Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]


Groups > sci.logic > #166488 > unrolled thread

Cantor's argument and non-standard reals

Started byZuhair <zaljohar@gmail.com>
First post2017-11-08 05:30 -0800
Last post2017-11-09 22:23 -0800
Articles 20 on this page of 28 — 8 participants

Back to article view | Back to sci.logic


Contents

  Cantor's argument and non-standard reals Zuhair <zaljohar@gmail.com> - 2017-11-08 05:30 -0800
    Re: Cantor's argument and non-standard reals Zuhair <zaljohar@gmail.com> - 2017-11-08 10:07 -0800
      Re: Cantor's argument and non-standard reals FredJeffries <fredjeffries@gmail.com> - 2017-11-08 14:26 -0800
      Re: Cantor's argument and non-standard reals "Shobe, Martin" <martin.shobe@yahoo.com> - 2017-11-09 06:41 -0600
        Re: Cantor's argument and non-standard reals Zuhair <zaljohar@gmail.com> - 2017-11-09 06:03 -0800
    Re: Cantor's argument and non-standard reals WM <wolfgang.mueckenheim@hs-augsburg.de> - 2017-11-08 11:04 -0800
      Re: Cantor's argument and non-standard reals George Greene <greeneg@email.unc.edu> - 2017-11-08 11:58 -0800
        Re: Cantor's argument and non-standard reals WM <wolfgang.mueckenheim@hs-augsburg.de> - 2017-11-08 13:13 -0800
          Re: Cantor's argument and non-standard reals Zeit Geist <Tucsondrew@me.com> - 2017-11-08 13:19 -0800
            Re: Cantor's argument and non-standard reals WM <wolfgang.mueckenheim@hs-augsburg.de> - 2017-11-09 01:20 -0800
          Re: Cantor's argument and non-standard reals Ben Bacarisse <ben.usenet@bsb.me.uk> - 2017-11-09 16:47 +0000
            Re: Cantor's argument and non-standard reals WM <wolfgang.mueckenheim@hs-augsburg.de> - 2017-11-10 07:02 -0800
              Re: Cantor's argument and non-standard reals Ben Bacarisse <ben.usenet@bsb.me.uk> - 2017-11-10 20:32 +0000
                Re: Cantor's argument and non-standard reals WM <wolfgang.mueckenheim@hs-augsburg.de> - 2017-11-11 01:57 -0800
                  Re: Cantor's argument and non-standard reals Ben Bacarisse <ben.usenet@bsb.me.uk> - 2017-11-11 15:44 +0000
                    Re: Cantor's argument and non-standard reals WM <wolfgang.mueckenheim@hs-augsburg.de> - 2017-11-11 13:20 -0800
                      Re: Cantor's argument and non-standard reals Ben Bacarisse <ben.usenet@bsb.me.uk> - 2017-11-11 21:43 +0000
                        Re: Cantor's argument and non-standard reals "Ross A. Finlayson" <ross.finlayson@gmail.com> - 2017-11-12 13:56 -0800
                          Re: Cantor's argument and non-standard reals Ben Bacarisse <ben.usenet@bsb.me.uk> - 2017-11-13 00:49 +0000
                            Re: Cantor's argument and non-standard reals "Ross A. Finlayson" <ross.finlayson@gmail.com> - 2017-11-12 17:18 -0800
      Re: Cantor's argument and non-standard reals Zeit Geist <Tucsondrew@me.com> - 2017-11-08 13:10 -0800
        Re: Cantor's argument and non-standard reals WM <wolfgang.mueckenheim@hs-augsburg.de> - 2017-11-08 13:17 -0800
          Re: Cantor's argument and non-standard reals Zeit Geist <Tucsondrew@me.com> - 2017-11-08 13:27 -0800
            Re: Cantor's argument and non-standard reals WM <wolfgang.mueckenheim@hs-augsburg.de> - 2017-11-09 01:27 -0800
    Re: Cantor's argument and non-standard reals FredJeffries <fredjeffries@gmail.com> - 2017-11-08 14:21 -0800
    Re: Cantor's argument and non-standard reals "Ross A. Finlayson" <ross.finlayson@gmail.com> - 2017-11-08 19:24 -0800
      Re: Cantor's argument and non-standard reals Zeit Geist <Tucsondrew@me.com> - 2017-11-08 19:46 -0800
        Re: Cantor's argument and non-standard reals "Ross A. Finlayson" <ross.finlayson@gmail.com> - 2017-11-09 22:23 -0800

Page 1 of 2  [1] 2  Next page →


#166488 — Cantor's argument and non-standard reals

FromZuhair <zaljohar@gmail.com>
Date2017-11-08 05:30 -0800
SubjectCantor's argument and non-standard reals
Message-ID<a10cb388-2b53-44a6-9888-a6cbddb4c30e@googlegroups.com>
The usual reals can have duplicate infinite decimal representation, i.e.
two distinct infinite decimal sequences representing that real. However,
there is a proof that for each real that has duplicate representation
one of these sequences must end with a trailing 9s and the other with a trailing 0s. Now are there NON-STANDARD reals in which this fails? whereby reals might have many (more than two, or even infinitely many) infinite decimal sequences representing them. IF so then how can Cantor's diagonal argument relate to such kind of reals, should those exist.

[toc] | [next] | [standalone]


#166495

FromZuhair <zaljohar@gmail.com>
Date2017-11-08 10:07 -0800
Message-ID<e0199b5d-63ba-414a-88e6-d532a73cc6a4@googlegroups.com>
In reply to#166488
On Wednesday, November 8, 2017 at 4:30:05 PM UTC+3, Zuhair wrote:
> The usual reals can have duplicate infinite decimal representation, i.e.
> two distinct infinite decimal sequences representing that real. However,
> there is a proof that for each real that has duplicate representation
> one of these sequences must end with a trailing 9s and the other with a trailing 0s. Now are there NON-STANDARD reals in which this fails? whereby reals might have many (more than two, or even infinitely many) infinite decimal sequences representing them. IF so then how can Cantor's diagonal argument relate to such kind of reals, should those exist.

although I'm not really familiar with such non-standard grounds. I have
a simple attempt as an example of what can possibly constitute such reals.

Now a real in the interval [0,1] can be defined as the limit to the sum of items (x_i/n^i) of some sequence, for all i=1,2,3,...., and 0=<x_i=<n-1.

Now the STANDARD reals use standard naturals i, and n.

Now can we get a non-standard real by taking n to be some non-standard
natural, while keeping i to range over standard naturals.

Would that non-standard real be different from the standard reals, I mean as far as having maximally two duplicate representations per base n infinite sequences? More explicitly can we have infinitely many representations for each real for some non-standard base n?


 

[toc] | [prev] | [next] | [standalone]


#166511

FromFredJeffries <fredjeffries@gmail.com>
Date2017-11-08 14:26 -0800
Message-ID<f9c3a0f5-9974-4e86-bcff-45ec99e02a4e@googlegroups.com>
In reply to#166495
On Wednesday, November 8, 2017 at 10:07:20 AM UTC-8, Zuhair wrote:
> On Wednesday, November 8, 2017 at 4:30:05 PM UTC+3, Zuhair wrote:
> > The usual reals can have duplicate infinite decimal representation, i.e.
> > two distinct infinite decimal sequences representing that real. However,
> > there is a proof that for each real that has duplicate representation
> > one of these sequences must end with a trailing 9s and the other with a trailing 0s. Now are there NON-STANDARD reals in which this fails? whereby reals might have many (more than two, or even infinitely many) infinite decimal sequences representing them. IF so then how can Cantor's diagonal argument relate to such kind of reals, should those exist.
> 
> although I'm not really familiar with such non-standard grounds. I have
> a simple attempt as an example of what can possibly constitute such reals.
> 
> Now a real in the interval [0,1] can be defined as the limit to the sum of items (x_i/n^i) of some sequence, for all i=1,2,3,...., and 0=<x_i=<n-1.
> 
> Now the STANDARD reals use standard naturals i, and n.
> 
> Now can we get a non-standard real by taking n to be some non-standard
> natural, while keeping i to range over standard naturals.

No. "The standard naturals" is an external set (notice the word "standard"). You can't use it as your index set.

[toc] | [prev] | [next] | [standalone]


#166537

From"Shobe, Martin" <martin.shobe@yahoo.com>
Date2017-11-09 06:41 -0600
Message-ID<ou1idv$e8k$1@dont-email.me>
In reply to#166495
On 11/8/2017 12:07 PM, Zuhair wrote:
> On Wednesday, November 8, 2017 at 4:30:05 PM UTC+3, Zuhair wrote:
>> The usual reals can have duplicate infinite decimal representation, i.e.
>> two distinct infinite decimal sequences representing that real. However,
>> there is a proof that for each real that has duplicate representation
>> one of these sequences must end with a trailing 9s and the other with a trailing 0s. Now are there NON-STANDARD reals in which this fails? whereby reals might have many (more than two, or even infinitely many) infinite decimal sequences representing them. IF so then how can Cantor's diagonal argument relate to such kind of reals, should those exist.
> 
> although I'm not really familiar with such non-standard grounds. I have
> a simple attempt as an example of what can possibly constitute such reals.
> 
> Now a real in the interval [0,1] can be defined as the limit to the sum of items (x_i/n^i) of some sequence, for all i=1,2,3,...., and 0=<x_i=<n-1.
> 
> Now the STANDARD reals use standard naturals i, and n.
> 
> Now can we get a non-standard real by taking n to be some non-standard
> natural, while keeping i to range over standard naturals.
> 
> Would that non-standard real be different from the standard reals, I mean as far as having maximally two duplicate representations per base n infinite sequences? More explicitly can we have infinitely many representations for each real for some non-standard base n?
> 

I don't see how the proof that there's only 1 or 2 representations for 
each real is changed by this. Can you point out where you think it might 
fail?

Martin Shobe

[toc] | [prev] | [next] | [standalone]


#166539

FromZuhair <zaljohar@gmail.com>
Date2017-11-09 06:03 -0800
Message-ID<65e76005-42c1-430a-9bbd-08695e1c7762@googlegroups.com>
In reply to#166537
On Thursday, November 9, 2017 at 3:41:37 PM UTC+3, Shobe, Martin wrote:
> On 11/8/2017 12:07 PM, Zuhair wrote:
> > On Wednesday, November 8, 2017 at 4:30:05 PM UTC+3, Zuhair wrote:
> >> The usual reals can have duplicate infinite decimal representation, i.e.
> >> two distinct infinite decimal sequences representing that real. However,
> >> there is a proof that for each real that has duplicate representation
> >> one of these sequences must end with a trailing 9s and the other with a trailing 0s. Now are there NON-STANDARD reals in which this fails? whereby reals might have many (more than two, or even infinitely many) infinite decimal sequences representing them. IF so then how can Cantor's diagonal argument relate to such kind of reals, should those exist.
> > 
> > although I'm not really familiar with such non-standard grounds. I have
> > a simple attempt as an example of what can possibly constitute such reals.
> > 
> > Now a real in the interval [0,1] can be defined as the limit to the sum of items (x_i/n^i) of some sequence, for all i=1,2,3,...., and 0=<x_i=<n-1.
> > 
> > Now the STANDARD reals use standard naturals i, and n.
> > 
> > Now can we get a non-standard real by taking n to be some non-standard
> > natural, while keeping i to range over standard naturals.
> > 
> > Would that non-standard real be different from the standard reals, I mean as far as having maximally two duplicate representations per base n infinite sequences? More explicitly can we have infinitely many representations for each real for some non-standard base n?
> > 
> 
> I don't see how the proof that there's only 1 or 2 representations for 
> each real is changed by this. Can you point out where you think it might 
> fail?
> 
> Martin Shobe

I just thought that it can make a difference because the structure of the non-standards is externally different from the standard naturals but internal to the theory they look the same. Apparently, it doesn't cause much of a difference.

[toc] | [prev] | [next] | [standalone]


#166497

FromWM <wolfgang.mueckenheim@hs-augsburg.de>
Date2017-11-08 11:04 -0800
Message-ID<ec9ed1d0-0f06-496f-a813-2c1b679f73e0@googlegroups.com>
In reply to#166488
Am Mittwoch, 8. November 2017 14:30:05 UTC+1 schrieb Zuhair:
> The usual reals can have duplicate infinite decimal representation, i.e.
> two distinct infinite decimal sequences representing that real.

Cantor has not bothered about limits. Cantor has simply used the binary digit sequence (expressed by the letters w and m).

The real numbers can in fact be expressed as limits of the digit sequences. But they *are* not these digit sequences. The sequence

0.9, 0.99, 0.999, ... has aleph_0 terms that differ from its limit 1.

0.9, 0.99, 0.999, ... = (...(((0.9)9)9)...) = 0.999...

That means, if we refrain from the confusion that 0.999... should express the limit, then it is clear that Cantor's list does not contain real numbers and does not supply a real number as its diagonal, let alone a transcendental number.

Regards, WM

[toc] | [prev] | [next] | [standalone]


#166499

FromGeorge Greene <greeneg@email.unc.edu>
Date2017-11-08 11:58 -0800
Message-ID<e6d9400d-61f2-4bff-acbd-e44f9feded62@googlegroups.com>
In reply to#166497
On Wednesday, November 8, 2017 at 2:04:18 PM UTC-5, WM wrote:
> The real numbers can in fact be expressed as limits of the digit sequences.

You don't have to bother with limits if the sequences are actually infinite.
You instead have SUMS.

[toc] | [prev] | [next] | [standalone]


#166503

FromWM <wolfgang.mueckenheim@hs-augsburg.de>
Date2017-11-08 13:13 -0800
Message-ID<4c874aeb-629c-48ba-a84a-860888a30f15@googlegroups.com>
In reply to#166499
Am Mittwoch, 8. November 2017 20:58:15 UTC+1 schrieb George Greene:
> On Wednesday, November 8, 2017 at 2:04:18 PM UTC-5, WM wrote:
> > The real numbers can in fact be expressed as limits of the digit sequences.
> 
> You don't have to bother with limits if the sequences are actually infinite.
> You instead have SUMS.

Wrong. The sum 0.9 + 0.09 + 0.009 + ... < 1
This holds for all indexed digits (at finite positions). The infinite set of all natural indices does not contain other than finite indices. Therefore the infinite sum is less than 1.

Regards, WM

[toc] | [prev] | [next] | [standalone]


#166505

FromZeit Geist <Tucsondrew@me.com>
Date2017-11-08 13:19 -0800
Message-ID<389aa74c-3cd7-40ce-80af-68d1ee7bd699@googlegroups.com>
In reply to#166503
On Wednesday, November 8, 2017 at 2:13:31 PM UTC-7, WM wrote:
> Am Mittwoch, 8. November 2017 20:58:15 UTC+1 schrieb George Greene:
> > On Wednesday, November 8, 2017 at 2:04:18 PM UTC-5, WM wrote:
> > > The real numbers can in fact be expressed as limits of the digit sequences.
> > 
> > You don't have to bother with limits if the sequences are actually infinite.
> > You instead have SUMS.
> 
> Wrong. The sum 0.9 + 0.09 + 0.009 + ... < 1
> This holds for all indexed digits (at finite positions). The infinite set of all natural indices does not contain other than finite indices. Therefore the infinite sum is less than 1.

It holds for UP TO every partial sum.  However, the entire sum equals 1.

You have issues with universal qualification.  No shock!

> Regards, WM

ZG

[toc] | [prev] | [next] | [standalone]


#166530

FromWM <wolfgang.mueckenheim@hs-augsburg.de>
Date2017-11-09 01:20 -0800
Message-ID<3f2a0ff2-a3e9-417a-a469-5b6f47aeac84@googlegroups.com>
In reply to#166505
Am Mittwoch, 8. November 2017 22:19:12 UTC+1 schrieb Zeit Geist:
> On Wednesday, November 8, 2017 at 2:13:31 PM UTC-7, WM wrote:
> > Am Mittwoch, 8. November 2017 20:58:15 UTC+1 schrieb George Greene:
> > > On Wednesday, November 8, 2017 at 2:04:18 PM UTC-5, WM wrote:
> > > > The real numbers can in fact be expressed as limits of the digit sequences.
> > > 
> > > You don't have to bother with limits if the sequences are actually infinite.
> > > You instead have SUMS.
> > 
> > Wrong. The sum 0.9 + 0.09 + 0.009 + ... < 1
> > This holds for all indexed digits (at finite positions). The infinite set of all natural indices does not contain other than finite indices. Therefore the infinite sum is less than 1.
> 
> It holds for UP TO every partial sum.  However, the entire sum equals 1.

The "entire sum" does never surpass finite summands 9/10^n.
> 
> You have issues with universal qualification. 

I use it: For every n in |N: a_n = 9/10 + ... + 9/10^n < 1.

This excludes that a sum of different terms 9/10^n could be 1.

Regards, WM


[toc] | [prev] | [next] | [standalone]


#166553

FromBen Bacarisse <ben.usenet@bsb.me.uk>
Date2017-11-09 16:47 +0000
Message-ID<87efp7761p.fsf@bsb.me.uk>
In reply to#166503
WM <wolfgang.mueckenheim@hs-augsburg.de> writes:

> Am Mittwoch, 8. November 2017 20:58:15 UTC+1 schrieb George Greene:
>> On Wednesday, November 8, 2017 at 2:04:18 PM UTC-5, WM wrote:
>> > The real numbers can in fact be expressed as limits of the digit sequences.
>> 
>> You don't have to bother with limits if the sequences are actually infinite.
>> You instead have SUMS.
>
> Wrong. The sum 0.9 + 0.09 + 0.009 + ... < 1
> This holds for all indexed digits (at finite positions). The infinite
> set of all natural indices does not contain other than finite
> indices. Therefore the infinite sum is less than 1.

No, that's wrong.  See, for example, page 203 of "Mathematik für die
ersten Semester" by Prof. Dr. rer. nat. habil. Wolfgang Mückenheim.

-- 
Ben.

[toc] | [prev] | [next] | [standalone]


#166652

FromWM <wolfgang.mueckenheim@hs-augsburg.de>
Date2017-11-10 07:02 -0800
Message-ID<0ed22766-42f3-4334-9959-ceeb80a3356c@googlegroups.com>
In reply to#166553
Am Donnerstag, 9. November 2017 17:47:32 UTC+1 schrieb Ben Bacarisse:
> WM <wolfgang.mueckenheim@hs-augsburg.de> writes:
> 
> > Am Mittwoch, 8. November 2017 20:58:15 UTC+1 schrieb George Greene:
> >> On Wednesday, November 8, 2017 at 2:04:18 PM UTC-5, WM wrote:
> >> > The real numbers can in fact be expressed as limits of the digit sequences.
> >> 
> >> You don't have to bother with limits if the sequences are actually infinite.
> >> You instead have SUMS.
> >
> > Wrong. The sum 0.9 + 0.09 + 0.009 + ... < 1
> > This holds for all indexed digits (at finite positions). The infinite
> > set of all natural indices does not contain other than finite
> > indices. Therefore the infinite sum is less than 1.
> 
> No, that's wrong.  See, for example, page 203 of "Mathematik für die
> ersten Semester" by Prof. Dr. rer. nat. habil. Wolfgang Mückenheim.
> 
I cannot find an error there. Do you confuse the sum from 1 to oo with the sum over all n in |N? 

Regards, WM

[toc] | [prev] | [next] | [standalone]


#166664

FromBen Bacarisse <ben.usenet@bsb.me.uk>
Date2017-11-10 20:32 +0000
Message-ID<87d14pyivn.fsf@bsb.me.uk>
In reply to#166652
WM <wolfgang.mueckenheim@hs-augsburg.de> writes:

> Am Donnerstag, 9. November 2017 17:47:32 UTC+1 schrieb Ben Bacarisse:
>> WM <wolfgang.mueckenheim@hs-augsburg.de> writes:
>> 
>> > Am Mittwoch, 8. November 2017 20:58:15 UTC+1 schrieb George Greene:
>> >> On Wednesday, November 8, 2017 at 2:04:18 PM UTC-5, WM wrote:
>> >> > The real numbers can in fact be expressed as limits of the digit sequences.
>> >> 
>> >> You don't have to bother with limits if the sequences are actually infinite.
>> >> You instead have SUMS.
>> >
>> > Wrong. The sum 0.9 + 0.09 + 0.009 + ... < 1
>> > This holds for all indexed digits (at finite positions). The infinite
>> > set of all natural indices does not contain other than finite
>> > indices. Therefore the infinite sum is less than 1.
>> 
>> No, that's wrong.  See, for example, page 203 of "Mathematik für die
>> ersten Semester" by Prof. Dr. rer. nat. habil. Wolfgang Mückenheim.
>> 
> I cannot find an error there.

No, page 203 is fine.  It all looks like conventional mathematics.

> Do you confuse the sum from 1 to oo with
> the sum over all n in |N?

The term "infinite sum" (to an English speaker) refers to the sums you
show how to calculate in your book, so I think you are just playing
words games again.  You've made up some sort of sum (without defining
it) that "is less than 1" and called it by a confusing name.  That seems
to be all you ever do.

-- 
Ben.

[toc] | [prev] | [next] | [standalone]


#166687

FromWM <wolfgang.mueckenheim@hs-augsburg.de>
Date2017-11-11 01:57 -0800
Message-ID<bd139f07-dd50-4d77-aad8-b24c3f536038@googlegroups.com>
In reply to#166664
Am Freitag, 10. November 2017 21:32:48 UTC+1 schrieb Ben Bacarisse:
> WM <wolfgang.muec
> 


> >> > Wrong. The sum 0.9 + 0.09 + 0.009 + ... < 1
> >> > This holds for all indexed digits (at finite positions). The infinite
> >> > set of all natural indices does not contain other than finite
> >> > indices. Therefore the infinite sum is less than 1.
> >> 
> >> No, that's wrong.  See, for example, page 203 of "Mathematik für die
> >> ersten Semester" by Prof. Dr. rer. nat. habil. Wolfgang Mückenheim.
> >> 
> > I cannot find an error there.
> 
> No, page 203 is fine.  It all looks like conventional mathematics.
> 
> > Do you confuse the sum from 1 to oo with
> > the sum over all n in |N?
> 
> The term "infinite sum" (to an English speaker) refers to the sums you
> show how to calculate in your book, so I think you are just playing
> words games again.

That shows your lack of understanding.

The sum from 1 to oo is a limit because oo is not a natural number. But if you take the sum over all natural numbers, then you cannot have the limit because logic shows: for all natural numbers, the sum is less than 1.

< You've made up some sort of sum (without defining
> it) that "is less than 1" and called it by a confusing name.  That seems
> to be all you ever do.

I have explained the difference clearly as soon as I understood it clearly, namely on p. 194f of the fourth edition "Mathematik für die ersten Semester", 4th ed., De Gruyter, Berlin 2015. Concerning the sum over |N:

Ausdrücke wie a1 + a2 + a3 + ... oder 0,111 ... werden in der Literatur gewöhnlich nicht als Reihen, sondern stillschweigend als deren Grenzwerte aufgefasst, so dass eine korrekte Bezeichnung wie 0,111... → 1/9 kurz als 0,111 ...= 1/9 notiert wird. Diese vereinfachende Konvention ist darauf zurückzuführen, dass ein Ausdruck wie 0,111 ...sowohl zur Berechnung des Grenzwertes und jeder Partialsumme dienen kann, als auch zur Kennzeichnung der unendlichen Reihe, also der Folge der Partialsummen selbst. Sie führt weder beim Rechnen, noch in der Mathematik des potentiell Unendlichen zu Irrtümern, weil von den Gliedern einer unendlichen Reihe ohnehin niemals Vollständigkeit erwartet werden kann. Wird dagegen diese Vollständigkeit axiomatisch oder mit anderer Begründung gefordert oder vorausgesetzt, so muss zwischen der Reihe, d. h. der Partialsummenfolge oder Ziffernfolge, und ihrem Grenzwert
unterschieden werden. Denn es ist mit mathematischer Strenge unvereinbar, die unendliche Folge aller Endziffern von Näherungen, die für jeden noch so großen Index den Grenzwert verfehlen, mit eben diesem zu identifizieren.

Regards, WM

[toc] | [prev] | [next] | [standalone]


#166700

FromBen Bacarisse <ben.usenet@bsb.me.uk>
Date2017-11-11 15:44 +0000
Message-ID<87shdkx1jc.fsf@bsb.me.uk>
In reply to#166687
WM <wolfgang.mueckenheim@hs-augsburg.de> writes:

> Am Freitag, 10. November 2017 21:32:48 UTC+1 schrieb Ben Bacarisse:
>> WM <wolfgang.muec
>> 
>
>
>> >> > Wrong. The sum 0.9 + 0.09 + 0.009 + ... < 1
>> >> > This holds for all indexed digits (at finite positions). The infinite
>> >> > set of all natural indices does not contain other than finite
>> >> > indices. Therefore the infinite sum is less than 1.
>> >> 
>> >> No, that's wrong.  See, for example, page 203 of "Mathematik für die
>> >> ersten Semester" by Prof. Dr. rer. nat. habil. Wolfgang Mückenheim.
>> >> 
>> > I cannot find an error there.
>> 
>> No, page 203 is fine.  It all looks like conventional mathematics.
>> 
>> > Do you confuse the sum from 1 to oo with
>> > the sum over all n in |N?
>> 
>> The term "infinite sum" (to an English speaker) refers to the sums you
>> show how to calculate in your book, so I think you are just playing
>> words games again.
>
> That shows your lack of understanding.

The infinite sum is 1 (in English).  If you don't mean what other people
mean you should pick another phrase.  But that won't do for you because
all you want is to argue over undefined terms, and to hook people in you
need to misuse at least one word or phrase.  If you had said "the
Mueckenheim sum (which I decline to define) is less than 1" I doubt
anyone would care.

<snip>
-- 
Ben.

[toc] | [prev] | [next] | [standalone]


#166735

FromWM <wolfgang.mueckenheim@hs-augsburg.de>
Date2017-11-11 13:20 -0800
Message-ID<202661f6-8174-47d2-9a17-f0bcc7756a2b@googlegroups.com>
In reply to#166700
Am Samstag, 11. November 2017 16:44:57 UTC+1 schrieb Ben Bacarisse:
> WM <wolfgang.mueckenheim@hs-augsburg.de> writes:
> 
> > Am Freitag, 10. November 2017 21:32:48 UTC+1 schrieb Ben Bacarisse:
> >> WM <wolfgang.muec
> >> 
> >
> >
> >> >> > Wrong. The sum 0.9 + 0.09 + 0.009 + ... < 1
> >> >> > This holds for all indexed digits (at finite positions). The infinite
> >> >> > set of all natural indices does not contain other than finite
> >> >> > indices. Therefore the infinite sum is less than 1.
> >> >> 
> >> >> No, that's wrong.  See, for example, page 203 of "Mathematik für die
> >> >> ersten Semester" by Prof. Dr. rer. nat. habil. Wolfgang Mückenheim.
> >> >> 
> >> > I cannot find an error there.
> >> 
> >> No, page 203 is fine.  It all looks like conventional mathematics.
> >> 
> >> > Do you confuse the sum from 1 to oo with
> >> > the sum over all n in |N?
> >> 
> >> The term "infinite sum" (to an English speaker) refers to the sums you
> >> show how to calculate in your book, so I think you are just playing
> >> words games again.
> >
> > That shows your lack of understanding.
> 
> The infinite sum is 1 (in English).

Yes, although the expression infinite sum is somewhat misleading. More precise would it be to say the limit of the partial sum.

>  If you don't mean what other people
> mean you should pick another phrase.

I mean what other people mean. That does not imply that I accept all their mistakes too.

>  But that won't do for you because
> all you want is to argue over undefined terms

In set theory the set of all finite indices is well defined. It is the set that does not contain an infinite index. It is a limit but it is not that limit oo that is never reached by natural numbers.

, If you had said "the
> Mueckenheim sum (which I decline to define) is less than 1" I doubt
> anyone would care.

In fact it is a disaster that anybody cares about Cantor's nonsense. But those who do should do it more consequently: There exist all terms of the sequence of partial sums, but none is 1. To claim that this changes when ll are written together and including all finite terms (but nothing else) would be 1 is the summit of nonsensical and unmathematical thinking. But if the average mathematicians were brighter this question would never haven been raised.

Regards, WM

[toc] | [prev] | [next] | [standalone]


#166737

FromBen Bacarisse <ben.usenet@bsb.me.uk>
Date2017-11-11 21:43 +0000
Message-ID<87h8u0wkxx.fsf@bsb.me.uk>
In reply to#166735
WM <wolfgang.mueckenheim@hs-augsburg.de> writes:

> Am Samstag, 11. November 2017 16:44:57 UTC+1 schrieb Ben Bacarisse:
>> WM <wolfgang.mueckenheim@hs-augsburg.de> writes:
>> 
>> > Am Freitag, 10. November 2017 21:32:48 UTC+1 schrieb Ben Bacarisse:
>> >> WM <wolfgang.muec
>> >> >> > ... Therefore the infinite sum is less than 1.
>> >> >> 
>> >> >> No, that's wrong.
<snip>
>> The infinite sum is 1 (in English).
>
> Yes, although the expression infinite sum is somewhat misleading.

That's all I was saying.  I am glad we agree for once.

<snip>
-- 
Ben.

[toc] | [prev] | [next] | [standalone]


#166756

From"Ross A. Finlayson" <ross.finlayson@gmail.com>
Date2017-11-12 13:56 -0800
Message-ID<8867dbd6-c061-4716-a731-82d4c5d89c27@googlegroups.com>
In reply to#166737
On Saturday, November 11, 2017 at 1:43:24 PM UTC-8, Ben Bacarisse wrote:
> WM <wolfgang.mueckenheim@hs-augsburg.de> writes:
> 
> > Am Samstag, 11. November 2017 16:44:57 UTC+1 schrieb Ben Bacarisse:
> >> WM <wolfgang.mueckenheim@hs-augsburg.de> writes:
> >> 
> >> > Am Freitag, 10. November 2017 21:32:48 UTC+1 schrieb Ben Bacarisse:
> >> >> WM <wolfgang.muec
> >> >> >> > ... Therefore the infinite sum is less than 1.
> >> >> >> 
> >> >> >> No, that's wrong.
> <snip>
> >> The infinite sum is 1 (in English).
> >
> > Yes, although the expression infinite sum is somewhat misleading.
> 
> That's all I was saying.  I am glad we agree for once.
> 
> <snip>
> -- 
> Ben.


WM doesn't have some coherent rejection of the 
infinite or unbounded, and neither do I, then 
WM doesn't have some coherent rejection of the 
uncountability of the reals - I do.


This is where not everyone shares my view, 
on mathematical matters, and they can make 
what absurd, trivial, crippled mathematics 
they want and call it their world, but, 
that's a blind man's in a continuous, infinite 
universe, a blind man's affrontery that others 
are, too.  

And it's not "the" foundation, which includes 
a continuous, infinite universe (and even 
physically).

Anthropocentrism isn't unreasonable, but 
we're in the universe, not above it.

This simple post-axiomatic approach is 
extra-Cantorian and directly suited for 
founding real analysis with simpler and 
less principles, which conscientious 
mathematicians know embodies more of 
what "the" foundation is, the plainly 
logical (and scientific) foundation of 
mathematics (singular).

[toc] | [prev] | [next] | [standalone]


#166761

FromBen Bacarisse <ben.usenet@bsb.me.uk>
Date2017-11-13 00:49 +0000
Message-ID<87mv3ruhob.fsf@bsb.me.uk>
In reply to#166756
"Ross A. Finlayson" <ross.finlayson@gmail.com> writes:
<snip>
> WM doesn't have some coherent rejection of the 
> infinite or unbounded, and neither do I, then 
> WM doesn't have some coherent rejection of the 
> uncountability of the reals - I do.
<snip more...>

Sorry, but I don't know what your post is saying.  That's fine -- I
don't need an explanation -- I'm just letting you know why I won't be
replying to anything in the content.

-- 
Ben.

[toc] | [prev] | [next] | [standalone]


#166764

From"Ross A. Finlayson" <ross.finlayson@gmail.com>
Date2017-11-12 17:18 -0800
Message-ID<e6a694dc-404f-4630-b481-405002595c03@googlegroups.com>
In reply to#166761
On Sunday, November 12, 2017 at 4:49:11 PM UTC-8, Ben Bacarisse wrote:
> "Ross A. Finlayson" <ross.finlayson@gmail.com> writes:
> <snip>
> > WM doesn't have some coherent rejection of the 
> > infinite or unbounded, and neither do I, then 
> > WM doesn't have some coherent rejection of the 
> > uncountability of the reals - I do.
> <snip more...>
> 
> Sorry, but I don't know what your post is saying.  That's fine -- I
> don't need an explanation -- I'm just letting you know why I won't be
> replying to anything in the content.
> 
> -- 
> Ben.

Thanks for your courtesy.

I have quite an "agen-da".

Physics and "infinities" has that physics 
finds infinities in quantities in the equations 
that otherwise model the system then that 
careful qualifications and truncations 
leave them out (that they remain, then).  
As physics improves, various sub-fields 
grew apart from each other, about here then 
that space-time really is pretty much flat 
(and actually "infinitesimal" in curvature) 
that occupies physics these days with that 
now SR is local while GR is global and 
inflationary theories have a big re-think on.

Mathematics and infinities (and here beyond 
the unbounded) have clear, direct, usual 
recognitions that the same paradoxes that 
for example Russell doomed Frege with still 
exist with the same objects, that then instead 
(in some extra-ordinary setting, again per 
Russell) the properties of the infinite and 
the continuum have a bit of a re-think on in 
terms of the modern foundations as well.

So, and for example I can agree with someone 
who says "there's no infinity in my mathematics", 
but not that that's all of mathematics which 
does include mathematics of infinity.

Or, retro-finitists crankety trolls can get bent.

[toc] | [prev] | [next] | [standalone]


Page 1 of 2  [1] 2  Next page →

Back to top | Article view | sci.logic


csiph-web