# 1 is 0.9999999999999............

Discussion in 'Alternative Theories' started by chinglu, Oct 27, 2013.

1. ### rpennerFully WiredValued Senior Member

Messages:
4,833
Captain Kremmen,

$\pi = \lim_{n \to \infty} \sum_{k=0}^n \frac{(-1)^{k} 4}{2 k + 1} = \lim_{n \to \infty} \sum_{k=0}^n \frac{(-1)^{k} ( 16 \times 5^{-2k-1} - 4 \times 239^{-2k-1} ) }{2 k + 1} = \lim_{n \to \infty} \sum_{k=0}^n \frac{(-1)^{k} 2 \times 3^{\frac{1}{2}-k}}{2 k + 1} = \lim_{n \to \infty} \frac{(k!)^2 2^{k+1}}{(2k+1)!} = ...$

Until you understand limits, you do not understand continuity. Until you understand continuity, you do not understand what separates the concept of real numbers from the rational numbers.

3. ### TruthSeekerFancy Virtual Reality MonkeyValued Senior Member

Messages:
15,162
Yes. Without those limits within continuity, there is no way to define mathematics and no way of using it.

5. ### TrippyALEA IACTA ESTStaff Member

Messages:
10,890
No, what I am saying is that your assertion that the statments are only true because they are defined that way is false.

the only definition that is relevant is the use of ... or () to represent a set of recurring decimals.

The statements are true because they are mathmatically true not because they are defined to be true.

7. ### quantum_waveContemplating the "as yet" unknownValued Senior Member

Messages:
6,677
Thanks, I know they are true. I also was under the impression that the "..." was, as you say, to represent the infinite sequence, and so I used terms that I am familiar with like "notation" and "designation" to refer to what the use of "..." in place of an infinite sequence meant. But thanks for reminding me that mathematicians are clear about .999... and 1 being the same value.

8. ### Captain KremmenAll aboard, me Hearties!Valued Senior Member

Messages:
12,738
I shouldn't have mentioned the words "rounding up"
It really has nothing to do with my argument.

My argument is that the number 0.3333 recurring can be seen in two ways.
1. As the decimal equivalent of one third. For this purpose it is perfectly serviceable, but will forever be infinitesimally inaccurate,
even with an infinity of 3's. Three times this amount is equal to 1, not 0.9999 recurring.
2. As the number 0.3333 recurring. For this purpose the number is accurate. Three times this amount is 0.9999 recurring.

1 and 2 are not the same thing.

9. ### SarkusHippomonstrosesquippedalo phobeValued Senior Member

Messages:
10,155
I don't see it...

X = 0.9999... "we have one object with a value of 0.9999..."
10X = 9.9999... "we have 10 objects of 0.9999, but the VALUE is 9.9999...."
10X-X = 9 "we have 9 objects of 0.99999 and we have a value of 9."

There is no mixing that was not inherent in the opening statement.
You are reading 10X-X to be a quantity - value... when it is still quantity - quantity.
On the left of the "=" we continue to use quantity; on the right we continue to use value.

As said, the key is in accepting that 10X = 9.9999....

But I do not see a mixing of contexts.
But the bolded is NOT what the proof does.
That step should read: "10x - 1x = 10a - 1a"

So no context mixing.

If anything it simply begs the question in that it relies, as stated, on 10 lots of 0.9999.... being 9.9999.... and not 0.9999....0 (i.e. with a 0 at the end of the infinith(?) decimal place - if such is even meaningful to talk about), and that 9.9999... - 0.9999... (both values) = 9 (value).
But it is not context mixing as far as I can tell.

10. ### kwhilbornBannedBanned

Messages:
2,088
1 divided by 3 = .333...

yet

0.333 times 3 = 0.999... , and not the original 1 (Can't get there from here).

When the 9 is continued to infinity, there needs to be a logical assumption not based on the math that it is close enough. It keeps getting 9/10's closer infinitely.

One way to pose this problem using math and no logic assumptions is to try to subtract 0.999... from 1.

1 - 0.999... = 0.000... Which goes on infinitely and would have a one at the end if infinity had an end, but to anybody doing this in long division for a hundred years would see zero difference between 1 and 0.999...

So 0.999 = 1 is based on math as well as assumption.

Using Ellipsis is a one of several accepted methods of showing a repeating decimal.

11. ### Quantum QuackLife's a tease...Valued Senior Member

Messages:
23,328
it is an interesting logic issue and one worth pursuing IMO.

Possibly another thread devoted to this sort of thing might be in order.

Possibly one of the problems I see, after thinking about it a bit, is that
by using the simple : x= 0.9999....
we are transposing an irrational infinite repeater into a finite quantity called "x"
The value of x is "real" or "complete" or "whole" where as the value we wish to place upon it is irrational. [infinite repeating, infinitely incomplete]

As soon as we infer x=0.999.... we are already stating that x= 1x = 1(0.9999....) because there is only one x
in writing [including the the implied value of 1x]:
1x = 0.9999.....
1= 0.9999...
of course if x = 0.999.... there is no one x as x now equals less than one.

To then go on and claim x as being a real number quantity (1) is false as it is not (1) it is 0.999.....

the value of 1x is now an irrational rather than real 1x.
(I am not sure I have the terminology correct re: "real" and "irrational" - numbers )

any ways... thoughts..only

12. ### Quantum QuackLife's a tease...Valued Senior Member

Messages:
23,328
The problem is 1/3 does not equal 0.333....unless you add an infinitesimal to it.

There is a need to prove that 1/3 = 0.333... to begin with and I believe that it does not and can not equal exactly 1/3 unless an infinitesimal is added.

Can I suggest that that (1) at the end of an infinite set of (0)'s would be what is commonly refereed to as an "infinitesimal"

1 - 0.999... = 0 + (1/infinity)
or
1- 0.999... = 0 + 1(infinitesimal)

13. ### quantum_waveContemplating the "as yet" unknownValued Senior Member

Messages:
6,677
Hmm, interesting way to describe 'infinitesimal". Could we write it .000...1?

14. ### Quantum QuackLife's a tease...Valued Senior Member

Messages:
23,328
I am sure you could, however I thought, probably mistakenly, that an infinitesimal is described as 1(1/infinity)

15. ### rpennerFully WiredValued Senior Member

Messages:
4,833
For all numbers, there is an number greater than that number. Because x + 1 > x.
1 is a number.
0.999... is a number.
If 0.999... = 1 then 1 - 0.999... = 0. But this is not what the OP assumes.

Our hypothesis must be 1 > 0.999.. from which the following follows:

If 1 > 0.999... then 1 - 0.999... > 0.
Define x = 1 - 0.999....
Then x is a number. And x > 0.
Then 1/x is a number. Therefore there is a number N > 1/x.

Therefore our hypothesis leads invariably to $\exists N\quad N \gt \frac{1}{1 - \lim_{n \to \infty } \sum_{k=1}^{n} 9 \times 10^{-k} } = \lim_{n \to \infty } \frac{1}{1 - \sum_{k=1}^{n} 9 \times 10^{-k} } = \lim_{n \to \infty } \frac{1}{ 10^{-n} } = \lim_{n \to \infty } 10^n$.

But no such N can exists, because there is always a value of M which makes the above untrue for all values of n > M. We can even write:
$M = 1 + \lceil \log_{10} N \rceil$ and prove
$\forall n \geq M \quad N \not\gt 10^M \leq 10^n$ .

Therefore the hypothesis that $1 > 0.999...$ cannot be true because 1 - 0.999... is smaller than any positive number.
Not even if N = $(\textrm{1 googol})^{\tiny \textrm{1 googol}} = (10^{100})^{(10^{100})} = (10^{10^2})^{(10^{100})} = 10^{\left( 10^2 \times 10^{100} \right)} = 10^{10^{102}}$ can we avoid this Archimedean property of numbers.
There is always a larger number -- the natural numbers are without bound.
And so $M = 1 + \lceil \log_{10} 10^{10^{102}} \rceil = \textrm{100 googol} + 1$

Last edited: Oct 30, 2013
16. ### TrippyALEA IACTA ESTStaff Member

Messages:
10,890
This goes back to the comment I made earlier about needing to brush up on your long division.

17. ### Quantum QuackLife's a tease...Valued Senior Member

Messages:
23,328
I believe in my naive way that the invention of the infinitesimal was to allow Zeno's paradox to be resolved mathematically with out the Hare ever catching the Tortoise.

In other words zero can only be proved as a nul value when everything else reduces to the infinitesimal.

18. ### rpennerFully WiredValued Senior Member

Messages:
4,833
Define $\epsilon_Q = 1/3 - 0.333... \\ \epsilon_c = 1 - 0.999...$ in honor of QQ and the OP.

Then it follows that $3 \epsilon_Q = 3 \times ( 1/3 - 0.333... ) = 1 - 0.999... = \epsilon_c$ so $3 \epsilon_Q = \epsilon_c$, right?

So why is $\epsilon_Q \neq \epsilon_c$?

Doesn't your scheme require an infinite number of infinitesimals if $\epsilon_c \neq 0$ ? How is this an improvement on the Real Numbers which have no such infinitesimals?

19. ### Quantum QuackLife's a tease...Valued Senior Member

Messages:
23,328
I am not sure that I understand the question fully, [perhaps one day I will understand the notation you are using] but I tend to believe that the use of an infinitesimal is a way that mathematics can resolve the paradox using the legitimate "fudge" factor of 1(1/infinity).

It has to do I think with the philosophical question raised by Zeno's paradox and acutely demonstrated in the notion of 10/3 or 1/3 using base 10.

Extended:
Philosophically it sums up to the ability for mathematics to prove that "zero" is indeed a null value.
From what I have come to understand after exhaustive philosophical discussion, is that zero can only be proved as a "nul" quantity or value with the use of infinitesimals as being the smallest quantity or value, we can go with out arriving at zero. Thus proving zero only by default and not directly. Deductive reasoning or logic? perhaps...
An infinitesimal is unable to be rationally multiplied or divided as it's definition is always going to be 1(1/infinity) regardless of the number of times it is used.
Same argument would I think apply to infinity * infinity = infinity...

To cut a portion of cake exactly 1/3 [in absolutum] is impossible [just like it is to resolve Pi] as infinite reduction towards the infinitesimal, as a part of measuring that 1/3, is all that is possible.

The other issue worth mentioning is that looking at any whole number... it may be realised that the "wholeness" is only available when used in the context of "Quantity" and not "Quality" [aka value]

Example:
We have 10 apples each weighing 1 kg.

10 apples is true [quantity]
1 kg exactly is false [ due to infinite reduction.] [quality - value] Suffice to say it can never be exactly 1kg [in absolutum]

The issue displayed in the process Sarkus mentioned is I believe erroneous due to the fact that it mixes quantity with quality and there is a subtle but important loss of contextual consistency in that process. [as exampled above with the apples]

Last edited: Oct 30, 2013
20. ### Quantum QuackLife's a tease...Valued Senior Member

Messages:
23,328
using number of apples and weight

1 apple = 0.999...kg
10 apples = 10 (0.999...kg)

10 apples - 1 apple = 9 apples
9 apples = 9 apples
1 apple = 1 apple

the issue appears to be when we multiply our apples by their weight. instead of maintaining Quantity as context we shift to a mix of quality and quantity.
To arrive at 1= 0.999... is simply saying that yes 1 apple weighs 0.999...kg

21. ### kwhilbornBannedBanned

Messages:
2,088
@ QQ,

No. saying 1 = 0.999... is an equation. If you are going to state that both sides of the equation are unrelated and that one side is an apple, yet the other side is the weight is incorrect.

The matter is best seen if you subtract 0.999... from 1 as I suggested in my last post, followed by similar statements by Rpenner.

22. ### Quantum QuackLife's a tease...Valued Senior Member

Messages:
23,328
the problem is we start out by not saying 1= 0.999.... we start out by saying X= 0.999...

big difference..
x is an algebraic symbol that is being replaced with the value 0.999....
thus the equation is simply
x = 0.999...
or transposed
0.999.... = 0.999...

X is being treated as whole number where as it is not a whole number. It is merely an algebraic symbol.
so to say that x= 0.999... is simply replacing x for the value of 0.999... hence the result 0.999...= 0.999....

so if
X = 0.999.....
10X - X = 9x
9X = 9x
X = X

Therefore X = X = 0.999.... = 0.999...
x cannot be replace by a 1
the other lines are redundant and unnecessary if one sticks to the value of x= x

The original version again:
X = 0.99999.....
10X = 9.999999....
10X - X = 9 <<<:replaces 9X with a real number... logical inconsistency (loss of context)
9X = 9
X = 1 <<: irrational number now converted to real

Therefore X = 1 = 0.9999.... <<: 1 is now an irrational number

Why is my version incorrect and the one below it deemed correct?
try

x = 0.7777...
10x=7.7777...
10x-x = 7
9x = 7
therefore 1= 1.285714586 (9/7)
which of course is total BS...

using the method suggested could lead to 1 being just about anything...

23. ### PeteIt's not rocket surgeryRegistered Senior Member

Messages:
10,167
QQ, you don't seem to understand how algebra works.

So far so good.
Next step:
x = 7/9

You've divided the let hand side by 9x, and the right hand side by 9. Are you thinking that x=1, instead of x=0.777...?