix close to being singular,
or not? (So much for using the determinant at all
to decide whether a matrix is singular)
> > You might want to have a look at Golub - van Loan's
> > "Matrix computations".
>
> No need. I have read dozens of books and journal articles on
> the subject. I also have published books, and written scientific
> papers in mathematical, statistical, and computational journals on
> statistical subjects in which "matrix computations" and especially
> the NUMERICAL precision of matrix computations are of fundamental
> importance.
Let me guess: your "scientific papers" were rejected by the reviewers.
Assuming that these "scientific papers" exist at all.
Thomas
Sid Ismail
wrote:
: There exist methods for deciding whether a matrix
: is singular which are significantly less vulnerable
: to roundoff errors than computing the determinant.
"less vulnerable" ... in a bridge newsgroup... ???
Sid
Mike Vaughn
Obviously you don't appreciate German humor :-)
--
Mike Vaughn
"A ship is safe in a harbor --
But that is not what ships are built for."
Peter Clinch
Apropos of nothing in particular - in these modern times, is is still
true that constipated mathematicians first try to work everything out
with a pencil, and failing that resort to logs?
Peter.
Peter Clinch
In The Village.
What do you want?
Information.
Which side are you on?
That would be telling. We want information, information, information...
You won't get it.
By hook or by crook we will.
Who are you?
The new Number Two.
Who is Number One?
You are Number Six.
I am not a number. I am a free man.
Ha, ha, ha, ha.
Reef Fish
"Thomas Dehn" <thomas...@arcor.de> wrote in message news:<bl3fsj$7l2be$5...@ID-57266.news.uni-berlin.de>...
TD > The determinant is, for all practical purposes, completely
TD > useless to check whether a matrix is singular, as its
TD > computation is vulnerable to round-off errors.
TD > There exist methods for deciding whether a matrix
TD > is singular which are significantly less vulnerable
TD > to roundoff errors than computing the determinant.
Both statement are patently false. So I explained WHY they are
false in general, and these FACTS:
RF> The vulnerability to roundoff errors depends PRIMARILY on:
RF> 1. The precision of the computational software, and 2. How
RF> ill-conditioned the matrix is.
TD > A singular value decomposition approach would be much more
TD > reliable than trying to compute the determinant.
RF > This broad stroke blanket statement is patently false!
I then extended to Mr. Dehn the courtesy of asking:
RF> Just a matter of curiosity -- what field (and what areas within
RF> that field) is your specialty that involves NUMERICAL linear
RF> algebra and matrix computations?
so that I can better judge the level of his expertise on the
subject as well as the appropariate level to address his reply.
Mr. Dehn came back without addressing the SUBSTANCE of what I had to
say, nor did he answer the preceding question about his qualification.
Instead, he resorted to insults (which are by now the modus operandi
of all posters who responded to my posts in this thread, with the
exception of Frances Hinden whose post prompted my questions), such as
> Let me guess: your "scientific papers" were rejected by the reviewers.
> Assuming that these "scientific papers" exist at all.
>
>
> Thomas
Okay, I'll respond to your present post point-by-point, but only after
you tell me what I asked you about your field of expertise, AND after
you respond to the same questions I posed to John D'Errico, because
these question address precisely where you (and he) erred:
RF> Let M be the 3 x 3 (symmetric) matrix with elements
RF>
RF> ( 7 3 5 )
RF> M = ( 3 1 2 )
RF> ( 5 2 1 )
RF>
RF> I can do BOTH the determinant and the inverse of M in my head
RF> (without pencil, paper, or any calculator) -- with no roundoff
RF> error of any kind.
The comment was directed toward your rash statements about the
uselessness of the determinant, about its vulnerability to roundoff,
etc.
RF> 1. What are the eigenvalues (singular values) of M?
RF> 2. I assume you know the determinant of M is the product of
RF> these eigenvalues. What value do you get for the determinant
RF> computed this way (using the product of the singular values)?
RF> 3. WHY do you NEED the eigenvalues of M to find the INVERSE of
RF> M which I can do in my head?
This question is particularly relevant to the present discussion, because
in Frances Hinden's post about her method of ranking players, the ONLY
requirement is the computation of INVERSES of matrices -- nothing about
the necessity of singular values (eigenvalues)!
FH> I once spent a long time creating an individual ranking system with
FH> Jonathan Mestel for a club based on a long history of club results.
FH> It involved inverting quite a few matrices.
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
RF> 4. What is the inverse of M?
Does Inverse(M) in anyway depend on your answer to (1) and (3)?
I almost forgot what I said to John D'Errico about those problems.
I offer you the same choice of your favorite program/algorithm:
RF> you can use whatever computer program and whatever numerical
RF> method you choose:
RF> Let see what kind of "mathematician" you are in your answers to
RF> the 4 questions above.
It's your turn, Mr. Dehn.
-- Bob.
Reef Fish
> Where am I?
A Clueless Newbie <tm> What else?
You are being Flamed Because:
[X] You posted a blatantly obvious troll
[X] Your life time posting history too short to try to be cute
[X] You aspire to be a r.g.b. Village Idiot when you grow up
[X] You didn't do anything specific, but appear to be so
generally worthless that you are being flamed anyway
To Repent, You Must:
[X] Actually post something relevant
[X] Be the guest of honor in alt.flame for a month
[X] Give up your Yahoo account
[X] Smash you modem with a hammer and eat it
[X] Tell your Mommy you've been a bad boyl
[X] Run your monitor/hair drier/toaster in the bath
In Closing, I'd Like to Say:
[X] Get a life
[X] Yo Mama was soooooo ugly ...
[X] Age 10 more years before posting again
[X] Get lost, you pathetic loser
Julian Lighton
Reef Fish <Large_Nass...@Yahoo.com> wrote:
>double...@yahoo.com (Peter Clinch) wrote in message
>news:<50e18684.03092...@posting.google.com>...
>
>> Where am I?
>
>A Clueless Newbie <tm> What else?
>
>You are being Flamed Because:
> [X] You posted a blatantly obvious troll
Actually, he posted a joke. You failed to get it. This is about par
for the course.
> [X] Your life time posting history too short to try to be cute
For somebody who has only recently graced this newsgroup with his
presence, you're awfully quick to label others, who have been here
noticably longer than you, as newbies.
>To Repent, You Must:
> [X] Give up your Yahoo account
Heh.
>In Closing, I'd Like to Say:
Flame checklists were old ten years ago, and yours is hardly an
exemplar of the species.
John Crinnion
> ...
>
> <SNIPPED BOASTING, LONG MATHS QUESTION AND SOME MORE BOASTING from Reef Fish>
>
> <SNIPPED TREMENDOUS GOOD SENSE FRON IAN PAYN>>
> > other John (Crinnion) for the title of "Village Idiot of math
> > in rgb", would you?
>
> ++++I don't think John Crinnion was "Village Idiot of math", I think he
> was just "Village Idiot". Whether he is directly answerable to "Chief
> Moron" Gordon Rainsford is unclear. I don't think I'll bother with
> another "Confusing morons with Mormons" joke. The last one fell
> completely flat.
>
I am the Village Idiot for maintaining that 'average' is a generic
term for 'mean', 'median' and 'mode'. This happens to be one of the
senses given in the Merriam-Webster dictionary, which (unfortunately)
is wrong - according to Reef Fish.
Why has poor benighted Gordon Rainsford been dragged into this?
Could we agree on David Stevenson as a compromise candidate for Chief
Moron?
John D'Errico
Large_Nass...@Yahoo.com (Reef Fish) wrote:
> John D'Errico <dropthispa...@flare.net.thisonetoo> wrote in message
> news:<dropthispart_derrico-...@news.newsguy.com>...
> > In article <8fb7380b.0309...@posting.google.com>,
> > Large_Nass...@Yahoo.com (Reef Fish) wrote:
> >
> > > "Thomas Dehn" <thomas...@arcor.de> wrote in message
> > > news:<bkdb3p$hg3a$1...@ID-57266.news.uni-berlin.de>...
> >
> > > > You might want to have a look at Golub - van Loan's
> > > > "Matrix computations".
> > >
> > > No need. I have read dozens of books and journal articles on
> > > the subject. I also have published books, and written scientific
> > > papers in mathematical, statistical, and computational journals on
> > > statistical subjects in which "matrix computations" and especially
> > > the NUMERICAL precision of matrix computations are of fundamental
> > > importance. Singular value decomposition and eigenvalue analysis
> > > are basic in these areas.
> >
> > Sorry, you gave yourself away as a mathematical poser
> > by simply suggesting the determinant.
>
>
> Geez, this ng is definitely not in short supply of Village Idiots
> in mathematics related topics, isn't it?
I won't stoop to your level.
> > I'll admit that you may have taken a class as an
> > undergrad, maybe even used some linear algebra in
> > the dark past.
>
> Can't you read the paragraph you cited above? I taught my first
> GRADUATE course in Numerical Analysis probably before you were born!
Gosh. Who cares? It just means you have not learned anything
since the 1950's. Something you should learn is that most
books on numerical analysis written that long ago are
virtually worthless now. Maybe ok as kindling.
As for the citations, all I heard were some vague claims,
loudly expressed.
> > Its a wee bit funny though when someone
> > makes it so obvious that they are not as good as they
> > loudly proclaim. Merely using the words "singular value
> > decompsition" and "sparse matrix" in moderately coherant
> > sentences do not a mathematician make.
>
> That is correct, and certainly applies to YOU. So, let me pose
> a trivial little matrix problem for you.
>
>
> Let M be the 3 x 3 (symmetric) matrix with elements
>
> ( 7 3 5 )
> M = ( 3 1 2 )
> ( 5 2 1 )
>
> I can do BOTH the determinant and the inverse of M in my head
> (without pencil, paper, or any calculator) -- with no roundoff
> error of any kind.
Again. Who cares?
Arithmetic. You make yourself appear more ridiculous
with every statement.
You are kind of funny, but I've no more interest in your
antics.
Plonk.
--
My real address sounds like "derrico ahht flare dawt net"
Reef Fish
> > > Sorry, you gave yourself away as a mathematical poser
> > > by simply suggesting the determinant.
> > > I'll admit that you may have taken a class as an
> > > undergrad, maybe even used some linear algebra in
> > > the dark past.
> > > Its a wee bit funny though when someone
> > > makes it so obvious that they are not as good as they
> > > loudly proclaim. Merely using the words "singular value
> > > decompsition" and "sparse matrix" in moderately coherant
> > > sentences do not a mathematician make.
John D'Errico, you sure did a lot of mouthing and dished out
insults freely, but as soon as I posted a few simple questions
that would have put you in your place, on a very concrete example
that addressed ALL of the key points EXPLICITLY ...
> > That is correct, and certainly applies to YOU. So, let me pose
> > a trivial little matrix problem for you.
> >
> >
> > Let M be the 3 x 3 (symmetric) matrix with elements
> >
> > ( 7 3 5 )
> > M = ( 3 1 2 )
> > ( 5 2 1 )
> >
> > I can do BOTH the determinant and the inverse of M in my head
> > (without pencil, paper, or any calculator) -- with no roundoff
> > error of any kind.
>
> Again. Who cares?
What did you do? ...
You snipped the 4 questions I asked you, tucked your tail between
your legs and started running ... yet kept on insulting though you
have absolutely nothing to show about ANY knowledge on the subject!
>
> Arithmetic. You make yourself appear more ridiculous
> with every statement.
>
> You are kind of funny, but I've no more interest in your
> antics.
>
> Plonk.
Sure, do you even know what an "eigenvalue is"? I would like to see
you do the arithmetic to answer my FIRST question.
Go away, you phoney!
Now we'll wait and see what John Dehn has to show for the insults
he dished out. :-) You have flunked miserably.
-- Bob.
Ian Payn
John Crinnion wrote:
>
> Why has poor benighted Gordon Rainsford been dragged into this?
++++I don't know. Nor does he.
> Could we agree on David Stevenson as a compromise candidate for Chief
> Moron?
++++Now there's an idea. In my official capacity as self-appointed
village cop/idiot I will approach DWS and ask him if he's willing to
stand. It may take some while, however, as, with my foot (apparently) in
my mouth all of the time it's a bit hard to get a conversation going.
Ian Payn
Reef Fish wrote:
<DIDN'T REALLY READ MUCH OF THE REST, SO CAN'T COMMENT>
> Ian, who appointed you to be the rgb Cop to tell anyone what they
> should or shouldn't post or discuss?
++++Well, nobody, but then I don't tell people what to do. I did suggest
that you dropped the pointless arguments and got involved in other
threads related more to bridge than mathematics.
Read my reply to Village Idiot
> Crinnion about Usenet and ngs.
++++I can't imagine why you think anyone would have any interest in what
you have written before. It's not as though anything you have to write
on the subject of usenet and newsgroups has any authority at all.
> Why don't YOU drop the foot from your mouth, and simply try to
> contribute something (ANYTHING) of Bridge substance to this ng,
> rather than being an officious arse that you are?
++++Now who's being "The Cop"? You're doing what I did. Only I was being
polite, and trying to be helpful. As for contributing anything of bridge
substance to this newsgroup, I think my record is rather better than yours.
> What is amicable or non-amicable discussions is a matter of quid
> pro quo. It takes TWO to have mutual respect. It takes only ONE
> Idiot to earn disdain from his/her idiotic posts.
++++As fine an example as one could wish for of one hoisting oneself on
ones own petard. Game, set and match to me, I think.
This correspondence is now, as far as I'm concerned, closed. Write what
you like. My views have been expressed, and I've tried to be helpful but
even my almost legendary tolerance has been worn out. I don't care what
you think, what you write or what you do. Many regular r.g.b.
contributors have tried to reason with you, some who know (or care) what
a matrix is, some who don't. All you've done is spew forth
not-very-amusing venom and droned on about mathematical terms, probably
in an attempt to befuddle or impress.
Carry on. Be my guest. Enjoy yourself. I'm bored with rattling your
cage, because the results are never interesting. Try and learn some new
invective for your next set of tirades.
Reef Fish
been placed in appropriate SUBJECTS for their NOISE in the serious
thread of
Subject: Re: A problem in matchpoint scoring
Ian Payn <Ian....@charterchambers.com> wrote in message news:<3f77efa4.0@entanet>...
who already EARNED his honor in the thread:
Subject: Re: Ian Payn the Village Idiot (was
Re: A problem in matchpoint scoring)
further polluted the original thread by going back to the matchpoint
thread to spew forth more noise with John Crinnion who had ALSO
earned his own honor in a thread titled:
Subject: Re: John Crinnion the Village Idiot (was
Re: A problem in matchpoint scoring)
This paragraph is what EARNED Ian Payn the new SUBJECT title, the
Self-Portrait of a Village Idiot!
This Self-Portrait is especially significant, because Ian Payn had
just posted in his OWN thread "Ian Payn the Village Idiot" this:
QUOTE Ian Payn -----------------------------------------------
My views have been expressed, and I've tried to be helpful but
even my almost legendary tolerance has been worn out. I don't care what
you think, what you write or what you do. Many regular r.g.b.
contributors have tried to reason with you, some who know (or care) what
a matrix is, some who don't. All you've done is spew forth
not-very-amusing venom and droned on about mathematical terms, probably
in an attempt to befuddle or impress.
ENDQUOTE Ian Payn ---------------------------------------------
Ian, we can now address you by a title perhaps more to your liking, a
Sanctimonious Hypocrite ?
I have cited Ian's entire post. I am still looking for any BRIDGE
content in any of his posts, or ANY contribution to this ng, in
this and related threads in which he starred as a Village Idiot!
Congratulations, Ian.
Your Idiotic FAME in r.g.b. and Usenet ngs is now secure!
John Crinnion -- you can respond to Ian in this NEW thread now.
-- Bob.
Bill Bickford
alleged degree; it is abundantly obvious you could not get one at a
reputable college or university.
Cheers............................/Bill Bickford
Reef Fish
You're talking to Ian Payn or John Crinnion, I presume.
Are you envious of their Village Idiot titles and wants to apply for
one yourself too?
-- Bob.
james
I think Reef Fish was a Professor of Statistics who has become
somewhat ecccentric in his old age.Every university town has a few
such characters.
James
Bill Bickford
similar question.
Cheers................................./Bill Bickford
"Reef Fish" <Large_Nass...@Yahoo.com> wrote in message
news:8fb7380b.0309...@posting.google.com...
Julian Lighton
james <jame...@hotmail.com> wrote:
>I think Reef Fish was a Professor of Statistics who has become
>somewhat ecccentric in his old age.Every university town has a few
>such characters.
If you want to see a professor gone eccentric, search Google (or
Google groups) for "alexander abian", one of the classic Usenet kooks.
Reef Fish has just gone obnoxious. (Assuming he wasn't always so.)
Reef Fish
> Reef Fish, this was directed directly and solely to you as was my previous
> similar question.
>
> Cheers................................./Bill Bickford
>
> > > You never did answer my question as to how much you had to pay for your
> > > alleged degree; it is abundantly obvious you could not get one at a
> > > reputable college or university.
> > >
> > > Cheers............................/Bill Bickford
> >
> >
> > You're talking to Ian Payn or John Crinnion, I presume.
Bill the Village Idiot,
If I had answered it earlier (and chose to do it), you and others would
have (a) accuse me of being presumptuous that it was addressed to me;
and (b) that whatever answer I gave would have been "boasting", just
as the rest of the answers I had given as a matter of fact to those
who chose to insult me publicly.
So here's the answer to you Bill the IDIOT:
I had to pay NOTHING for EARNING my Ph.D. degree at Yale. The
nationally competed NDEA Fellowship not only paid for my entire
tuition and fees, it paid for all my LIVING EXPENSES as well.
If you're not so obnoxious and dumb, you could find out how much
that total cost is, in today's dollar or Pound Stirling value.
> > Are you envious of their Village Idiot titles and wants to apply for
> > one yourself too?
> >
> > -- Bob.
You have EARNED your Idiotship now, Bill Bickford. Enjoy.
-- Bob.
Reef Fish
I take it as a compliment, James.
I think you're thinking of eccentric mathematicians such as Paul Erdos
(met him several times), John Nash (before my time, in movie "A
Beautiful Mind"), and other geniusses in that field who turned out
to be eccentric.
Nah, I wouldn't dream of putting myself in their company in terms of
eccentricity (I use is only for ellipses). Besides, I consider them
to SHALLOW in their interests, as brilliant as they are in one field.
Sort of a high class "Idiot Savant". :-))
Thanks just the same, James.
Julian Lighton the aspiring Village Idiot chimed in with,
JL> Reef Fish has just gone obnoxious. (Assuming he wasn't always so.)
Why assume anything, Julian? Just learn to use the ng archives,
and you'll know FOR SURE that I have always been obnoxious to the
"mental midget", morons, imbecils, and idiots in newsgroups who
chose me as the target of their idiocy!
If you keep following the other idiots in this ng, Julian, and
make your own gratuitous remarks when you should have heeded the
saying "Silence is Golden" on those occasions, you'll soon get your
own TITLE too (as the subject heading of the present one).
cheerio,
-- Bob.
Julian Lighton
Reef Fish <Large_Nass...@Yahoo.com> wrote:
>jame...@hotmail.com (james) wrote in message
>news:<e5870d4.03093...@posting.google.com>...
>> somewhat ecccentric in his old age.Every university town has a few
>> such characters.
>
>I take it as a compliment, James.
>
>I think you're thinking of eccentric mathematicians such as Paul Erdos
>(met him several times), John Nash (before my time, in movie "A
>Beautiful Mind"), and other geniusses in that field who turned out
>to be eccentric.
Erdos was eccentric. Nash was nuts. You fail to rise above "annoying".
>Nah, I wouldn't dream of putting myself in their company in terms of
>eccentricity
I doubt you're in their league in intellect, either. No such evidence
is on display here.
>Julian Lighton the aspiring Village Idiot chimed in with,
>
>JL> Reef Fish has just gone obnoxious. (Assuming he wasn't always so.)
>
>Why assume anything, Julian? Just learn to use the ng archives,
>and you'll know FOR SURE that I have always been obnoxious to the
>"mental midget", morons, imbecils, and idiots in newsgroups who
>chose me as the target of their idiocy!
Frankly, you're not amusing enough to be worth the time required for
digging for past posts in the archives.
>If you keep following the other idiots in this ng, Julian, and
>make your own gratuitous remarks when you should have heeded the
>saying "Silence is Golden" on those occasions, you'll soon get your
>own TITLE too (as the subject heading of the present one).
I doubt you can comprehend the depths of the terror I suffer from that
threat.
Micha Keijzers
[snip]
>
> I think you're thinking of eccentric mathematicians such as Paul Erdos
> (met him several times), John Nash (before my time, in movie "A
> Beautiful Mind"), and other geniusses in that field who turned out
> to be eccentric.
>
> Nah, I wouldn't dream of putting myself in their company in terms of
> eccentricity (I use is only for ellipses). Besides, I consider them
> to SHALLOW in their interests, as brilliant as they are in one field.
One thing I seem to remember from "A Beautiful Mind", but one could
possibly debate whether the movie was really representable for his life,
was that John Nash was working on more subjects than game theory alone.
[snip]
>
> If you keep following the other idiots in this ng, Julian, and
> make your own gratuitous remarks when you should have heeded the
> saying "Silence is Golden" on those occasions, you'll soon get your
> own TITLE too (as the subject heading of the present one).
Are you going to call everyone on this ng an idiot. It sure begins to
look like it. You'd be the only non-idiot. And guess what we think of
that... ;-)
>
> cheerio,
Likewise, Micha
>
> -- Bob.
james
> > I think Reef Fish was a Professor of Statistics who has become
> > somewhat ecccentric in his old age.Every university town has a few
> > such characters.
>
> I take it as a compliment, James.
>
> I think you're thinking of eccentric mathematicians such as Paul Erdos
> (met him several times), John Nash (before my time, in movie "A
> Beautiful Mind"), and other geniusses in that field who turned out
> to be eccentric.
>
> Nah, I wouldn't dream of putting myself in their company in terms of
> eccentricity (I use is only for ellipses). Besides, I consider them
> to SHALLOW in their interests, as brilliant as they are in one field.
>
> Sort of a high class "Idiot Savant". :-))
>
> Thanks just the same, James.
>
Do you still do any academic work? It would be a shame if you were
wasting your talents on usenet.
James
Reef Fish
No.
When it reached the point that the university adminstrators are more
interested in placating morons (such as a few we've seen here :-)
rather
than upholding a minimum level of educational "standard", that was
when I took an early retirement, and quit "cold turkey" you might say.
My scientific journals and most of my text and reference books are now
scattered in the libraries of 10 universities in China and one in
India.
The students and library users there need them much more than I do.
> It would be a shame if you were wasting your talents on usenet.
>
> James
James, even if you intended that to be a sarcasm, I'll respond to it
my way. The term "waste" is in the eye of the beholder, as well as
other who behold the "waste".
I am the antithesis of an "Idiot Savant", of the low-brow or the high-
brow kind. I am interested in many subjects throughout my
productive professional life.
Those who have read my posts in some less visible LISTSERV lists had
long known that I am already into my 7th or 8th life, in terms of
activities in which I have studied and practiced in depth -- and
that's not counting "mathematics" or "statistics". Bridge is only
ONE of those.
As for "wasting my talents" on usenet:
1. It requires NO talent, and hence it wastes no talent.
2. When I repeatedly find in usenet newsgroups, posters who are
supposedly "educated" people, yet not only do not know the
simplest
term in statistics that are used everyday ("mode" and "median"
to name just two), that is taught in a required course of
EVERY curriculum in even the 2nd and 3rd rate universities today,
I don't consider my TIME (though it required no talent whatsoever)
wasted, by educating the public (while kicking the arses of those
Village Idiots who should have known better not to disrupt valid
and legit discussions because of their own ignorance) that there's
no room for their noise-making, even in a RECREATIONAL newsgroup.
3. I don't consider anything I do as a "waste of my talent", or else
I wouldn't have done it. As I often lecture the obtuse, this is
a Free Market in a Free Country, and a public forum that supports
the Freedom of Speech, everyone is entitled to let their views
known whether they merely made fools of themselves or contribute
a small part to the world at large.
That's all I have time, for my "sermon of the day". :-)
-- Bob.
Henry Raymond
"Reef Fish" <Large_Nass...@Yahoo.com> wrote in message
Apparently English wasn't one of your subjects at Yale. I haven't seen such
an abundance of bad grammar in one post in quite some time. Bravo!
Jeremy Rickard
> 2. When I repeatedly find in usenet newsgroups, posters who are
> supposedly "educated" people, yet not only do not know the
> simplest
> term in statistics that are used everyday ("mode" and "median"
> to name just two),
I've not noticed any people like that. People on r.g.b. have said that
they believe that the statement "the mode and median are kinds of
average" is true, because they disagree with you about the definition
(or definitions) of the word "average". But I think you'll find that
all of these people know what the terms "mode" and "median" mean.
Jeremy.
--
Jeremy Rickard
Email: j.ri...@bristol.ac.uk
WWW: http://www.maths.bris.ac.uk/~pure/staff/majcr/
Bill Bickford
is obviously an out and out lie. I doubt you would even have the chutzpah
to apply to Yale. Any degree you claim to have could only have been from
some fraudulent offer on the Internet.
Cheers............................/Bill Bickford
"Reef Fish" <Large_Nass...@Yahoo.com> wrote in message
news:8fb7380b.03093...@posting.google.com...
Reef Fish
Clueless Henry copied 100 lines just to say:
> Apparently English wasn't one of your subjects at Yale. I haven't seen such
> an abundance of bad grammar in one post in quite some time. Bravo!
I had already replied to this kind of noise in r.g.b. Below is the
verbatim clip, with only "Henry Raymond" replacing the name previously
used:
======== begin excerpt
It was simply a case of MORONS like <Henry Raymond> who didn't
understand anything of substance being discussed, and chose to make
his usual NOISE by picking on typos, spelling, grammar, sentence
structure, etc. when there was no ambiguity caused by any of them.
Is he a low-level copy editor for some Brit tabloid?
To <Henry Raymond> and his likes -- typos and gender misreference
seldom create and problem of comprehension on a subject matter, and
are usually self-corrected or easily corrected in the next
occurence of such. OTOH, as Sir Winston Churchill said to a female,
something to the effect, "Madam, if I am drunk, I shall be sober
tomorrow; but you'll still be ugly." :-) The analogue to
<Henry Raymond> would be "Sir, if I made a typo slip, it'll be
self-corrected tomorrow; but the lack of brain cells in your head
may take generations to correct."
======== end excerpt
-- Bob.
Reef Fish
> Still refusing to answer? I note you claim to have a degree from Yale; this
> is obviously an out and out lie.
Bill Bickford,
You are a MUCH BIGGER idiot than I had thought you were.
Even the other Village Idiots here are laughing at you now.
> doubt you would even have the chutzpah
> to apply to Yale. Any degree you claim to have could only have been from
> some fraudulent offer on the Internet.
>
> Cheers............................/Bill Bickford
John Crinnion,
You have just lost your Chief Village Idiot title to Mr. Bickford.
-- Bob.
jimf...@cox.net
news:vnkv5th...@corp.supernews.com...
> In article <8fb7380b.03093...@posting.google.com>,
> Reef Fish <Large_Nass...@Yahoo.com> wrote:
> >jame...@hotmail.com (james) wrote in message
> >news:<e5870d4.03093...@posting.google.com>...
> >> I think Reef Fish was a Professor of Statistics who has become
> >> somewhat ecccentric in his old age.Every university town has a few
> >> such characters.
> >
> >I take it as a compliment, James.
> >
> >I think you're thinking of eccentric mathematicians such as Paul Erdos
> >(met him several times), John Nash (before my time, in movie "A
> >Beautiful Mind"), and other geniusses in that field who turned out
> >to be eccentric.
>
> Erdos was eccentric. Nash was nuts. You fail to rise above "annoying".
Nash *is* whatever he is. It is amazing how many people think he is dead but
those rumors are false.
Mmbridge
Julian Lighton
I knew he was alive. My understanding is that he is more or less no
longer nuts.
james
No sarcasm was intended.I was just hoping you might tell us about yourself.
I am sorry you had some troubles with the university administrators.
James
Reef Fish
> Large_Nass...@Yahoo.com (Reef Fish) writes:
>
> > 2. When I repeatedly find in usenet newsgroups, posters who are
> > supposedly "educated" people, yet not only do not know the
> > simplest term in statistics that are used everyday ("mode" and
> > "median" to name just two),
>
> I've not noticed any people like that. People on r.g.b. have said that
> they believe that the statement "the mode and median are kinds of
> average" is true, because they disagree with you about the definition
> (or definitions) of the word "average". But I think you'll find that
> all of these people know what the terms "mode" and "median" mean.
>
> Jeremy.
Even Thomas Dehn, who has since been exposed as an IMPOSTER of a
mathematician,for having extended his limited knowledge about matrices
to hurl insults and was caught red-handed, was correct about THIS:
TD> I strongly doubt that common english language
TD> usage would refer to 10 as being the "average" of
TD> 1,10,10,3, and 10. The M-W definition for average
TD> simply is crap.
Mike Vaughn was both spotless correct as well as perceptive about
the probable cause of this wide-spread IGNORANCE:
MV> There are non-statistical uses for the word 'average', but
MV> in a statistical context, the use should be correct.
MV>
MV> One hopes the M-W editors are (were) not _all_ 1950s liberal
MV> arts majors for whom ignorance of mathematics was worn as a
MV> badge of honor.
Jeremy, you are apparently still wearing this "badge of honor" of
the 1950s.
This is a new CENTURY, and a new MILLENNIUM now. Get an education,
Jeremy.
-- Bob.
Thomas Dehn
"Julian Lighton" <jl...@fragment.com> wrote:
> In article <bldppe$b3aom$1...@ID-57266.news.uni-berlin.de>,
> Thomas Dehn <thomas...@arcor.de> wrote:
> >x-no-archive: yes
> >
> >"Reef Fish" <Large_Nass...@Yahoo.com> wrote:
> >> I had to pay NOTHING for EARNING my Ph.D. degree at Yale. The
> >> nationally competed NDEA Fellowship not only paid for my entire
> >> tuition and fees, it paid for all my LIVING EXPENSES as well.
> >
>
> Don't insult dogs.
"On the internet, nobody knows you're Bob Silverman.
Except if you tell them."
Better? ;-)
Thomas
Reef Fish
Thank you, James, for your clarification.
As for the university administrators, it is pretty well-known and
widely-known that MANY of them (in the USA anyway), are the lowest form
of creatures in any "educational institution" that is "educational"
in reality.
-- Bob.