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 Historical content remains viewable.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

Re: WHAT HAPPENS IF YOU CHANGE THE DIAGONAL TO THE ANTIDIAGONAL?

54 views
Skip to first unread message

Graham Cooper

unread,
Nov 9, 2012, 4:13:09 PM11/9/12
to
Why this is marked as abuse? It has been marked as abuse.
Report not abuse
On Nov 9, 7:18 pm, forbisga...@gmail.com wrote:
> On Thursday, November 8, 2012 10:18:43 PM UTC-8, Graham Cooper wrote:
> > On Nov 9, 4:03 pm, forbisga...@gmail.com wrote:
> > > On Thursday, November 8, 2012 8:38:12 PM UTC-8, Graham Cooper wrote:
>
> > > > You change the DIAGONAL to the ANTI-DIAGONAL and no existing real can
> > > > be left on the list!
>
> > > Tell me how to produce the antidiagonal of an infinite list.
>
> > One method is to input a stream of digits and interpret them as a
> > transpose sort of the list.
>
> Wait a second.
> I'm accepting a minor repurposing of the word as defined athttp://en.wiktionary.org/wiki/antidiagonal
> What do you mean by the antidiagonal of an infinite list of
> real numbers with infinite digits after the decimal point?
> Where do you find the upper right or bottom left corner of
> an infinitely large matrix?  I get the diagonal because it
> defines the first digit after the decimal point in the first
> member of the set, the second digit after the decimal point in
> the second member of the set, the nth digit after the decimal
> point in the nth member of the set.
>
> The part I cut dealt with finite sets and an arbitrary cutoff
> of digits after the decimal point.  The routine you provided
> doesn't apply to Cantor's diagonalization let alone refute it.
>
> Here's a version of Cantor's argument.
> Given any ordered set of unique reals on the interval (0,1)
> represented base ten where the order is identified via a map
> to the natural numbers in their normal order, then...
> given the identification of the elements using the index function
> R=f(x) where x is index into f and R is the value the real number
> at that index... then the real number developed by summing
> case (floor(f(x) * 10^(x+1)) mod 10 ) = 0 then 7 * 0.1^n
> else ((floor(f(x) * 10^(x+1)) mod 10) - 1) * 0.1^n
> for all x won't be in the set.  The reason is it differs from
> f(x) in at least the xth digit after the decimal.

If it differs from all reals on the list, then by what Epsilon>0
is the difference?


> If the number
> isn't in the ordered set it won't be in the unordered set containing
> the same reals.

The "MISSING REALS" here :

http://tinyurl.com/antidiagonals

are different to some R at the 1st digit pos
are different to some R at the 2nd digit pos
are different to some R at the 3rd digit pos

What do you notice about the "MISSING REALS" ?

Herc

William Hughes

unread,
Nov 9, 2012, 6:56:28 PM11/9/12
to
Why this is marked as abuse? It has been marked as abuse.
Report not abuse
On Nov 9, 5:13 pm, Graham Cooper <grahamcoop...@gmail.com> wrote:
> On Nov 9, 7:18 pm, forbisga...@gmail.com wrote:

> > Here's a version of Cantor's argument.
> > Given any ordered set of unique reals on the interval (0,1)
> > represented base ten where the order is identified via a map
> > to the natural numbers in their normal order, then...
> > given the identification of the elements using the index function
> > R=f(x) where x is index into f and R is the value the real number
> > at that index... then the real number developed by summing
> > case (floor(f(x) * 10^(x+1)) mod 10 ) = 0 then 7 * 0.1^n
> > else ((floor(f(x) * 10^(x+1)) mod 10) - 1) * 0.1^n
> > for all x won't be in the set.  The reason is it differs from
> > f(x) in at least the xth digit after the decimal.
>
> If it differs from all reals on the list, then by what Epsilon>0
> is the difference?


There is no Epsilon of course. Compare: sqrt(2) differs from all
rationals
but there is no Epsilon>0 by which it differs from all rationals

Graham Cooper

unread,
Nov 9, 2012, 11:39:28 PM11/9/12
to
A powerful analogy that turned the entire maths world into fools.

Herc

--
"Nothing supernatural there!" ~ Bob C
www.chatzombie.com/BUDDHA-CLOUD.gif

William Hughes

unread,
Nov 10, 2012, 12:33:35 AM11/10/12
to
Not an analogy, a counterexample

proposition, if x differs from every element of S. then there is
an
Epsilon>0 such that x differs by at least epsilon

counterexample: x=sqrt(2), S the irrationals

William Hughes

unread,
Nov 10, 2012, 12:57:24 AM11/10/12
to
make that S the rationals (increment my Oops counter)

Graham Cooper

unread,
Nov 10, 2012, 1:06:04 AM11/10/12
to
And <1 2 3...> is different to all of

<1 2 1 2 1 2...>
<3 3 3 3 3 3...>
<9 8 7 9 8 7...>

for a specific reason
=================


<9 9 8 6 2 5 8 3 4...>

is not different to all of

<4 2 0 5 3 4 3 4 2..>
<2 4 3 5 6 3 4 4 3..>
<0 3 3 3 0 4 4 5 5..>
...
AND SO ON..

just because you can formulate a pinpoint intersection of digit
inequality.

MODIFY-ALL-OF-SEQUENCE <4 4 3 ... > = <9 9 8 ...>

It's total nonsense if you work with infinite )sets_ of reals.

WTF is 4 4 3...? to a infinite SET NXN?

You can't PROVE ANYTHING with the nomenclature surrounded by
'uncountable' because it's an INTRINSIC DEFINITION.

That doesn't mean it HOLDS UP TO SCRUITY FOR 100 YEARS - it just HAS
NO SCRUTINY!


Herc

Uirgil

unread,
Nov 10, 2012, 1:28:37 AM11/10/12
to
In article
<f2ab8d18-7c72-4851...@m4g2000pbd.googlegroups.com>,
On the contrary, it works quite well for countably infinite sets of
reals in lists.

Graham Cooper

unread,
Nov 10, 2012, 1:34:49 AM11/10/12
to
On Nov 10, 4:28 pm, Uirgil <uir...@uirgil.ur> wrote:
> In article
> <f2ab8d18-7c72-4851-92f1-e49dda4e2...@m4g2000pbd.googlegroups.com>,
You equate LIST with COUNTABLE SET
hence you missed the point completely.

There is no diagonal of an infinite set.

Herc

Uirgil

unread,
Nov 10, 2012, 1:54:21 AM11/10/12
to
In article
<27de2871-d7a9-4cef...@u4g2000pbo.googlegroups.com>,
Lists ARE countable sets, though not all sets need to be lists.
>
> There is no diagonal of an infinite set.

There is an anti-diagonal for every finite or countably infinite list of
infinite binary sequences. There is no anti-diagonal for, say, the set
of all reals or the set of al binary sequnces.
>

William Hughes

unread,
Nov 10, 2012, 8:39:16 AM11/10/12
to
On Nov 10, 2:06 am, Graham Cooper <grahamcoop...@gmail.com> wrote:

<snip>

> MODIFY-ALL-OF-SEQUENCE <4 4 3 ... > = <9 9 8 ...>
>
> It's total nonsense if you work with infinite )sets_ of reals.

sqrt(2) an infinite sequence of digits you can define.
ANTIDIAG(L) an infinite sequence of digits you cannot define.

Graham Cooper

unread,
Nov 10, 2012, 7:40:13 PM11/10/12
to
why not?

.... TM# TAPE
UTM( index, digitpos ) e {0 1 2 3 4 5 6 7 8 9} TAPE AFTER HALT
<->
reals(index)_digitpos e {0 1 2 3 4 5 6 7 8 9}

reals(index)_digitpos = 0 OTHERWISE

AD(i) = flip( reals(i)_i )

Herc

Graham Cooper

unread,
Nov 10, 2012, 7:42:50 PM11/10/12
to
> > You equate LIST with COUNTABLE SET
> > hence you missed the point completely.
>
> Lists ARE countable sets, though not all sets need to be lists.
>


This PROVES you MISSED THE POINT!

Countable SETS are NOT LISTS.

This set has no diagonal.

0.102 0.434 0.543
0.245 0.545 0.544
0.095 0.111 0.999


Herc

Uirgil

unread,
Nov 10, 2012, 11:35:52 PM11/10/12
to
In article
<a1128ee0-b224-45a9...@v6g2000pbb.googlegroups.com>,
Graham Cooper <graham...@gmail.com> wrote:

> > > You equate LIST with COUNTABLE SET
> > > hence you missed the point completely.
> >
> > Lists ARE countable sets, though not all sets need to be lists.
> >
>
>
> This PROVES you MISSED THE POINT!
>
> Countable SETS are NOT LISTS.

But their members necessarily are capable of being listed by the very
proof that they are countable.
>
> This set has no diagonal.
>
> 0.102 0.434 0.543
> 0.245 0.545 0.544
> 0.095 0.111 0.999
>
>
> Herc

0.102000000
0.434000000
0.543000000
0.245000000
0.545000000
0.544000000
0.095000000
0.111000000
0.999000000

Now it does!

Graham Cooper

unread,
Nov 11, 2012, 4:59:42 PM11/11/12
to
On Nov 11, 2:35 pm, Uirgil <uir...@uirgil.ur> wrote:
> In article
> <a1128ee0-b224-45a9-bae9-56160c2ac...@v6g2000pbb.googlegroups.com>,
>  Graham Cooper <grahamcoop...@gmail.com> wrote:
>
> > > > You equate LIST with COUNTABLE SET
> > > > hence you missed the point completely.
>
> > > Lists ARE countable sets, though not all sets need to be lists.
>
> > This PROVES you MISSED THE POINT!
>
> > Countable SETS are NOT LISTS.
>
> But their members necessarily are capable of being listed by the very
> proof that they are countable.
>
>
>
> > This set has no diagonal.
>
> > 0.102  0.434  0.543
> > 0.245  0.545  0.544
> > 0.095  0.111  0.999
>
> > Herc
>
> 0.102000000
> 0.434000000
> 0.543000000
> 0.245000000
> 0.545000000
> 0.544000000
> 0.095000000
> 0.111000000
> 0.999000000
>
> Now it does!
>


like looking at the position of an electron with a microscope and
claiming the position is fixed.


Herc

Graham Cooper

unread,
Nov 12, 2012, 4:03:44 PM11/12/12
to
On Nov 12, 6:16 pm, David Bernier <david...@videotron.ca> wrote:
> On 11/12/2012 03:02 AM, Uirgil wrote:
>
> > In article
> > <acc4f4cb-7f4d-4a0d-bb27-c9682b41e...@r10g2000pbd.googlegroups.com>,
> >   Graham Cooper<grahamcoop...@gmail.com>  wrote:
>
> >> On Nov 12, 10:23 am, forbisga...@gmail.com wrote:
> >>> On Sunday, November 11, 2012 1:59:42 PM UTC-8, Graham Cooper wrote:
> >>>> like looking at the position of an electron with a microscope and
> >>>> claiming the position is fixed.
>
> >>> The problem is Cantor dealt with a well ordered set.  A well ordered
> >>> set has a fixed index.  The same number will alway appear at the same
> >>> index.  You appear to be claiming there is no fixed index for a well
> >>> ordered set.
>
> >> I claim 20 flaws in Cantor's proof!
>
> > A lot of nuts have claimed flaws in Cantor's proofs, but, as yet, none
> > of those nuts has proved not to be cracked.
>
> [...]
>
> I think one ontological (non-mathematical)
> problematic is that if something can't be
> listed, enumerated, then it can't be shown (like
> in "show the story", don't just "tell the story"
> in journalism).
>
> The story of the reals:  if it could be shown completely
> in a movie, that would be enumerable.
>
> The non-believers are not satisfied with a logical proof ...
>
> Dave
>


It's NOTHING CLOSE to a logical proof!

2OL

ALL(f):N->R E(r):R A(n):N

f(n) =/= r

clearly is 2OL!

---------------

ZFC AXIOM 9

ALL(X) E(R) (R well-orders X)

THIS IS NOT 1OL or even 2OL!

----------------

YOU ALL LIE YOU HAVE PROVEN X>OO IN FIRST ORDER LOGIC
YOU ALL LIE YOU HAVE A FORMAL PROOF OF X>OO
YOU DON'T MAKE ANY TESTABLE CLAIM
YOU DON'T UTILISE ANY CARDINALITY THEOREM
YOU FAIL TO ANSWER ANY QUESTIONS

*****************************************************
*****************************************************
Q4
How can there be uncountable many GODEL NUMBERS like this?

20130415
a01(0,1)
MIDPOINT(0,1)

A CHOICE FUNCTION
*****************************************************
*****************************************************
Q5
Which 1 of these does not hold?

a) N <-BIJECTS-> GODEL NUMBERS

b) GODEL NUMBERS <-BIJECT-> FUNCTIONS

c) FUNCTIONS <-BIJECT-> CHOICE FUNCTIONS

c) CHOICE FUNCTIONS <-BIJECT-> SETS

d) |SETS| > |N|
*****************************************************
*****************************************************
Q6
Does this Anti-Diagonal Method produce any unique
digit segment not listed?
AD METHOD
Choose the number 0.a_1a_2a_3...., where a_i = 1 if the i-th
number in your list had zero in its i-position, a_i = 0 otherwise.

LIST
R1= < <314><15><926><535><8979><323> ... >
R2= < <27><18281828><459045><235360> ... >
R3= < <333><333><333><333><333><333> ... >
R4= < <888888888888888888888><8><88> ... >
R5= < <0123456789><0123456789><01234 ... >
R6= < <1><414><21356><2373095><0488> ... >
....

G. Cooper (BInfTech)

William Hughes

unread,
Nov 12, 2012, 4:55:55 PM11/12/12
to
On Nov 12, 5:03 pm, Graham Cooper <grahamcoop...@gmail.com> wrote:
*****************************************************
> Q5
> Which 1 of these does not hold?
>
> a) N <-BIJECTS-> GODEL NUMBERS
>
> b) GODEL NUMBERS <-BIJECT-> FUNCTIONS
>
> c) FUNCTIONS <-BIJECT-> CHOICE FUNCTIONS
>
> c) CHOICE FUNCTIONS <-BIJECT-> SETS
>
> d) |SETS| > |N|
> *****************************************************
> *****************************************************

b) is false. What is true is that

GODEL NUMBERS <-BIJECT-> CONSTRUCTABLE FUNCTIONS

However

CONSTRUCTABLE FUNCTIONS <-NOT BIJECT-> CHOICE FUNCTIONS

Graham Cooper

unread,
Nov 13, 2012, 4:10:23 AM11/13/12
to
So what's an unlistable choice function example?

Are you using an uncountable infinite alphabet of Logic Symbols?

Or are you imagining infinite length of logic symbols choice functions
for every set?

Herc

William Hughes

unread,
Nov 13, 2012, 8:52:23 AM11/13/12
to
On Nov 13, 5:10 am, Graham Cooper <grahamcoop...@gmail.com> wrote:
> On Nov 13, 7:55 am, William Hughes <wpihug...@gmail.com> wrote:
>
>
>
>
>
>
>
>
>
> > On Nov 12, 5:03 pm, Graham Cooper <grahamcoop...@gmail.com> wrote:
> >  *****************************************************
>
> > > Q5
> > > Which 1 of these does not hold?
>
> > > a) N <-BIJECTS-> GODEL NUMBERS
>
> > > b) GODEL NUMBERS <-BIJECT-> FUNCTIONS
>
> > > c) FUNCTIONS <-BIJECT-> CHOICE FUNCTIONS
>
> > > c) CHOICE FUNCTIONS <-BIJECT-> SETS
>
> > > d) |SETS| > |N|
> > > *****************************************************
> > > *****************************************************
>
> > b) is false. What is true is that
>
> >  GODEL NUMBERS <-BIJECT-> CONSTRUCTABLE FUNCTIONS
>
> > However
>
> >       CONSTRUCTABLE FUNCTIONS <-NOT BIJECT-> CHOICE FUNCTIONS
>
> So what's an unlistable choice function example?

There is of course, no such thing (given any x, you can always
start a list with x).

However, given any x there is a list L(x) such that x is
in L(x), does not mean there is a single list L that
contains all x.

There is no list of all choice functions.


0 new messages