• Re: A game like billards

    From WM@21:1/5 to William on Thu Dec 14 16:22:50 2023
    On 14.12.2023 15:23, William wrote:
    On Thursday, December 14, 2023 at 6:08:40 AM UTC-4, WM wrote:
    On 13.12.2023 20:41, William wrote:
    need dark numbers.
    Only producing the putative A shows that "dark" numbers are needed.
    No. Producing B

    does not involve a limit

    It involves the realization of all natural numbers
    k = (m + n - 1)(m + n - 2)/2 + m
    at matrix places
    1/1, 1/2, 2/1, 1/3, 2/2, 3/1, 1/4, 2/3, 3/2, 4/1, 1/5, 2/4, 3/3, 4/2,
    5/1, 1/6, 2/5, 3/4, 4/3, 5/2, 6/1, ...

    This requires a limit or a last step. Your belief alone is not sufficient.

    Regards, WM

    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)
  • From Mikko@21:1/5 to All on Fri Dec 15 11:58:20 2023
    On 2023-12-14 15:22:50 +0000, WM said:

    On 14.12.2023 15:23, William wrote:
    On Thursday, December 14, 2023 at 6:08:40 AM UTC-4, WM wrote:
    On 13.12.2023 20:41, William wrote:
    need dark numbers.
    Only producing the putative A shows that "dark" numbers are needed.
    No. Producing B

    does not involve a limit

    It involves the realization of all natural numbers
    k = (m + n - 1)(m + n - 2)/2 + m
    at matrix places
    1/1, 1/2, 2/1, 1/3, 2/2, 3/1, 1/4, 2/3, 3/2, 4/1, 1/5, 2/4, 3/3, 4/2,
    5/1, 1/6, 2/5, 3/4, 4/3, 5/2, 6/1, ...

    This requires a limit or a last step.

    You say but don't show. Your belief alone is not sufficient.

    Mikko

    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)
  • From WM@21:1/5 to Mikko on Fri Dec 15 13:12:36 2023
    On 15.12.2023 10:58, Mikko wrote:
    On 2023-12-14 15:22:50 +0000, WM said:

    This requires a limit or a last step.

    You say but don't show.
    What proves completeness in your opinion?

    Regards, WM

    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)
  • From Jim Burns@21:1/5 to All on Fri Dec 15 10:57:07 2023
    On 12/15/2023 7:12 AM, WM wrote:
    On 15.12.2023 10:58, Mikko wrote:
    On 2023-12-14 15:22:50 +0000, WM said:

    This requires a limit or a last step.

    You say but don't show.

    What proves completeness in your opinion?

    In my opinion,
    the answer is as difficult as it is
    because it's too obvious.
    https://en.wikipedia.org/wiki/The_Purloined_Letter
    | however, Dupin conjectured that it would instead
    | be hidden in plain sight.

    Write something which is true of
    each one of multiple objects.
    For example,
    |
    | x is the point between the fore- and hindpart
    | of a non-empty split F,H of the rationals ℚ

    What's hidden in plain sight is:
    when those are the points we're discussing,
    |
    | x is the point between the fore- and hindpart
    | of a non-empty split F,H of the rationals ℚ
    |
    is true for each of what we're discussing.
    Complete.

    In instances involving infinity,
    we do not _accomplish_ completeness.
    We do not step
    from incomplete to complete.
    (Unless we are Chuck Norris.)
    (Chuck Norris counted to infinity.)
    (Twice.)

    For lesser beings, our justified claim of
    completeness involving infinity
    has a too.obvious initial completeness
    and only makes completeness.preserving
    further claims.

    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)
  • From WM@21:1/5 to Jim Burns on Fri Dec 15 18:09:24 2023
    On 15.12.2023 16:57, Jim Burns wrote:
    On 12/15/2023 7:12 AM, WM wrote:
    On 15.12.2023 10:58, Mikko wrote:
    On 2023-12-14 15:22:50 +0000, WM said:

    This requires a limit or a last step.

    You say but don't show.

    What proves completeness in your opinion?

    Write something which is true of
    each one of multiple objects.

    That does not make it true. There are differences.
    Example: Every fraction is the quotient of two natnumbers. That is true.
    Every fraction can be isolated from all smaller fractions by an eps
    between them. That is wrong.

    For example,
    |
    | x is the point between the fore- and hindpart
    | of a non-empty split F,H of the rationals ℚ

    What's hidden in plain sight is:
    when those are the points we're discussing,
    |
    | x is the point between the fore- and hindpart
    | of a non-empty split F,H of the rationals ℚ
    |
    is true for each of what we're discussing.
    Complete.

    As far as I can understand, you say the same as I.

    In instances involving infinity,
    we do not _accomplish_ completeness.

    That's it!

    Regards, WM

    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)
  • From Jim Burns@21:1/5 to All on Fri Dec 15 13:31:27 2023
    On 12/15/2023 12:09 PM, WM wrote:
    On 15.12.2023 16:57, Jim Burns wrote:
    On 12/15/2023 7:12 AM, WM wrote:
    On 15.12.2023 10:58, Mikko wrote:
    On 2023-12-14 15:22:50 +0000, WM said:

    This requires a limit or a last step.

    You say but don't show.

    What proves completeness in your opinion?

    Write something which is true of
    each one of multiple objects.

    That does not make it true.

    No.
    Something which is true of
    each one of multiple objects
    is
    something which is true of
    each one of multiple objects.

    Perhaps this is too obvious,
    hidden in plain sight.

    If,
    on the other hand,
    you have picked something else to say,
    something which _isn't_ true of
    each one of the multiple objects
    which you intend to discuss,
    then
    pick a different thing to say.

    For example,
    |
    | x is the point between the fore- and hindpart
    | of a non-empty split F,H of the rationals ℚ

    I picked this thing to say because
    a function of _all_ those points,
    including points.between (real numbers),
    which is continuous at each point
    doesn't jump.

    I'm familiar with contexts in which
    it is useful to say of a function that
    it doesn't jump.

    This is not a fantasy.example,
    not Scrooge McDuck's vaults.
    How we say _what we intend to say_
    which is that a function doesn't jump,
    is to say something once which is
    true in infinitely.many ways.

    In instances involving infinity,
    we do not _accomplish_ completeness.

    That's it!

    Our methods take that into account.
    We begin at completeness and
    maintain completeness as we learn
    by not-first-false augmenting.

    We don't accomplish completeness and
    we don't need to.

    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)
  • From Richard Damon@21:1/5 to All on Fri Dec 15 20:05:46 2023
    On 12/15/23 11:51 AM, WM wrote:
    Fritz Feldhase schrieb am Freitag, 15. Dezember 2023 um 14:36:48 UTC+1:
    On Friday, December 15, 2023 at 1:12:39 PM UTC+1, WM wrote:

    What proves completeness [...]?

    A DEFINITION of /completeness/ and a PROOF (of a statement referring to completeness).

    Completeness in counting the natural numbers cannot be proven because the contrary can be proven. And that is simple: For every eps > 0 there are only finitely many unit fractions in (eps, 1] counted, but infinitely many in (0, eps] uncounted and
    uncountable because eps cannot be made small enough to leave only finitely many uncounted. Therefore every definition of completeness in this respect is nonsense.

    Regards, WM
    So NOTHING in your logic system can completely talk about the whole of
    Natural Numbers, so it just can't handle them.

    Your logic is BOUNDED, but the Numbers are UNBOUNDED.

    Thus, your logic fails you.

    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)
  • From Mikko@21:1/5 to All on Sat Dec 16 10:07:03 2023
    On 2023-12-15 12:12:36 +0000, WM said:

    On 15.12.2023 10:58, Mikko wrote:
    On 2023-12-14 15:22:50 +0000, WM said:

    This requires a limit or a last step.

    You say but don't show.
    What proves completeness in your opinion?

    A proof of completeness. Details depend on what kind of completeness
    you mean.

    Mikko

    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)
  • From WM@21:1/5 to All on Sat Dec 16 12:18:21 2023
    Le 16/12/2023 à 09:07, Mikko a écrit :
    On 2023-12-15 12:12:36 +0000, WM said:

    On 15.12.2023 10:58, Mikko wrote:
    On 2023-12-14 15:22:50 +0000, WM said:

    This requires a limit or a last step.

    You say but don't show.
    What proves completeness in your opinion?

    A proof of completeness. Details depend on what kind of completeness
    you mean.

    You "prove" that Cantor enumerates all fractions. That is wrong. The
    matrix
    XOOO...
    XOOO...
    XOOO...
    XOOO...
    ..will never be covered by indeXes. The indeX transferred from matrix
    position (k, j) to matrix position (m, n) implies that, in exchange, O is transferred from matrix position (m, n) to matrix position (k, j). An O is indicating a not indexed position. No O will ever get off of the matrix.
    Hence not all positions will be indeXed. Therefore your "proofs" are
    worthless.

    Regards, WM

    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)
  • From WM@21:1/5 to All on Sat Dec 16 12:26:45 2023
    Le 16/12/2023 à 02:05, Richard Damon a écrit :
    On 12/15/23 11:51 AM, WM wrote:
    Fritz Feldhase schrieb am Freitag, 15. Dezember 2023 um 14:36:48 UTC+1:
    On Friday, December 15, 2023 at 1:12:39 PM UTC+1, WM wrote:

    What proves completeness [...]?

    A DEFINITION of /completeness/ and a PROOF (of a statement referring to
    completeness).

    Completeness in counting the natural numbers cannot be proven because the
    contrary can be proven. And that is simple: For every eps > 0 there are only >> finitely many unit fractions in (eps, 1] counted, but infinitely many in (0, eps]
    uncounted and uncountable because eps cannot be made small enough to leave only
    finitely many uncounted. Therefore every definition of completeness in this respect
    is nonsense.

    So NOTHING in your logic system can completely talk about the whole of Natural Numbers, so it just can't handle them.

    Neither can yours.
    XOOO...
    XOOO...
    XOOO...
    XOOO...
    ..
    The indeX transferred from matrix position (k, j) to matrix position (m,
    n) implies that, in exchange, O is transferred from matrix position (m, n)
    to matrix position (k, j). An O is indicating a not indexed position. No O
    will ever get off of the matrix. Hence not all positions will be indeXed.

    But you believe that. So you are provably wrong.

    Regards, WM

    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)
  • From Richard Damon@21:1/5 to All on Sat Dec 16 07:48:04 2023
    On 12/16/23 7:26 AM, WM wrote:
    Le 16/12/2023 à 02:05, Richard Damon a écrit :
    On 12/15/23 11:51 AM, WM wrote:
    Fritz Feldhase schrieb am Freitag, 15. Dezember 2023 um 14:36:48 UTC+1: >>>> On Friday, December 15, 2023 at 1:12:39 PM UTC+1, WM wrote:

    What proves completeness [...]?

    A DEFINITION of /completeness/ and a PROOF (of a statement referring
    to completeness).

    Completeness in counting the natural numbers cannot be proven because
    the contrary can be proven. And that is simple: For every eps > 0
    there are only finitely many unit fractions in (eps, 1] counted, but
    infinitely many in (0, eps] uncounted and uncountable because eps
    cannot be made small enough to leave only finitely many uncounted.
    Therefore every definition of completeness in this respect is nonsense.

    So NOTHING in your logic system can completely talk about the whole of
    Natural Numbers, so it just can't handle them.

    Neither can yours. XOOO...
    XOOO...
    XOOO...
    XOOO...
    ..
    The indeX transferred from matrix position (k, j) to matrix position (m,
    n) implies that, in exchange, O is transferred from matrix position (m,
    n) to matrix position (k, j). An O is indicating a not indexed position.
    No O will ever get off of the matrix. Hence not all positions will be indeXed.

    But you believe that. So you are provably wrong.

    Regards, WM


    But that arguement is irrelevent. There are many ways that fail to index
    the set, but the key is that they are shown equal if there exist An
    indexing that matches them.

    So, restricting yourself to only an indexing that doesn;t work is
    meaningless.

    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)
  • From WM@21:1/5 to All on Sat Dec 16 13:20:55 2023
    Le 16/12/2023 à 13:48, Richard Damon a écrit :
    On 12/16/23 7:26 AM, WM wrote:
    Le 16/12/2023 à 02:05, Richard Damon a écrit :
    On 12/15/23 11:51 AM, WM wrote:
    Fritz Feldhase schrieb am Freitag, 15. Dezember 2023 um 14:36:48 UTC+1: >>>>> On Friday, December 15, 2023 at 1:12:39 PM UTC+1, WM wrote:

    What proves completeness [...]?

    A DEFINITION of /completeness/ and a PROOF (of a statement referring >>>>> to completeness).

    Completeness in counting the natural numbers cannot be proven because
    the contrary can be proven. And that is simple: For every eps > 0
    there are only finitely many unit fractions in (eps, 1] counted, but
    infinitely many in (0, eps] uncounted and uncountable because eps
    cannot be made small enough to leave only finitely many uncounted.
    Therefore every definition of completeness in this respect is nonsense. >>>>
    So NOTHING in your logic system can completely talk about the whole of
    Natural Numbers, so it just can't handle them.

    Neither can yours. XOOO...
    XOOO...
    XOOO...
    XOOO...
    ..
    The indeX transferred from matrix position (k, j) to matrix position (m,
    n) implies that, in exchange, O is transferred from matrix position (m,
    n) to matrix position (k, j). An O is indicating a not indexed position.
    No O will ever get off of the matrix. Hence not all positions will be
    indeXed.

    But you believe that. So you are provably wrong.

    But that arguement is irrelevent.

    No,it is exactly Cantor's way. Only one step is inserted: The first column
    is put in bojection with the natural numbers: 1/n <--> n. But if that is
    not posible, then all is impossible.

    There are many ways that fail to index
    the set,

    There are only ways that fail. Otherwise every injection would be a
    bijection.

    but the key is that they are shown equal if there exist An
    indexing that matches them.

    It is hard to understand how Cantor could convince the matheologians of
    this silly idea. But in order to help you I did use Cantor's formula k =
    (m + n - 1)(m + n - 2)/2 + m with the resulting sequence 1/1, 1/2, 2/1,
    1/3, 2/2, 3/1, 1/4, 2/3, 3/2, 4/1, 1/5, 2/4, 3/3, 4/2, 5/1, 1/6, 2/5, 3/4,
    4/3, 5/2, 6/1, ...

    Regards, WM

    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)
  • From WM@21:1/5 to All on Sat Dec 16 13:27:22 2023
    Le 15/12/2023 à 19:31, Jim Burns a écrit :
    On 12/15/2023 12:09 PM, WM wrote:

    Something which is true of
    each one of multiple objects
    is
    something which is true of
    each one of multiple objects.

    Never an O leaves the matrix.

    I'm familiar with contexts in which
    it is useful to say of a function that
    it doesn't jump.

    Here for instance: ∀n ∈ ℕ: 1/n - 1/(n+1) = d_n > 0

    This is not a fantasy.example,
    not Scrooge McDuck's vaults.
    How we say _what we intend to say_
    which is that a function doesn't jump,
    is to say something once which is
    true in infinitely.many ways.

    In instances involving infinity,
    we do not _accomplish_ completeness.

    That's it!

    Our methods take that into account.
    We begin at completeness and
    maintain completeness as we learn
    by not-first-false augmenting.

    The set of O remains completely in the matrix.

    We don't accomplish completeness and
    we don't need to.

    A bijection woul require it.

    Regards, WM

    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)
  • From Richard Damon@21:1/5 to All on Sat Dec 16 09:25:39 2023
    On 12/16/23 8:20 AM, WM wrote:
    Le 16/12/2023 à 13:48, Richard Damon a écrit :
    On 12/16/23 7:26 AM, WM wrote:
    Le 16/12/2023 à 02:05, Richard Damon a écrit :
    On 12/15/23 11:51 AM, WM wrote:
    Fritz Feldhase schrieb am Freitag, 15. Dezember 2023 um 14:36:48
    UTC+1:
    On Friday, December 15, 2023 at 1:12:39 PM UTC+1, WM wrote:

    What proves completeness [...]?

    A DEFINITION of /completeness/ and a PROOF (of a statement
    referring to completeness).

    Completeness in counting the natural numbers cannot be proven
    because the contrary can be proven. And that is simple: For every
    eps > 0 there are only finitely many unit fractions in (eps, 1]
    counted, but infinitely many in (0, eps] uncounted and uncountable
    because eps cannot be made small enough to leave only finitely many
    uncounted. Therefore every definition of completeness in this
    respect is nonsense.

    So NOTHING in your logic system can completely talk about the whole
    of Natural Numbers, so it just can't handle them.

    Neither can yours. XOOO...
    XOOO...
    XOOO...
    XOOO...
    ..
    The indeX transferred from matrix position (k, j) to matrix position
    (m, n) implies that, in exchange, O is transferred from matrix
    position (m, n) to matrix position (k, j). An O is indicating a not
    indexed position. No O will ever get off of the matrix. Hence not all
    positions will be indeXed.

    But you believe that. So you are provably wrong.

    But that arguement is irrelevent.

    No,it is exactly Cantor's way. Only one step is inserted: The first
    column is put in bojection with the natural numbers: 1/n <--> n. But if
    that is not posible, then all is impossible.

    Yes, you can biject n <--> 1/n, you can also biject k <--> n/m with a
    different mapping, This is NOT a "Contradiction"

    The fact that the first mapping missed an infinite number of terms of
    the second is irrelvent.


    There are many ways that fail to index the set,

    There are only ways that fail. Otherwise every injection would be a bijection.

    Nope, you are confusing your qualifiers. If there exists at least one
    mapping that is ONE to ONE covering all, we have a bijection.

    If we can establish a mapping of every element of the first set to some
    unique element in the second (maybe with some left over) and another
    mapping of every element of the second to a unique element of the first
    (maybe with some left over) we can also establish they are of the same size.


    but the key is that they are shown equal if there exist An indexing
    that matches them.

    It is hard to understand how Cantor could convince the matheologians of
    this silly idea. But in order to help you I did use Cantor's formula k =
    (m + n - 1)(m + n - 2)/2 + m with the resulting sequence 1/1, 1/2, 2/1,
    1/3, 2/2, 3/1, 1/4, 2/3, 3/2, 4/1, 1/5, 2/4, 3/3, 4/2, 5/1, 1/6, 2/5,
    3/4, 4/3, 5/2, 6/1,  ...

    Regards, WM


    So, there exists A mapping of N to N/N so they are the same size.

    This is just a FACT of the behavior of infinite sets.

    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)
  • From Jim Burns@21:1/5 to All on Sat Dec 16 14:27:27 2023
    On 12/16/2023 8:27 AM, WM wrote:
    Le 15/12/2023 à 19:31, Jim Burns a écrit :
    On 12/15/2023 12:09 PM, WM wrote:

    In instances involving infinity,
    we do not _accomplish_ completeness.

    That's it!

    Our methods take that into account.
    We begin at completeness and
    maintain completeness as we learn
    by not-first-false augmenting.

    The set of O remains completely
    in the matrix.

    We don't accomplish completeness and
    we don't need to.

    A bijection [would] require it.

    ⟨i,j⟩ ⟼ ⟨k,1⟩
    k = i+(i+j-1)(i+j-2)/2

    ⟨k,1⟩ ⟼ ⟨i,j⟩
    sₖ = max{h| (H-1)(h-2)/2 < k }
    i+j = sₖ
    i = k-(sₖ-1)(sₖ-2)/2
    j = sₖ(sₖ-1)/2-k+1

    That's complete.
    We didn't accomplishᵂᴹ that completeness.
    We didn't need to accomplishᵂᴹ it.
    It's never incomplete.

    We began with some statements true of
    each one which we're discussing, such as
    | k is in
    | ordered ⟨1,…,k⟩ such that,
    | for each non-empty split of ⟨1,…,k⟩
    | i‖i+1 is last‖first in F‖H and
    | 1‖k is first‖last in ⟨1,…,k⟩
    |
    Initially.complete.

    We augmented with only
    claims not.first.false of
    each one of what we're discussing.

    If the descriptive claims are true,
    as they are selected to be for
    each one of what we're discussing,
    then
    the not.first.false augmenting claims are in
    a finite sequence of only not.first.false

    A finite sequence of only not.first.false
    cannot include a false claim.

    If the descriptive claims are true,
    as they are selected to be true for
    each one of what we're discussing,
    then
    the not.first.false augmenting claims are
    also true of each one of what we're discussing.
    Maintained.complete.

    Something which is true of
    each one of multiple objects
    is
    something which is true of
    each one of multiple objects.

    Never an O leaves the matrix.

    If O is at ⟨i,j⟩
    then ⟨i,j⟩↔⟨kᵢⱼ,1⟩ is not swapped
    kᵢⱼ = i+(i+j-1)(i+j-2)/2

    If, for each ⟨i,j⟩
    ⟨i,j⟩↔⟨kᵢⱼ,1⟩ is swapped
    then, for each ⟨i,j⟩
    O is not at it.

    After all ⟨i,j⟩↔⟨kᵢⱼ,1⟩ are swapped
    O is not at each ⟨i,j⟩

    The O's are overwritten by X's from ⟨k,1⟩
    from proper.subset first.column,
    which is same.sized as the whole matrix.

    Herds of sheep and bags of pebbles
    do not have same.sized proper.subsets

    However,
    these rows, columns, fractions, and swaps
    are not.like.sheep in that respect.
    They do have same.sized proper.subsets.

    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)
  • From Mikko@21:1/5 to All on Sun Dec 17 10:53:45 2023
    On 2023-12-16 12:18:21 +0000, WM said:

    Le 16/12/2023 à 09:07, Mikko a écrit :
    On 2023-12-15 12:12:36 +0000, WM said:

    On 15.12.2023 10:58, Mikko wrote:
    On 2023-12-14 15:22:50 +0000, WM said:

    This requires a limit or a last step.

    You say but don't show.
    What proves completeness in your opinion?

    A proof of completeness. Details depend on what kind of completeness
    you mean.

    You "prove" that Cantor enumerates all fractions.

    No, I don't. I prefer to prove that one can associate a natural
    number (or several natural numbers) to each positive rational number
    so that every natural number is associated to at most one raltional
    number. Or, if that is not sufficient, that there is a bijection
    between natural numbers and rationals.

    That is wrong. The matrix
    XOOO...
    XOOO...
    XOOO...
    XOOO...
    ..will never be covered by indeXes.

    Yes it is. Every position in the matrix has a row index and a column
    indes. Positions not covered are not part of the matrix.

    The indeX transferred from matrix position (k, j) to matrix position (m, n)

    What does "transferred" mean?

    implies that,

    and how would a "transferred" index imply anything?

    Mikko

    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)
  • From WM@21:1/5 to All on Sun Dec 17 19:36:59 2023
    Le 17/12/2023 à 09:53, Mikko a écrit :
    On 2023-12-16 12:18:21 +0000, WM said:

    That is wrong. The matrix
    XOOO...
    XOOO...
    XOOO...
    XOOO...
    ..will never be covered by indeXes.

    Yes it is. Every position in the matrix has a row index and a column
    indes. Positions not covered are not part of the matrix.

    There are not enough indices.

    The indeX transferred from matrix position (k, j) to matrix position (m, n)

    What does "transferred" mean?

    If (m, n) is to be indexed according to Cantors formula, then the indeX is taken from its position (k, j).

    Regards, WM

    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)
  • From WM@21:1/5 to All on Sun Dec 17 19:41:20 2023
    Le 16/12/2023 à 20:27, Jim Burns a écrit :
    On 12/16/2023 8:27 AM, WM wrote:

    Never an O leaves the matrix.

    The O's are overwritten by X's from ⟨k,1⟩

    But they remain in the matrix. The set of positione covered by O minus the
    set of positions covered by X is infinite and will never decrease.

    Regards, WM

    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)
  • From WM@21:1/5 to All on Sun Dec 17 19:44:50 2023
    Le 16/12/2023 à 15:25, Richard Damon a écrit :

    Yes, you can biject n <--> 1/n, you can also biject k <--> n/m with a different mapping, This is NOT a "Contradiction"

    The fact that the first mapping missed an infinite number of terms of
    the second is irrelvent.

    Not in mathematics!

    But the fact that the second mapping misses an infinite number of terms,
    proved by the eternal presence of O, proves that Cantor's way fails.

    Regards, WM

    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)
  • From Richard Damon@21:1/5 to All on Sun Dec 17 15:23:54 2023
    On 12/17/23 2:44 PM, WM wrote:
    Le 16/12/2023 à 15:25, Richard Damon a écrit :

    Yes, you can biject n <--> 1/n, you can also biject k <--> n/m with a
    different mapping, This is NOT a "Contradiction"

    The fact that the first mapping missed an infinite number of terms of
    the second is irrelvent.

    Not in mathematics!

    But the fact that the second mapping misses an infinite number of terms, proved by the eternal presence of O, proves that Cantor's way fails.

    Regards, WM

    Yes, in mathematics of infinite sets, the fact that one mapping misses
    even an infinite number of members means nothing, only the existance of
    some mapping that meets the requirements.

    You just don't understand the infinite.

    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)
  • From WM@21:1/5 to All on Sun Dec 17 21:18:02 2023
    Le 17/12/2023 à 21:23, Richard Damon a écrit :

    Yes, in mathematics of infinite sets, the fact that one mapping misses
    even an infinite number of members means nothing,

    But here Cantor's mapping misses an infinite number of fractions.

    Regards, WM

    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)
  • From Richard Damon@21:1/5 to All on Sun Dec 17 19:09:37 2023
    On 12/17/23 4:18 PM, WM wrote:
    Le 17/12/2023 à 21:23, Richard Damon a écrit :

    Yes, in mathematics of infinite sets, the fact that one mapping misses
    even an infinite number of members means nothing,

    But here Cantor's mapping misses an infinite number of fractions.

    Regards, WM





    You aren't using the right mapping, showing your ignorance.

    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)
  • From Mikko@21:1/5 to All on Mon Dec 18 11:44:10 2023
    On 2023-12-17 19:36:59 +0000, WM said:

    Le 17/12/2023 à 09:53, Mikko a écrit :
    On 2023-12-16 12:18:21 +0000, WM said:

    That is wrong. The matrix
    XOOO...
    XOOO...
    XOOO...
    XOOO...
    ..will never be covered by indeXes.

    Yes it is. Every position in the matrix has a row index and a column
    indes. Positions not covered are not part of the matrix.

    There are not enough indices.

    Then the matrix is too small, but every entry in the matrix has
    tu indices.

    The indeX transferred from matrix position (k, j) to matrix position (m, n) >>
    What does "transferred" mean?

    If (m, n) is to be indexed according to Cantors formula, then the indeX
    is taken from its position (k, j).

    How does that imply anything?

    Mikko

    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)
  • From WM@21:1/5 to All on Mon Dec 18 10:47:35 2023
    Le 18/12/2023 à 01:09, Richard Damon a écrit :
    On 12/17/23 4:18 PM, WM wrote:
    Le 17/12/2023 à 21:23, Richard Damon a écrit :

    Yes, in mathematics of infinite sets, the fact that one mapping misses
    even an infinite number of members means nothing,

    But here Cantor's mapping misses an infinite number of fractions.

    You aren't using the right mapping, showing your ignorance.

    Where does my mapping deviate from Cantor's
    k = (m + n - 1)(m + n - 2)/2 + m
    with the result
    1/1, 1/2, 2/1, 1/3, 2/2, 3/1, 1/4, 2/3, 3/2, 4/1, 1/5, 2/4, 3/3, 4/2, 5/1,
    1/6, 2/5, 3/4, 4/3, 5/2, 6/1, ...
    for the first time?

    Regards, WM

    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)
  • From WM@21:1/5 to All on Mon Dec 18 10:52:59 2023
    Le 18/12/2023 à 10:44, Mikko a écrit :
    On 2023-12-17 19:36:59 +0000, WM said:

    The indeX transferred from matrix position (k, j) to matrix position (m, n)

    What does "transferred" mean?

    If (m, n) is to be indexed according to Cantors formula, then the indeX
    is taken from its position (k, j).

    How does that imply anything?

    The indices are distributet according to Cantor's formula
    k = (m + n - 1)(m + n - 2)/2 + m
    resulting in
    1/1, 1/2, 2/1, 1/3, 2/2, 3/1, 1/4, 2/3, 3/2, 4/1, 1/5, 2/4, 3/3, 4/2, 5/1,
    1/6, 2/5, 3/4, 4/3, 5/2, 6/1, ...
    but where the indices are removed, there are no indices.

    Index 1 remains at fraction 1/1, the first term of the sequence.

    1, 1/2, 1/3, 1/4, ...
    2, 2/2, 2/3, 2/4, ...
    3, 3/2, 3/3, 3/4, ...
    4, 4/2, 4/3, 4/4, ...
    5, 5/2, 5/3, 5/4, ...
    ..


    The next term, 1/2, is indexed with 2 which is taken from its initial
    position 2/1

    XXOO...
    OOOO...
    XOOO...
    XOOO...
    XOOO...
    ..

    Then index 3 is taken from its initial position 3/1 and is attached to 2/1

    XXOO...
    XOOO...
    OOOO...
    XOOO...
    XOOO...
    ..

    Then index 4 is taken from its initial position 4/1 and is attached to 1/3

    XXXO...
    XOOO...
    OOOO...
    OOOO...
    XOOO...
    ..

    Then index 5 is taken from its initial position 5/1 and is attached to 2/2

    XXXO...
    XXOO...
    OOOO...
    OOOO...
    OOOO...
    ..

    The O will remain forever, indicating not indexed positions.

    Regards, WM

    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)
  • From Richard Damon@21:1/5 to All on Mon Dec 18 10:00:22 2023
    On 12/18/23 5:47 AM, WM wrote:
    Le 18/12/2023 à 01:09, Richard Damon a écrit :
    On 12/17/23 4:18 PM, WM wrote:
    Le 17/12/2023 à 21:23, Richard Damon a écrit :

    Yes, in mathematics of infinite sets, the fact that one mapping
    misses even an infinite number of members means nothing,

    But here Cantor's mapping misses an infinite number of fractions.

    You aren't using the right mapping, showing your ignorance.

    Where does my mapping deviate from Cantor's
    k = (m + n - 1)(m + n - 2)/2 + m
    with the result
    1/1, 1/2, 2/1, 1/3, 2/2, 3/1, 1/4, 2/3, 3/2, 4/1, 1/5, 2/4, 3/3, 4/2,
    5/1, 1/6, 2/5, 3/4, 4/3, 5/2, 6/1,  ...
    for the first time?

    Regards, WM





    Which shows that your matrix B having values like

    1/1 2/1 3/1 ...
    1/2 2/2 3/2 ...
    1/3 2/3 3/3 ...
    .
    .
    .

    is exactly indexed by a matrix form like:

    1 3 6 10
    2 5 9
    4 8 13
    7 12
    11

    Where the number at the m,n location of your fractions is indexed by the
    number at the m,n location of the Natural Numbers organized in that fashion.

    You seem to think that just because you can make a pattern that doesn't
    cover it (like trying to go all the way down first), it means you can't
    cover it.

    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)
  • From Richard Damon@21:1/5 to All on Mon Dec 18 11:06:37 2023
    On 12/18/23 5:52 AM, WM wrote:
    Le 18/12/2023 à 10:44, Mikko a écrit :
    On 2023-12-17 19:36:59 +0000, WM said:

    The indeX transferred from matrix position (k, j) to matrix
    position (m, n)

    What does "transferred" mean?

    If (m, n) is to be indexed according to Cantors formula, then the
    indeX is taken from its position (k, j).

    How does that imply anything?

    The indices are distributet according to Cantor's formula
    k = (m + n - 1)(m + n - 2)/2 + m
    resulting in
    1/1, 1/2, 2/1, 1/3, 2/2, 3/1, 1/4, 2/3, 3/2, 4/1, 1/5, 2/4, 3/3, 4/2,
    5/1, 1/6, 2/5, 3/4, 4/3, 5/2, 6/1,  ...
    but where the indices are removed, there are no indices.

    Index 1 remains at fraction 1/1, the first term of the sequence.

    1, 1/2, 1/3, 1/4, ...
    2, 2/2, 2/3, 2/4, ...
    3, 3/2, 3/3, 3/4, ...
    4, 4/2, 4/3, 4/4, ...
    5, 5/2, 5/3, 5/4, ...
    ..


    The next term, 1/2, is indexed with 2 which is taken from its initial position 2/1

    XXOO...
    OOOO...
    XOOO...
    XOOO...
    XOOO...
    ..

    Then index 3 is taken from its initial position 3/1 and is attached to 2/1

    XXOO...
    XOOO...
    OOOO...
    XOOO...
    XOOO...
    ..

    Then index 4 is taken from its initial position 4/1 and is attached to 1/3

    XXXO...
    XOOO...
    OOOO...
    OOOO...
    XOOO...
    ..

    Then index 5 is taken from its initial position 5/1 and is attached to 2/2

    XXXO...
    XXOO...
    OOOO...
    OOOO...
    OOOO...
    ..

    The O will remain forever, indicating not indexed positions.

    Regards, WM






    And the problem is you aren't supposed to be taking the indexes out of
    B, but out of the matrix A where the value k was at the position
    indicated by m,n

    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)
  • From Jim Burns@21:1/5 to All on Mon Dec 18 16:23:21 2023
    On 12/17/2023 2:41 PM, WM wrote:
    Le 16/12/2023 à 20:27, Jim Burns a écrit :
    On 12/16/2023 8:27 AM, WM wrote:

    Never an O leaves the matrix.

    The O's are overwritten by X's from ⟨k,1⟩

    But they remain in the matrix.

    For each ⟨i,j⟩ in the matrix
    there is ⟨i,j⟩⇄⟨k,1⟩ in the swaps such that
    ⟨k,1⟩ is in the matrix
    ⟨i,j⟩⇄⟨k,1⟩ is first.swap for ⟨k,1⟩
    ⟨i,j⟩⇄⟨k,1⟩ is last.swap for ⟨i,j⟩

    For each ⟨i,j⟩ in the matrix
    after all swaps,
    X which was initially at ⟨k,1⟩ is at ⟨i,j⟩
    and O is not-at ⟨i,j⟩

    The set of positione covered by O
    minus
    the set of positions covered by X
    is infinite and will never decrease.

    Attention PISA.students. https://en.wikipedia.org/wiki/Programme_for_International_Student_Assessment

    Infinity minus infinity
    is not an operation.

    An operation needs
    no less than one result and
    no more than one result.
    1+1 is an operation because there is
    no less than one result (2) and
    no more than one result (nothing other than 2)

    Infinity minus infinity
    is not an operation because
    there _is_ more than one result,
    depending upon what is being asked.

    The matrix has X's in the first column
    and O's elsewhere
    infinite.matrix - infinite.column =
    infinite.elsewhere

    infinite.⟨0,1,…⟩ - infinite.⟨17,18,…⟩ =
    finite.⟨0,1,…,15,16⟩

    More than one result. Not an operation.

    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)
  • From Mikko@21:1/5 to All on Tue Dec 19 11:58:09 2023
    On 2023-12-18 10:52:59 +0000, WM said:

    Le 18/12/2023 à 10:44, Mikko a écrit :
    On 2023-12-17 19:36:59 +0000, WM said:

    The indeX transferred from matrix position (k, j) to matrix position (m, n)

    What does "transferred" mean?

    If (m, n) is to be indexed according to Cantors formula, then the indeX
    is taken from its position (k, j).

    How does that imply anything?

    The indices are distributet according to Cantor's formula
    k = (m + n - 1)(m + n - 2)/2 + m
    resulting in
    1/1, 1/2, 2/1, 1/3, 2/2, 3/1, 1/4, 2/3, 3/2, 4/1, 1/5, 2/4, 3/3, 4/2,
    5/1, 1/6, 2/5, 3/4, 4/3, 5/2, 6/1, ...
    but where the indices are removed, there are no indices.

    If any of the indices is removed you no longer have a matrix.

    Mikko

    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)
  • From WM@21:1/5 to All on Tue Dec 19 11:57:20 2023
    Le 18/12/2023 à 22:23, Jim Burns a écrit :

    Infinity minus infinity
    is not an operation.

    Sometimes it is an operation yielding the correct result. Compare Leibniz' summation of the infinite series 1/3 + 1/8 + 1/15 + ... = 3/4 where he subtracted two harmonic series.

    An operation needs
    no less than one result and
    no more than one result.

    Leibniz got exactly one result. My matrix without the first column gives exactly one result.
    Cantor gets many results, for instance in Hilbert's hotel.

    Regards, wM

    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)
  • From WM@21:1/5 to All on Tue Dec 19 12:00:46 2023
    Le 19/12/2023 à 10:58, Mikko a écrit :
    On 2023-12-18 10:52:59 +0000, WM said:

    Le 18/12/2023 à 10:44, Mikko a écrit :
    On 2023-12-17 19:36:59 +0000, WM said:

    The indeX transferred from matrix position (k, j) to matrix position (m, n)

    What does "transferred" mean?

    If (m, n) is to be indexed according to Cantors formula, then the indeX >>>> is taken from its position (k, j).

    How does that imply anything?

    Please try to understand by yourself in future. Or shut up.

    The indices are distributet according to Cantor's formula
    k = (m + n - 1)(m + n - 2)/2 + m
    resulting in
    1/1, 1/2, 2/1, 1/3, 2/2, 3/1, 1/4, 2/3, 3/2, 4/1, 1/5, 2/4, 3/3, 4/2,
    5/1, 1/6, 2/5, 3/4, 4/3, 5/2, 6/1, ...
    but where the indices are removed, there are no indices.

    If any of the indices is removed you no longer have a matrix.

    The matrix remains what it is.

    1/2 is indexed with 2 which is taken from its initial
    position 2/1

    XXOO...
    OOOO...
    XOOO...
    XOOO...
    XOOO...
    .

    Then index 3 is taken from its initial position 3/1 and is attached to 2/1

    XXOO...
    XOOO...
    OOOO...
    XOOO...
    XOOO...
    .

    Then index 4 is taken from its initial position 4/1 and is attached to 1/3

    XXXO...
    XOOO...
    OOOO...
    OOOO...
    XOOO...
    .

    Then index 5 is taken from its initial position 5/1 and is attached to 2/2

    XXXO...
    XXOO...
    OOOO...
    OOOO...
    OOOO...
    .

    The O will remain forever, indicating not indexed positions.

    Regards, WM

    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)
  • From WM@21:1/5 to All on Tue Dec 19 18:52:31 2023
    Le 18/12/2023 à 16:00, Richard Damon a écrit :
    On 12/18/23 5:47 AM, WM wrote:
    Le 18/12/2023 à 01:09, Richard Damon a écrit :
    On 12/17/23 4:18 PM, WM wrote:
    Le 17/12/2023 à 21:23, Richard Damon a écrit :

    Yes, in mathematics of infinite sets, the fact that one mapping
    misses even an infinite number of members means nothing,

    But here Cantor's mapping misses an infinite number of fractions.

    You aren't using the right mapping, showing your ignorance.

    Where does my mapping deviate from Cantor's
    k = (m + n - 1)(m + n - 2)/2 + m
    with the result
    1/1, 1/2, 2/1, 1/3, 2/2, 3/1, 1/4, 2/3, 3/2, 4/1, 1/5, 2/4, 3/3, 4/2,
    5/1, 1/6, 2/5, 3/4, 4/3, 5/2, 6/1,  ...
    for the first time?


    Which shows that your matrix B having values like

    1/1 2/1 3/1 ...
    1/2 2/2 3/2 ...
    1/3 2/3 3/3 ...
    .
    .
    .

    is exactly indexed by a matrix form like:

    Now it is the right mapping not showing ignorance?

    1 3 6 10
    2 5 9
    4 8 13
    7 12
    11

    Thus is a triangle, never leaving the shape of a triangle.


    Where the number at the m,n location of your fractions is indexed by the number at the m,n location of the Natural Numbers organized in that fashion.

    You seem to think that just because you can make a pattern that doesn't
    cover it (like trying to go all the way down first),

    By what is ithat forbidden?

    it means you can't
    cover it.

    It is impossible to remove an O (indicating a not indexed position) from
    the matrix.

    Regards, WM

    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)
  • From WM@21:1/5 to All on Tue Dec 19 18:54:50 2023
    Le 18/12/2023 à 17:06, Richard Damon a écrit :
    On 12/18/23 5:52 AM, WM wrote:

    The O will remain forever, indicating not indexed positions.

    And the problem is you aren't supposed to be taking the indexes out of
    B,

    What forbids to index the first column first? Or what forbids to use the integer fractions n/1 for indexing?


    Regards, WM

    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)
  • From Richard Damon@21:1/5 to All on Tue Dec 19 15:07:19 2023
    On 12/19/23 1:54 PM, WM wrote:
    Le 18/12/2023 à 17:06, Richard Damon a écrit :
    On 12/18/23 5:52 AM, WM wrote:

    The O will remain forever, indicating not indexed positions.

    And the problem is you aren't supposed to be taking the indexes out of B,

    What forbids to index the first column first? Or what forbids to use the integer fractions n/1 for indexing?


    Regards, WM



    Nothing "forbids" it, but the rule is that if ANY indexing works, then
    we have established that they are equal sizes.

    The fact that some attempts doesn't cover is irrelevant.

    You can't disprove that a mapping exists by showing that one mapping fails.

    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)
  • From WM@21:1/5 to All on Wed Dec 20 10:51:42 2023
    Le 19/12/2023 à 21:07, Richard Damon a écrit :
    On 12/19/23 1:54 PM, WM wrote:

    What forbids to index the first column first? Or what forbids to use the
    integer fractions n/1 for indexing?

    Nothing "forbids" it, but the rule is that if ANY indexing works, then
    we have established that they are equal sizes.

    First, this is a stupid rule. But second, my proof concerns the most
    famous and praised indexing and shows that this very mapping does not
    work.

    The fact that some attempts doesn't cover is irrelevant.

    You can't disprove that a mapping exists by showing that one mapping fails.

    I have shown that Cantor's mapping fails. And I can show that every other mapping will fail too.

    Regards, WM

    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)
  • From Mikko@21:1/5 to All on Wed Dec 20 14:05:36 2023
    On 2023-12-19 12:00:46 +0000, WM said:

    Le 19/12/2023 à 10:58, Mikko a écrit :
    On 2023-12-18 10:52:59 +0000, WM said:

    If any of the indices is removed you no longer have a matrix.

    The matrix remains what it is.

    You can't change the matrix so that it remanis as it was.
    If it remains as it was it is not changed.

    If you remove and index you change the matrix.
    If the matrix is not changed it still has the indices it had.

    Mikko

    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)
  • From Richard Damon@21:1/5 to All on Wed Dec 20 08:17:04 2023
    On 12/20/23 5:51 AM, WM wrote:
    Le 19/12/2023 à 21:07, Richard Damon a écrit :
    On 12/19/23 1:54 PM, WM wrote:

    What forbids to index the first column first? Or what forbids to use
    the integer fractions n/1 for indexing?

    Nothing "forbids" it, but the rule is that if ANY indexing works, then
    we have established that they are equal sizes.

    First, this is a stupid rule. But second, my proof concerns the most
    famous and praised indexing and shows that this very mapping does not work.


    Not "stupid", but necessary to make the system consistant.

    You are the stupid one for not understanding.

    Note, your indexing is NOT the indexing that Cantor was talking about,
    but you don't seem to understand enough to see that.

    The fact that some attempts doesn't cover is irrelevant.

    You can't disprove that a mapping exists by showing that one mapping
    fails.

    I have shown that Cantor's mapping fails. And I can show that every
    other mapping will fail too.

    Nope, you have shown that a mapping that has some surface simulatity to Cantor's fails.

    Cantor showed a mapping of k to m/n not of m/1 to m/n.

    Cantor maps the NATURAL NUMBERS (not the intergral rationals) to the
    rationals.


    Regards, WM



    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)
  • From WM@21:1/5 to All on Wed Dec 20 16:28:09 2023
    Le 20/12/2023 à 13:05, Mikko a écrit :
    On 2023-12-19 12:00:46 +0000, WM said:

    If any of the indices is removed you no longer have a matrix.

    The matrix remains what it is.

    You can't change the matrix so that it remanis as it was.
    If it remains as it was it is not changed.

    The matrix remains, only the covering by X changes.

    Regards, WM

    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)
  • From WM@21:1/5 to All on Wed Dec 20 17:44:59 2023
    Le 20/12/2023 à 14:17, Richard Damon a écrit :
    On 12/20/23 5:51 AM, WM wrote:

    I have shown that Cantor's mapping fails. And I can show that every
    other mapping will fail too.

    Nope, you have shown that a mapping that has some surface simulatity to Cantor's fails.

    What is the difference?

    Cantor showed a mapping of k to m/n not of m/1 to m/n.

    I first use the bijection between integers and integer fractions k <-->
    k/1. If that fails, then every mapping between infinite sets fails.

    Cantor maps the NATURAL NUMBERS (not the intergral rationals) to the rationals.

    So do I.

    Regards, WM

    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)
  • From immibis@21:1/5 to Richard Damon on Wed Dec 20 19:04:16 2023
    On 12/20/23 14:17, Richard Damon wrote:
    Cantor maps the NATURAL NUMBERS (not the intergral rationals) to the rationals.

    I am missing the start of this conversation. Aren't integral rationals
    exactly the integers? Are there any integers which are not rational?

    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)
  • From Richard Damon@21:1/5 to All on Wed Dec 20 17:25:10 2023
    On 12/20/23 12:44 PM, WM wrote:
    Le 20/12/2023 à 14:17, Richard Damon a écrit :
    On 12/20/23 5:51 AM, WM wrote:

    I have shown that Cantor's mapping fails. And I can show that every
    other mapping will fail too.

    Nope, you have shown that a mapping that has some surface simulatity
    to Cantor's fails.

    What is the difference?

    Cantor showed a mapping of k to m/n not of m/1 to m/n.

    I first use the bijection between integers and integer fractions k <-->
    k/1. If that fails, then every mapping between infinite sets fails.

    Cantor maps the NATURAL NUMBERS (not the intergral rationals) to the
    rationals.

    So do I.

    Regards, WM



    Nope, because the matrix B is filled with ratios, not just Natural Numbers.

    You talk about values left behind, but those numbers aren't Natural Numbers.

    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)
  • From Richard Damon@21:1/5 to immibis on Wed Dec 20 18:40:08 2023
    On 12/20/23 1:04 PM, immibis wrote:
    On 12/20/23 14:17, Richard Damon wrote:
    Cantor maps the NATURAL NUMBERS (not the intergral rationals) to the
    rationals.

    I am missing the start of this conversation. Aren't integral rationals exactly the integers? Are there any integers which are not rational?

    The difference is in the sets that they are embedded in.

    He seems to be matching the m/n element with the k/1 element in the
    first column, and noting that he has all the other columns that he
    hasn't matched to, so he is looking at the k/1 elements not a Natural
    Numbers but the Rationals that happen to be Natural Numbers.

    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)
  • From Mikko@21:1/5 to All on Thu Dec 21 11:41:41 2023
    On 2023-12-20 16:28:09 +0000, WM said:

    Le 20/12/2023 à 13:05, Mikko a écrit :
    On 2023-12-19 12:00:46 +0000, WM said:

    If any of the indices is removed you no longer have a matrix.

    The matrix remains what it is.

    You can't change the matrix so that it remanis as it was.
    If it remains as it was it is not changed.

    The matrix remains, only the covering by X changes.

    Your first "covering" is just another matrix with the same idicex.
    When it "changes" it does not really change, there just is another
    "covering", and after successive changes a sequence of such "coverings".

    Mikko

    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)
  • From WM@21:1/5 to All on Thu Dec 21 11:05:17 2023
    Le 20/12/2023 à 19:04, immibis a écrit :
    On 12/20/23 14:17, Richard Damon wrote:
    Cantor maps the NATURAL NUMBERS (not the intergral rationals) to the
    rationals.

    I am missing the start of this conversation. Aren't integral rationals exactly the integers?

    Of course. Therefore it is irrelevant whether we start with the first
    column of

    1/1, 1/2, 1/3, 1/4, ...
    2/1, 2/2, 2/3, 2/4, ...
    3/1, 3/2, 3/3, 3/4, ...
    4/1, 4/2, 4/3, 4/4, ...
    5/1, 5/2, 5/3, 5/4, ...
    ..

    or first biject the integer fractions with the natural numbers

    1, 1/2, 1/3, 1/4, ...
    2, 2/2, 2/3, 2/4, ...
    3, 3/2, 3/3, 3/4, ...
    4, 4/2, 4/3, 4/4, ...
    5, 5/2, 5/3, 5/4, ...
    ..

    In no case it is possible to cover the whole matrix by indeXes

    XOOO...
    XOOO...
    XOOO...
    XOOO...
    ..

    Regards, WM

    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)
  • From WM@21:1/5 to All on Thu Dec 21 10:57:10 2023
    Le 21/12/2023 à 10:41, Mikko a écrit :
    On 2023-12-20 16:28:09 +0000, WM said:

    The matrix remains, only the covering by X changes.

    Your first "covering" is just another matrix with the same idicex.

    It has the same (m, n), but not all positions have been counted by k
    k = (m + n - 1)(m + n - 2)/2 + m.

    XOOO...
    XOOO...
    XOOO...
    XOOO...
    ..

    When it "changes" it does not really change, there just is another "covering", and after successive changes a sequence of such "coverings".

    And none covers the whole matrix by indeXes.

    Regards, WM

    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)
  • From Mikko@21:1/5 to All on Thu Dec 21 13:48:23 2023
    On 2023-12-21 10:57:10 +0000, WM said:

    Le 21/12/2023 à 10:41, Mikko a écrit :
    On 2023-12-20 16:28:09 +0000, WM said:

    The matrix remains, only the covering by X changes.

    Your first "covering" is just another matrix with the same idicex.

    It has the same (m, n), but not all positions have been counted by k
    k = (m + n - 1)(m + n - 2)/2 + m.

    Positions are not covered by. Some of them are covered by X.

    Each position as m and n so each position has k.

    Mikko

    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)
  • From Ganzhinterseher@21:1/5 to FredJeffries on Thu Dec 21 12:53:11 2023
    On 21.12.2023 01:10, FredJeffries wrote:
    Cantor described a function from the SET of natural numbers and the SET of rational numbers,

    This function is a sequence described by
    k = (m + n - 1)(m + n - 2)/2 + m.
    The same that I use.

    whereas OBP 'constructs' a few 'mappings',

    What index k of Cantor's do I miss to apply?

    each between some particular natural number and some rational number.

    Precisely like Cantor, and with the same result: 1/1, 1/2, 2/1, 1/3,
    2/2, 3/1, 1/4, 2/3, 3/2, 4/1, 1/5, 2/4, 3/3, 4/2, 5/1, 1/6, 2/5, 3/4,
    4/3, 5/2, 6/1, ...

    Regards, WM

    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)
  • From WM@21:1/5 to All on Thu Dec 21 11:56:10 2023
    Le 21/12/2023 à 12:48, Mikko a écrit :
    On 2023-12-21 10:57:10 +0000, WM said:

    Le 21/12/2023 à 10:41, Mikko a écrit :
    On 2023-12-20 16:28:09 +0000, WM said:

    The matrix remains, only the covering by X changes.

    Your first "covering" is just another matrix with the same idicex.

    It has the same (m, n), but not all positions have been counted by k
    k = (m + n - 1)(m + n - 2)/2 + m.

    Positions are not covered by. Some of them are covered by X.

    Each position as m and n so each position has k.

    That is an error. Each O remains in the matrix at a not indexed place.

    Regards, WM

    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)
  • From Richard Damon@21:1/5 to All on Thu Dec 21 10:18:46 2023
    On 12/21/23 5:57 AM, WM wrote:
    Le 21/12/2023 à 10:41, Mikko a écrit :
    On 2023-12-20 16:28:09 +0000, WM said:

    The matrix remains, only the covering by X changes.

    Your first "covering" is just another matrix with the same idicex.

    It has the same (m, n), but not all positions have been counted by k
    k = (m + n - 1)(m + n - 2)/2 + m.

    XOOO...
    XOOO...
    XOOO...
    XOOO...
    ..

    When it "changes" it does not really change, there just is another
    "covering", and after successive changes a sequence of such "coverings".

    And none covers the whole matrix by indeXes.

    Regards, WM


    So, you just don't understand what the bijection is doing.

    You aren't indexing the values of k to the matrix, but the index
    locations m,n

    You are just doing it backwards, as seems to be your nature,

    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)
  • From Richard Damon@21:1/5 to All on Thu Dec 21 10:16:18 2023
    On 12/21/23 6:05 AM, WM wrote:
    Le 20/12/2023 à 19:04, immibis a écrit :
    On 12/20/23 14:17, Richard Damon wrote:
    Cantor maps the NATURAL NUMBERS (not the intergral rationals) to the
    rationals.

    I am missing the start of this conversation. Aren't integral rationals
    exactly the integers?

    Of course. Therefore it is irrelevant whether we start with the first
    column of

    1/1, 1/2, 1/3, 1/4, ...
    2/1, 2/2, 2/3, 2/4, ...
    3/1, 3/2, 3/3, 3/4, ...
    4/1, 4/2, 4/3, 4/4, ...
    5/1, 5/2, 5/3, 5/4, ...
    ..

    or first biject the integer fractions with the natural numbers

    1, 1/2, 1/3, 1/4, ...
    2, 2/2, 2/3, 2/4, ...
    3, 3/2, 3/3, 3/4, ...
    4, 4/2, 4/3, 4/4, ...
    5, 5/2, 5/3, 5/4, ...
    ..

    So, what are you trying to do here? This isn't a matrix of any normal
    domain.



    In no case it is possible to cover the whole matrix by indeXes

    XOOO...
    XOOO...
    XOOO...
    XOOO...
    ..

    Regards, WM



    Your arguement doesn't make sense.

    IF you are trying to say that there isn't a mapping for the values in
    the first column to the whole set, you are just wrong and the mapping
    has been given.

    If you are arguing that the shape of the first column doesn't ever cover
    the matrix, that is just an irrelevent stupidity, as no one says it should.

    Cantor said there was a bijection between the integers and the
    rationals, not that all rationals were integers.

    Your inability to understand the difference shows your lack of
    understanding and unsound reasoning.


    We cover the matrix:

    1/1, 1/2, 1/3, 1/4, ...
    2/1, 2/2, 2/3, 2/4, ...
    3/1, 3/2, 3/3, 3/4, ...
    4/1, 4/2, 4/3, 4/4, ...
    ...

    with values mappes as following:

    1 2 4 7 11 16 22
    3 5 8 12 17 23
    6 9 13 18 24
    10 14 19 25
    15 20 26
    21 26
    27


    Note, all of these *VALUES* can be found in the first column of the
    rationals, but that is irrelevant, as we are showing a mapping of
    indexes from the Natural Numbers to rational matrix.

    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)
  • From Mikko@21:1/5 to All on Thu Dec 21 18:45:40 2023
    On 2023-12-21 11:56:10 +0000, WM said:

    Le 21/12/2023 à 12:48, Mikko a écrit :
    On 2023-12-21 10:57:10 +0000, WM said:

    Le 21/12/2023 à 10:41, Mikko a écrit :
    On 2023-12-20 16:28:09 +0000, WM said:

    The matrix remains, only the covering by X changes.

    Your first "covering" is just another matrix with the same idicex.

    It has the same (m, n), but not all positions have been counted by k
    k = (m + n - 1)(m + n - 2)/2 + m.

    Positions are not covered by. Some of them are covered by X.

    Each position as m and n so each position has k.

    That is an error. Each O remains in the matrix at a not indexed place.

    By the definition of "matrix" there are only indexed places.
    A structure that contains unindexec places is not a matrix.

    Mikko

    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)
  • From Jim Burns@21:1/5 to All on Thu Dec 21 13:18:29 2023
    On 12/20/2023 5:51 AM, WM wrote:
    Le 19/12/2023 à 21:07, Richard Damon a écrit :
    On 12/19/23 1:54 PM, WM wrote:

    What forbids
    to index the first column first?
    Or what forbids
    to use the integer fractions n/1 for indexing?

    Nothing "forbids" it,
    but the rule is that
    if ANY indexing works,
    then we have established that
    they are equal sizes.

    First, this is a stupid rule.

    The rule is the same for finite and infinite sets.

    If,
    for each b ∈ B
    b has its own aᵇ ∈ A
    then
    A holds at least as many as B
    |B| ≤ |A|

    What _isn't_ a rule, and
    which would be a rule for only sets like
    flocks of sheep or bags of pebbles,
    is that
    if
    1.to.1 f: B ↣ A
    1.to.1 g: B ↣ A
    then
    f(B) = A and g(B) = A
    or
    f(B) ≠ A and g(B) ≠ A

    What you (WM) are probably calling stupid
    isn't so much a _rule_ rule
    as it is the _observation_ that
    the flock-of-sheep rule isn't correct
    for all sets.

    For example.

    The flock-of-sheep rule is correct for
    each ⟨0,…,i⟩, each ⟨1,…,j+1⟩

    Define 1.to.1 f(k) = k-1

    if i=j
    f⟨1,…,j+1⟩ = ⟨0,…,i⟩
    and
    for each 1.to.1 g: ⟨1,…,j+1⟩ ↣ ⟨0,…,i⟩
    g⟨1,…,j+1⟩ = ⟨0,…,i⟩

    if i≠j
    f⟨1,…,j+1⟩ ≠ ⟨0,…,i⟩
    and
    for each 1.to.1 g: ⟨1,…,j+1⟩ ↣ ⟨0,…,i⟩
    g⟨1,…,j+1⟩ ≠ ⟨0,…,i⟩

    However,
    we observe that
    the flock-of-sheep rule does not work
    for ᴸᵁᴮ⟨⟨0,…,i⟩⟩ ᴸᵁᴮ⟨⟨1,…,j+1⟩⟩
    least upper bounds
    of all ⟨0,…,i⟩ and of all ⟨1,…,j+1⟩

    for 1.to.1 f(k) = k-1
    fᴸᵁᴮ⟨⟨1,…,j+1⟩⟩ = ᴸᵁᴮ⟨⟨0,…,i⟩⟩

    for 1.to.1 g(k) = k
    gᴸᵁᴮ⟨⟨1,…,j+1⟩⟩ = ᴸᵁᴮ⟨⟨1,…,j+1⟩⟩ ᴸᵁᴮ⟨⟨1,…,j+1⟩⟩ ≠ ᴸᵁᴮ⟨⟨0,…,i⟩⟩

    ᴸᵁᴮ⟨⟨1,…,j+1⟩⟩ is
    a proper subset of ᴸᵁᴮ⟨⟨0,…,i⟩⟩
    and
    ᴸᵁᴮ⟨⟨1,…,j+1⟩⟩ is
    the same size as ᴸᵁᴮ⟨⟨0,…,i⟩⟩

    We observe that
    not all sets are flocks of sheep.

    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)
  • From WM@21:1/5 to All on Fri Dec 22 08:04:21 2023
    Le 21/12/2023 à 16:16, Richard Damon a écrit :
    On 12/21/23 6:05 AM, WM wrote:
    Le 20/12/2023 à 19:04, immibis a écrit :
    On 12/20/23 14:17, Richard Damon wrote:
    Cantor maps the NATURAL NUMBERS (not the intergral rationals) to the
    rationals.

    I am missing the start of this conversation. Aren't integral rationals
    exactly the integers?

    Of course. Therefore it is irrelevant whether we start with the first
    column of

    1/1, 1/2, 1/3, 1/4, ...
    2/1, 2/2, 2/3, 2/4, ...
    3/1, 3/2, 3/3, 3/4, ...
    4/1, 4/2, 4/3, 4/4, ...
    5/1, 5/2, 5/3, 5/4, ...
    ..

    or first biject the integer fractions with the natural numbers

    1, 1/2, 1/3, 1/4, ...
    2, 2/2, 2/3, 2/4, ...
    3, 3/2, 3/3, 3/4, ...
    4, 4/2, 4/3, 4/4, ...
    5, 5/2, 5/3, 5/4, ...
    ..

    In no case it is possible to cover the whole matrix by indeXes

    XOOO...
    XOOO...
    XOOO...
    XOOO...
    ..

    IF you are trying to say that there isn't a mapping for the values in
    the first column to the whole set, you are just wrong and the mapping
    has been given.

    It appears so, but the O indicate not indeXed fractions. This has to be
    taken into account.

    If you are arguing that the shape of the first column doesn't ever cover
    the matrix, that is just an irrelevent stupidity, as no one says it should.

    That is Cantor's claim. All indexes have to be appßlied. They can be
    taken from the first column.


    We cover the matrix:

    1/1, 1/2, 1/3, 1/4, ...
    2/1, 2/2, 2/3, 2/4, ...
    3/1, 3/2, 3/3, 3/4, ...
    4/1, 4/2, 4/3, 4/4, ...
    ...

    with values mappes as following:

    1 2 4 7 11 16 22
    3 5 8 12 17 23
    6 9 13 18 24
    10 14 19 25
    15 20 26
    21 26
    27


    Note, all of these *VALUES* can be found in the first column of the rationals, but that is irrelevant, as we are showing a mapping of
    indexes from the Natural Numbers to rational matrix.

    You are mistaken. All your indeXes up to every indeX applied in your
    mapping belong to a finite set {1, 2, 3, ..., X} which has ℵo
    successors. My proof shows that the O, indicating not indexed fractions,
    will remain forever within the matrix, being even the majority.


    Regards, WM

    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)
  • From WM@21:1/5 to All on Fri Dec 22 08:09:14 2023
    Le 21/12/2023 à 17:45, Mikko a écrit :
    On 2023-12-21 11:56:10 +0000, WM said:

    Le 21/12/2023 à 12:48, Mikko a écrit :
    On 2023-12-21 10:57:10 +0000, WM said:

    Le 21/12/2023 à 10:41, Mikko a écrit :
    On 2023-12-20 16:28:09 +0000, WM said:

    The matrix remains, only the covering by X changes.

    Your first "covering" is just another matrix with the same idicex.

    It has the same (m, n), but not all positions have been counted by k
    k = (m + n - 1)(m + n - 2)/2 + m.

    Positions are not covered by. Some of them are covered by X.

    Each position as m and n so each position has k.

    That is an error. Each O remains in the matrix at a not indexed place.

    By the definition of "matrix" there are only indexed places.

    But not indexed by Cantor's k = (m + n - 1)(m + n - 2)/2 + m.

    Regards, WM

    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)
  • From WM@21:1/5 to All on Fri Dec 22 08:07:55 2023
    Le 21/12/2023 à 16:18, Richard Damon a écrit :
    On 12/21/23 5:57 AM, WM wrote:
    Le 21/12/2023 à 10:41, Mikko a écrit :
    On 2023-12-20 16:28:09 +0000, WM said:

    The matrix remains, only the covering by X changes.

    Your first "covering" is just another matrix with the same idicex.

    It has the same (m, n), but not all positions have been counted by k
    k = (m + n - 1)(m + n - 2)/2 + m.

    XOOO...
    XOOO...
    XOOO...
    XOOO...
    ..

    When it "changes" it does not really change, there just is another
    "covering", and after successive changes a sequence of such "coverings".

    And none covers the whole matrix by indeXes.

    So, you just don't understand what the bijection is doing.

    A bijection should index every element of the matrix, but it does not,
    because the O remain. Therefore it is not a bijection.

    Regards, WM

    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)
  • From WM@21:1/5 to All on Fri Dec 22 08:13:05 2023
    Le 21/12/2023 à 19:18, Jim Burns a écrit :
    On 12/20/2023 5:51 AM, WM wrote:
    Le 19/12/2023 à 21:07, Richard Damon a écrit :
    On 12/19/23 1:54 PM, WM wrote:

    What forbids
    to index the first column first?
    Or what forbids
    to use the integer fractions n/1 for indexing?

    Nothing "forbids" it,
    but the rule is that
    if ANY indexing works,
    then we have established that
    they are equal sizes.

    First, this is a stupid rule.

    The rule is the same for finite and infinite sets.

    No. For finite sets of equal numerosity any injection is a surjection and
    any surjection is an injection.

    What you (WM) are probably calling stupid
    isn't so much a _rule_ rule
    as it is the _observation_ that
    the flock-of-sheep rule isn't correct
    for all sets.

    That is not an observation but an assumption of people who deny or are
    unable to understand actual infinity. Bob will never leave the matrix, Therefore your arguments are invalid and worthless.

    Regards, WM

    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)
  • From Richard Damon@21:1/5 to All on Fri Dec 22 09:05:05 2023
    On 12/22/23 3:04 AM, WM wrote:
    Le 21/12/2023 à 16:16, Richard Damon a écrit :
    On 12/21/23 6:05 AM, WM wrote:
    Le 20/12/2023 à 19:04, immibis a écrit :
    On 12/20/23 14:17, Richard Damon wrote:
    Cantor maps the NATURAL NUMBERS (not the intergral rationals) to
    the rationals.

    I am missing the start of this conversation. Aren't integral
    rationals exactly the integers?

    Of course. Therefore it is irrelevant whether we start with the first
    column of

    1/1, 1/2, 1/3, 1/4, ...
    2/1, 2/2, 2/3, 2/4, ...
    3/1, 3/2, 3/3, 3/4, ...
    4/1, 4/2, 4/3, 4/4, ...
    5/1, 5/2, 5/3, 5/4, ...
    ..

    or first biject the integer fractions with the natural numbers

    1, 1/2, 1/3, 1/4, ...
    2, 2/2, 2/3, 2/4, ...
    3, 3/2, 3/3, 3/4, ...
    4, 4/2, 4/3, 4/4, ...
    5, 5/2, 5/3, 5/4, ...
    ..

    In no case it is possible to cover the whole matrix by indeXes

    XOOO...
    XOOO...
    XOOO...
    XOOO...
    ..

    IF you are trying to say that there isn't a mapping for the values in
    the first column to the whole set, you are just wrong and the mapping
    has been given.

    It appears so, but the O indicate not indeXed fractions. This has to be
    taken into account.

    No, you are confusing the index with the indexed.

    the second x down (2) indexes the second element of the first row (1/2)
    which you have marked with an O.


    If you are arguing that the shape of the first column doesn't ever
    cover the matrix, that is just an irrelevent stupidity, as no one says
    it should.

    That is Cantor's claim. All indexes have to be appßlied. They can be
    taken from the first column.

    Right, and by the transformation of the index, they select all the
    element of the matrix.



    We cover the matrix:

    1/1, 1/2, 1/3, 1/4, ...
    2/1, 2/2, 2/3, 2/4, ...
    3/1, 3/2, 3/3, 3/4, ...
    4/1, 4/2, 4/3, 4/4, ...
    ...

    with values mappes as following:

    1    2     4    7   11  16  22
    3    5     8    12  17  23
    6    9     13   18  24
    10   14    19   25
    15   20    26
    21   26
    27


    Note, all of these *VALUES* can be found in the first column of the
    rationals, but that is irrelevant, as we are showing a mapping of
    indexes from the Natural Numbers to rational matrix.

    You are mistaken. All your indeXes up to every indeX applied in your
    mapping belong to a finite set {1, 2, 3, ..., X} which has ℵo
    successors. My proof shows that the O, indicating not indexed fractions,
    will remain forever within the matrix, being even the majority.

    No, you are showing you can create a non-covering mapping, but the rule
    says that if at least ONE covering mapping exists, we meet the requirement.

    You can't prove non-existence by showing an example of something that isn't.

    In other words, you just don't understand what Cantor is talking about,
    because your understanding of the infinite is incomplete.



    Regards, WM

    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)
  • From Richard Damon@21:1/5 to All on Fri Dec 22 09:08:30 2023
    On 12/22/23 3:07 AM, WM wrote:
    Le 21/12/2023 à 16:18, Richard Damon a écrit :
    On 12/21/23 5:57 AM, WM wrote:
    Le 21/12/2023 à 10:41, Mikko a écrit :
    On 2023-12-20 16:28:09 +0000, WM said:

    The matrix remains, only the covering by X changes.

    Your first "covering" is just another matrix with the same idicex.

    It has the same (m, n), but not all positions have been counted by k
    k = (m + n - 1)(m + n - 2)/2 + m.

    XOOO...
    XOOO...
    XOOO...
    XOOO...
    ..

    When it "changes" it does not really change, there just is another
    "covering", and after successive changes a sequence of such
    "coverings".

    And none covers the whole matrix by indeXes.

    So, you just don't understand what the bijection is doing.

    A bijection should index every element of the matrix, but it does not, because the O remain. Therefore it is not a bijection.

    Regards, WM

    So, you are showning that the bijecton for k to k/1 doesn't cover all of
    m/n, but that doesn't show that the bijecton for k to m/n where
    k = (m + n - 1)(m + n - 2)/2 + m doesn't work.

    Since the requirement is only that there eists *A* bijection that works,
    you haven't proven your point.

    You are trying to prove non-existence by example, which just doesn't work.

    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)
  • From Mikko@21:1/5 to All on Fri Dec 22 16:19:33 2023
    On 2023-12-22 08:09:14 +0000, WM said:

    Le 21/12/2023 à 17:45, Mikko a écrit :
    On 2023-12-21 11:56:10 +0000, WM said:

    Le 21/12/2023 à 12:48, Mikko a écrit :
    On 2023-12-21 10:57:10 +0000, WM said:

    Le 21/12/2023 à 10:41, Mikko a écrit :
    On 2023-12-20 16:28:09 +0000, WM said:

    The matrix remains, only the covering by X changes.

    Your first "covering" is just another matrix with the same idicex.

    It has the same (m, n), but not all positions have been counted by k >>>>> k = (m + n - 1)(m + n - 2)/2 + m.

    Positions are not covered by. Some of them are covered by X.

    Each position as m and n so each position has k.

    That is an error. Each O remains in the matrix at a not indexed place.

    By the definition of "matrix" there are only indexed places.

    But not indexed by Cantor's k = (m + n - 1)(m + n - 2)/2 + m.

    It is indexed by two natural numbers because the matrices meintioned
    above and earlier in this discussion are matrices that are indexed by
    two natural numbers. You have called these two numbers m and n.
    For each natural number k there is only one pair of natural numbers
    m and n so that k = (m + n - 1)(m + n - 2)/2 + m. Therefore k is
    sufficient to identify a position in any of the matrices. But it does
    not identify which matrix one is talking about so that must be told
    separately.

    Mikko

    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)
  • From immibis@21:1/5 to All on Fri Dec 22 16:14:59 2023
    On 12/22/23 09:09, WM wrote:
    Le 21/12/2023 à 17:45, Mikko a écrit :
    On 2023-12-21 11:56:10 +0000, WM said:

    Le 21/12/2023 à 12:48, Mikko a écrit :
    On 2023-12-21 10:57:10 +0000, WM said:

    Le 21/12/2023 à 10:41, Mikko a écrit :
    On 2023-12-20 16:28:09 +0000, WM said:

    The matrix remains, only the covering by X changes.

    Your first "covering" is just another matrix with the same idicex.

    It has the same (m, n), but not all positions have been counted by k >>>>> k = (m + n - 1)(m + n - 2)/2 + m.

    Positions are not covered by. Some of them are covered by X.

    Each position as m and n so each position has k.

    That is an error. Each O remains in the matrix at a not indexed place.

    By the definition of "matrix" there are only indexed places.

    But not indexed by Cantor's k = (m + n - 1)(m + n - 2)/2 + m.

    Regards, WM

    All of them are indexed by Cantor

    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)
  • From WM@21:1/5 to All on Fri Dec 22 16:15:19 2023
    Le 22/12/2023 à 16:15, immibis a écrit :
    On 12/22/23 09:13, WM wrote:
    Le 21/12/2023 à 19:18, Jim Burns a écrit :
    On 12/20/2023 5:51 AM, WM wrote:
    Le 19/12/2023 à 21:07, Richard Damon a écrit :
    On 12/19/23 1:54 PM, WM wrote:

    What forbids
    to index the first column first?
    Or what forbids
    to use the integer fractions n/1 for indexing?

    Nothing "forbids" it,
    but the rule is that
    if ANY indexing works,
    then we have established that
    they are equal sizes.

    First, this is a stupid rule.

    The rule is the same for finite and infinite sets.

    No. For finite sets of equal numerosity any injection is a surjection
    and any surjection is an injection.

    Wrong rule. If any injection between two sets exists, they have equal sizes.

    Yes, that is awrong rule. See the injection of natnumbers into the reals.

    Regards, WM

    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)
  • From immibis@21:1/5 to All on Fri Dec 22 16:15:58 2023
    On 12/22/23 09:13, WM wrote:
    Le 21/12/2023 à 19:18, Jim Burns a écrit :
    On 12/20/2023 5:51 AM, WM wrote:
    Le 19/12/2023 à 21:07, Richard Damon a écrit :
    On 12/19/23 1:54 PM, WM wrote:

    What forbids
    to index the first column first?
    Or what forbids
    to use the integer fractions n/1 for indexing?

    Nothing "forbids" it,
    but the rule is that
    if ANY indexing works,
    then we have established that
    they are equal sizes.

    First, this is a stupid rule.

    The rule is the same for finite and infinite sets.

    No. For finite sets of equal numerosity any injection is a surjection
    and any surjection is an injection.

    Wrong rule. If any injection between two sets exists, they have equal sizes.

    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)
  • From WM@21:1/5 to All on Fri Dec 22 16:16:22 2023
    Le 22/12/2023 à 16:15, immibis a écrit :
    On 12/22/23 09:09, WM wrote:
    Le 21/12/2023 à 17:45, Mikko a écrit :
    On 2023-12-21 11:56:10 +0000, WM said:

    Le 21/12/2023 à 12:48, Mikko a écrit :
    On 2023-12-21 10:57:10 +0000, WM said:

    Le 21/12/2023 à 10:41, Mikko a écrit :
    On 2023-12-20 16:28:09 +0000, WM said:

    The matrix remains, only the covering by X changes.

    Your first "covering" is just another matrix with the same idicex. >>>>>>
    It has the same (m, n), but not all positions have been counted by k >>>>>> k = (m + n - 1)(m + n - 2)/2 + m.

    Positions are not covered by. Some of them are covered by X.

    Each position as m and n so each position has k.

    That is an error. Each O remains in the matrix at a not indexed place.

    By the definition of "matrix" there are only indexed places.

    But not indexed by Cantor's k = (m + n - 1)(m + n - 2)/2 + m.

    All of them are indexed by Cantor

    The remaining not indexed fractions, covered by O, prove the contrary.

    Regards, WM

    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)
  • From WM@21:1/5 to All on Fri Dec 22 16:23:40 2023
    Le 22/12/2023 à 15:05, Richard Damon a écrit :

    No, you are showing you can create a non-covering mapping,

    I show that Cantor's mapping is not covering the matrix.

    but the rule
    says that if at least ONE covering mapping exists, we meet the requirement.

    Who made that rule?

    Regards, WM

    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)
  • From WM@21:1/5 to All on Fri Dec 22 16:19:08 2023
    Le 22/12/2023 à 15:19, Mikko a écrit :
    On 2023-12-22 08:09:14 +0000, WM said:

    Le 21/12/2023 à 17:45, Mikko a écrit :
    On 2023-12-21 11:56:10 +0000, WM said:

    Le 21/12/2023 à 12:48, Mikko a écrit :
    On 2023-12-21 10:57:10 +0000, WM said:

    Le 21/12/2023 à 10:41, Mikko a écrit :
    On 2023-12-20 16:28:09 +0000, WM said:

    The matrix remains, only the covering by X changes.

    Your first "covering" is just another matrix with the same idicex. >>>>>>
    It has the same (m, n), but not all positions have been counted by k >>>>>> k = (m + n - 1)(m + n - 2)/2 + m.

    Positions are not covered by. Some of them are covered by X.

    Each position as m and n so each position has k.

    That is an error. Each O remains in the matrix at a not indexed place.

    By the definition of "matrix" there are only indexed places.

    But not indexed by Cantor's k = (m + n - 1)(m + n - 2)/2 + m.

    It is indexed by two natural numbers because the matrices meintioned
    above and earlier in this discussion are matrices that are indexed by
    two natural numbers.

    As the O, not indexed by Cantor prove, most of the matrix' (m, n) are
    dark.
    Only the upper left corner contains visible m and n.

    You have called these two numbers m and n.
    For each natural number k there is only one pair of natural numbers
    m and n so that k = (m + n - 1)(m + n - 2)/2 + m. Therefore k is
    sufficient to identify a position in any of the matrices.

    The remaining O prove the contrary.

    Regards, WM

    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)
  • From WM@21:1/5 to All on Fri Dec 22 16:25:57 2023
    Le 22/12/2023 à 15:08, Richard Damon a écrit :
    On 12/22/23 3:07 AM, WM wrote:
    Le 21/12/2023 à 16:18, Richard Damon a écrit :
    On 12/21/23 5:57 AM, WM wrote:
    Le 21/12/2023 à 10:41, Mikko a écrit :
    On 2023-12-20 16:28:09 +0000, WM said:

    The matrix remains, only the covering by X changes.

    Your first "covering" is just another matrix with the same idicex.

    It has the same (m, n), but not all positions have been counted by k
    k = (m + n - 1)(m + n - 2)/2 + m.

    XOOO...
    XOOO...
    XOOO...
    XOOO...
    ..

    When it "changes" it does not really change, there just is another
    "covering", and after successive changes a sequence of such
    "coverings".

    And none covers the whole matrix by indeXes.

    So, you just don't understand what the bijection is doing.

    A bijection should index every element of the matrix, but it does not,
    because the O remain. Therefore it is not a bijection.

    So, you are showning that the bijecton for k to k/1 doesn't cover all of
    m/n, but that doesn't show that the bijecton for k to m/n where
    k = (m + n - 1)(m + n - 2)/2 + m doesn't work.

    If the first bijection k to k/1 is a bijection, then I apply in the second stepCantor's procedure. If not, then there are no infinite bijections at
    all.

    Regards, WM

    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)
  • From Richard Damon@21:1/5 to All on Fri Dec 22 11:59:01 2023
    On 12/22/23 11:23 AM, WM wrote:
    Le 22/12/2023 à 15:05, Richard Damon a écrit :

    No, you are showing you can create a non-covering mapping,

    I show that Cantor's mapping is not covering the matrix.

    No, you didn't, because you didn't do Cantor's mapping.


    but the rule says that if at least ONE covering mapping exists, we
    meet the requirement.

    Who made that rule?

    It wasn't "Made" but "Discovered". It is a natural consequence of the
    need for a consistent set of logical rules to describe infinite set.

    If you are willing for your logic to be in a system that creates
    inconsistent results, and thus worthless for establishing the actual
    truth of statement, go ahead and ignore it.

    Of course, the fact that you even ask the question says you do not
    understand what Cantor does, and thus your arguments are based on your ignorance.


    Regards, WM

    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)
  • From Richard Damon@21:1/5 to All on Fri Dec 22 12:04:01 2023
    On 12/22/23 11:15 AM, WM wrote:
    Le 22/12/2023 à 16:15, immibis a écrit :
    On 12/22/23 09:13, WM wrote:
    Le 21/12/2023 à 19:18, Jim Burns a écrit :
    On 12/20/2023 5:51 AM, WM wrote:
    Le 19/12/2023 à 21:07, Richard Damon a écrit :
    On 12/19/23 1:54 PM, WM wrote:

    What forbids
    to index the first column first?
    Or what forbids
    to use the integer fractions n/1 for indexing?

    Nothing "forbids" it,
    but the rule is that
    if ANY indexing works,
    then we have established that
    they are equal sizes.

    First, this is a stupid rule.

    The rule is the same for finite and infinite sets.

    No. For finite sets of equal numerosity any injection is a surjection
    and any surjection is an injection.

    Wrong rule. If any injection between two sets exists, they have equal
    sizes.

    Yes, that is awrong rule. See the injection of natnumbers into the reals.

    Regards, WM



    But the actual rule is either have a bijection, or an injection both ways.

    You can not inject the Reals to the Natural Numbers as you run out of
    Natural Numbers in all attempts to do so.

    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)
  • From Richard Damon@21:1/5 to All on Fri Dec 22 12:05:07 2023
    On 12/22/23 11:19 AM, WM wrote:
    Le 22/12/2023 à 15:19, Mikko a écrit :
    On 2023-12-22 08:09:14 +0000, WM said:

    Le 21/12/2023 à 17:45, Mikko a écrit :
    On 2023-12-21 11:56:10 +0000, WM said:

    Le 21/12/2023 à 12:48, Mikko a écrit :
    On 2023-12-21 10:57:10 +0000, WM said:

    Le 21/12/2023 à 10:41, Mikko a écrit :
    On 2023-12-20 16:28:09 +0000, WM said:

    The matrix remains, only the covering by X changes.

    Your first "covering" is just another matrix with the same idicex. >>>>>>>
    It has the same (m, n), but not all positions have been counted by k >>>>>>> k = (m + n - 1)(m + n - 2)/2 + m.

    Positions are not covered by. Some of them are covered by X.

    Each position as m and n so each position has k.

    That is an error. Each O remains in the matrix at a not indexed place. >>>>
    By the definition of "matrix" there are only indexed places.

    But not indexed by Cantor's k = (m + n - 1)(m + n - 2)/2 + m.

    It is indexed by two natural numbers because the matrices meintioned
    above and earlier in this discussion are matrices that are indexed by
    two natural numbers.

    As the O, not indexed by Cantor prove, most of the matrix' (m, n) are dark. Only the upper left corner contains visible m and n.

    Except the matrix with the O's isn't Cantor injection in the rationals,
    so doesn't prove anything but that you like to play with Strawmen, or
    you just don't understand Cantor.


    You have called these two numbers m and n.
    For each natural number k there is only one pair of natural numbers
    m and n so that k = (m + n - 1)(m + n - 2)/2 + m. Therefore k is
    sufficient to identify a position in any of the matrices.

    The remaining O prove the contrary.

    Regards, WM



    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)
  • From Richard Damon@21:1/5 to All on Fri Dec 22 12:08:03 2023
    On 12/22/23 11:25 AM, WM wrote:
    Le 22/12/2023 à 15:08, Richard Damon a écrit :
    On 12/22/23 3:07 AM, WM wrote:
    Le 21/12/2023 à 16:18, Richard Damon a écrit :
    On 12/21/23 5:57 AM, WM wrote:
    Le 21/12/2023 à 10:41, Mikko a écrit :
    On 2023-12-20 16:28:09 +0000, WM said:

    The matrix remains, only the covering by X changes.

    Your first "covering" is just another matrix with the same idicex.

    It has the same (m, n), but not all positions have been counted by k >>>>> k = (m + n - 1)(m + n - 2)/2 + m.

    XOOO...
    XOOO...
    XOOO...
    XOOO...
    ..

    When it "changes" it does not really change, there just is another >>>>>> "covering", and after successive changes a sequence of such
    "coverings".

    And none covers the whole matrix by indeXes.

    So, you just don't understand what the bijection is doing.

    A bijection should index every element of the matrix, but it does
    not, because the O remain. Therefore it is not a bijection.

    So, you are showning that the bijecton for k to k/1 doesn't cover all
    of m/n, but that doesn't show that the bijecton for k to m/n where
    k = (m + n - 1)(m + n - 2)/2 + m doesn't work.

    If the first bijection k to k/1 is a bijection, then I apply in the
    second stepCantor's procedure. If not, then there are no infinite
    bijections at all.

    Regards, WM

    Excpet that it CAN'T be a "bijection" to the set of m/n as it isn't an injection when reversed.

    Yes, it can be a bijection of natural to integral rationals, but that
    isn't the set you are talking about, so irrelevent.

    I don't think you understand anything that you are talking about.

    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)
  • From Jim Burns@21:1/5 to All on Fri Dec 22 18:06:33 2023
    On 12/22/2023 3:13 AM, WM wrote:
    Le 21/12/2023 à 19:18, Jim Burns a écrit :
    On 12/20/2023 5:51 AM, WM wrote:

    First, this is a stupid rule.

    What you (WM) are probably calling stupid
    isn't so much a _rule_ rule
    as it is the _observation_ that
    the flock-of-sheep rule isn't correct
    for all sets.

    That is not an observation

    𝕍 is the least.upper.bound of FISONs ⟨1,…,n⟩
    ∀A: ∀⟨1,…,n⟩ ⊆ A ⟹ ∀⟨1,…,n⟩ ⊆ 𝕍 ⊆ A
    Standardly, 𝕍 = ℕ, however,
    whatever ℕ is, consider least.upper.bound 𝕍

    𝕍 is all.visibleᵂᴹ

    cantor: 𝕍×𝕍 ↣ 𝕍×{1}
    cantor⟨i,j⟩ = ⟨kᵢⱼ,1⟩
    kᵢⱼ = i+(i+j-1)(i+j-2)/2

    cantor⁻¹⟨k,1⟩ = ⟨iₖ,jₖ⟩
    sₖ = max{h| (h-1)(h-2)/2 < k }
    iₖ+jₖ = sₖ
    iₖ = k-(sₖ-1)(sₖ-2)/2
    jₖ = sₖ(sₖ-1)/2-k+1

    We observe that
    cantor: 𝕍×𝕍 ↣ 𝕍×{1}
    is 1.to.1

    We observe that
    there are at least as many
    in 𝕍×{1} as there are in 𝕍×𝕍

    1.to.1 𝕍×𝕍 ↣ 𝕍×{1}
    |𝕍×𝕍| ≤ |𝕍×{1}|
    and
    𝕍×𝕍 ⊃≠ 𝕍×{1}

    We observe that
    𝕍×𝕍 and 𝕍×{1} are not flocks of sheep.

    That is not an observation
    but an assumption

    an assumption
    of the correctness of arithmetic

    of people who deny or are unable to
    understand actual infinity.

    Your (WM's) actually.infiniteᵂᴹ set A
    does _not_ have internal de.Bob.ification
    but has a _subset_ B ⊆ A
    which _does_ have internal de.Bob.ification.

    Your explanation is that A contains elements
    which "cannot be said" to self-equal.

    I don't see what I can add to that.

    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)
  • From Mikko@21:1/5 to All on Sat Dec 23 14:34:02 2023
    On 2023-12-22 16:19:08 +0000, WM said:

    Le 22/12/2023 à 15:19, Mikko a écrit :
    On 2023-12-22 08:09:14 +0000, WM said:

    Le 21/12/2023 à 17:45, Mikko a écrit :
    On 2023-12-21 11:56:10 +0000, WM said:

    Le 21/12/2023 à 12:48, Mikko a écrit :
    On 2023-12-21 10:57:10 +0000, WM said:

    Le 21/12/2023 à 10:41, Mikko a écrit :
    On 2023-12-20 16:28:09 +0000, WM said:

    The matrix remains, only the covering by X changes.

    Your first "covering" is just another matrix with the same idicex. >>>>>>>
    It has the same (m, n), but not all positions have been counted by k >>>>>>> k = (m + n - 1)(m + n - 2)/2 + m.

    Positions are not covered by. Some of them are covered by X.

    Each position as m and n so each position has k.

    That is an error. Each O remains in the matrix at a not indexed place. >>>>
    By the definition of "matrix" there are only indexed places.

    But not indexed by Cantor's k = (m + n - 1)(m + n - 2)/2 + m.

    It is indexed by two natural numbers because the matrices meintioned
    above and earlier in this discussion are matrices that are indexed by
    two natural numbers.

    As the O, not indexed by Cantor prove, most of the matrix' (m, n) are dark. Only the upper left corner contains visible m and n.

    Every m and every n is "visible". There are no other rows and no other
    columns in the matrix.

    Mikko

    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)
  • From WM@21:1/5 to All on Sat Dec 23 12:57:14 2023
    Le 22/12/2023 à 17:59, Richard Damon a écrit :
    On 12/22/23 11:23 AM, WM wrote:

    but the rule says that if at least ONE covering mapping exists, we
    meet the requirement.

    Who made that rule?

    It wasn't "Made" but "Discovered".

    It was "discovered" that there are as many fractions as prime numbers. No
    it is believed by matheologians who cannot think from 1 to 3. (They would discover that between every pair of natural numebers there are
    infinitely,many fraction.

    It is a natural consequence of the
    need for a consistent set of logical rules to describe infinite set.

    That is not consistent but foolish.

    Regards, WM

    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)
  • From WM@21:1/5 to All on Sat Dec 23 13:17:07 2023
    Le 23/12/2023 à 00:06, Jim Burns a écrit :

    𝕍 is the least.upper.bound of FISONs ⟨1,…,n⟩

    Does not exist.

    The collection of FISONs is potentially infinite.

    Your explanation is that A contains elements
    which "cannot be said" to self-equal.

    Which cannot be *proved* to be anything because they cannot be chosen as individuals. But I assume that they all are natural numbers, all have
    unique prime decompositions and all are equal to themselves.

    Regards, WM

    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)
  • From Richard Damon@21:1/5 to All on Sat Dec 23 08:44:03 2023
    On 12/23/23 7:57 AM, WM wrote:
    Le 22/12/2023 à 17:59, Richard Damon a écrit :
    On 12/22/23 11:23 AM, WM wrote:

    but the rule says that if at least ONE covering mapping exists, we
    meet the requirement.

    Who made that rule?

    It wasn't "Made" but "Discovered".

    It was "discovered" that there are as many fractions as prime numbers.
    No it is believed by matheologians who cannot think from 1 to 3. (They
    would discover that between every pair of natural numebers there are infinitely,many fraction.

    You don't seem to understand basic set theory.

    In the Natural Numbers, between 1 and 3 is only 2, and NO fractions, as
    they aren't part of the set.

    Yes, in the set of Rational Numbers (your fractions) it is well known
    that between any two members of the set there are numerically an
    infinite number of other members. That is because the Rationals are "dense".

    Rationals are a different set then the Natual Numbers (even if the
    values of the Natural Numbers are a subset of the values of the Rationals)

    So you are just caught in a basic lie about "Matheologians".

    It is a natural consequence of the need for a consistent set of
    logical rules to describe infinite set.

    That is not consistent but foolish.

    Nope. You have just shown your total ignorance of what you are talking
    about, not being able to distiguish which number system you are talking
    about.


    Regards, WM

    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)
  • From Jim Burns@21:1/5 to All on Sat Dec 23 12:47:43 2023
    On 12/23/2023 8:17 AM, WM wrote:
    Le 23/12/2023 à 00:06, Jim Burns a écrit :
    On 12/22/2023 3:13 AM, WM wrote:
    Le 21/12/2023 à 19:18, Jim Burns a écrit :
    On 12/20/2023 5:51 AM, WM wrote:

    First, this is a stupid rule.

    What you (WM) are probably calling stupid
    isn't so much a _rule_ rule
    as it is the _observation_ that
    the flock-of-sheep rule isn't correct
    for all sets.

    That is not an observation

    𝕍 is the least.upper.bound of FISONs ⟨1,…,n⟩

    Does not exist.

    The collection of FISONs is
    potentially infinite.

    ⟨⟨1,…,n⟩⟩ is the collection of only each FISON

    if ⟨1,…,k⟩ is not a FISON
    then ⟨1,…,k⟩ ∉ ⟨⟨1,…,n⟩⟩
    (only)
    if ⟨1,…,k⟩ is a FISON
    then ⟨1,…,k⟩ ∈ ⟨⟨1,…,n⟩⟩
    (each)

    ⟨1,…,k⟩ is a FISON
    means,
    for each non.∅ split F,H of ⟨1,…,k⟩
    i‖i⁺¹ exists last‖first in F‖H
    and 1‖k exists first‖last in ⟨1,…,k⟩

    i⁺¹ is not.1 not.doppelgänger not.final
    i⁺¹≠1 ∧ ¬∃h≠i:h⁺¹=i⁺¹ ∧ ∃k=(i⁺¹)⁺¹

    An example definition for which that's true is
    i⁺¹ = ⟨i,1⟩

    First, this is a stupid rule.

    What you (WM) are probably calling stupid
    isn't so much a _rule_ rule
    as it is the _observation_ that
    the flock-of-sheep rule isn't correct
    for all sets.

    That is not an observation

    𝕍 is the least.upper.bound of FISONs ⟨1,…,n⟩

    Does not exist.

    Each ⟨1,…,n⟩ subsets ℕ
    ∀⟨1,…,n⟩ ⊆ ℕ
    I'll say ℕ is _each.FISON_

    𝕍 is the intersection of
    all each.FISON subsets of ℕ
    𝕍 = ⋂{S⊆ℕ| ∀⟨1,…,n⟩ ⊆ ℕ }

    𝕍 is each.FISON

    For each each.FISON subset S of ℕ
    𝕍 subset S

    For any each.FISON set B not.subset ℕ
    B∩ℕ is each.FISON
    B∩ℕ subset ℕ
    𝕍 subset B∩ℕ


    Consider
    the intersection 𝕍′ of
    all each.FISON subsets of B
    𝕍′ = ⋂{S⊆B| ∀⟨1,…,n⟩ ⊆ ℕ }

    For each each.FISON subset S′ of B
    each.FISON 𝕍′ subset S′
    each.FISON 𝕍 subset B∩ℕ subset B
    𝕍′ subset 𝕍

    And
    each.FISON 𝕍′ subset 𝕍 subset ℕ
    𝕍 subset 𝕍′ subset 𝕍
    𝕍 = 𝕍′

    𝕍 is the least.upper.bound of FISONs ⟨1,…,n⟩

    Does not exist.

    𝕍 is the intersection of
    all each.FISON subsets of ℕ
    𝕍 = ⋂{S⊆ℕ| ∀⟨1,…,n⟩ ⊆ ℕ }

    For any each.FISON set B
    each.FISON 𝕍 subset B
    ∀B: ∀⟨1,…,n⟩ ⊆ B ⟹ ∀⟨1,…,n⟩ ⊆ 𝕍 ⊆ B

    𝕍 is the greatest.lower.bound of each.FISON sets
    and
    𝕍 is the least.upper.bound of FISONs

    Your explanation is that A contains elements
    which "cannot be said" to self-equal.

    Which cannot be *proved* to be anything
    because they cannot be chosen as individuals.
    But I assume that they all are natural numbers,
    all have unique prime decompositions and
    all are equal to themselves.

    Finite.length claims can be made about them.
    You (WM) just did that.

    We can claim
    that k ∈ 𝕍 ends FISON ⟨1,…,k⟩ [1]

    That's true in infinitely.many ways.
    Another way to say that is
    k ∈ 𝕍 and
    k ends ⟨1,…,k⟩
    are true in ways which
    have internal de.Bob.ification.

    We can augment [1] with only
    not.first.false claims and,
    in each of the ways [1] is true,
    the augmenting claims are in
    a finite sequence without first.false.

    Each claim in such a sequence is true.
    We know it's true because of what we know
    about finite sequence, not because of
    anything we know about k or 𝕍

    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)
  • From WM@21:1/5 to All on Sun Dec 24 08:31:17 2023
    Le 23/12/2023 à 14:44, Richard Damon a écrit :

    You don't seem to understand basic set theory.

    Teach me. Does NUF(x) increase from NUF(0) = 0 to NUF(x>0) > 0 in steps of height 1 or not?

    Regards, WM

    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)
  • From WM@21:1/5 to All on Sun Dec 24 08:30:01 2023
    Le 23/12/2023 à 18:47, Jim Burns a écrit :

    Each claim in such a sequence is true.

    Does NUF(x) increase from NUF(0) = 0 to NUF(x>0) > 0 in steps og height 1
    or not?

    Regards, WM

    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)
  • From Richard Damon@21:1/5 to All on Sun Dec 24 12:11:14 2023
    On 12/24/23 3:31 AM, WM wrote:
    Le 23/12/2023 à 14:44, Richard Damon a écrit :

    You don't seem to understand basic set theory.

    Teach me. Does NUF(x) increase from NUF(0) = 0 to NUF(x>0) > 0 in steps
    of height 1 or not?

    Regards, WM

    Not at finite numbers x.

    The "points" where NUF(x) might increase in units of height of 1, would
    be at the infintesimals, with a proper definition of NUF to handle them.

    Otherwise, it is just discontinuous, because it doesn't take into
    account unbounded nature of the unit fractions.


    Note, "Set Theory" doesn't help you here, a your attempt to define
    NUF(x) based on the number of unit fractions with values less than the
    current point doesn't actually say that it ever needs to be finite for x
    0, as your logic of NUF(x) being 1 at the "lowest" Unit Fraction is a
    flawed assumption, YOU need to prove the existance of such a point, you
    can't claim a "necessary continuity" that hasn't actually been established.

    You are just showing your ignorance of how mathematics actually works,
    because you make unwarrented assumptions based on finite system when you
    move to infinite systems.

    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)
  • From Jim Burns@21:1/5 to All on Sun Dec 24 15:44:22 2023
    On 12/24/2023 3:30 AM, WM wrote:
    Le 23/12/2023 à 18:47, Jim Burns a écrit :

    Each claim in such a sequence is true.

    Does NUF(x) increase
    from NUF(0) = 0 to NUF(x>0) > 0
    in steps [of] height 1 or not?

    You (WM) _want_ to only be told a claim
    without an argument for it.
    You are _demanding_ no.argument.

    The following isn't what you're demanding:
    ----
    This true claim:
    | for each unit.fraction ⅟j
    | there is a unit fraction ⅟kⱼ such that
    | ⅟kⱼ < ⅟j
    |
    | Example: Consider kⱼ = j+1
    | It's a true claim.
    |
    does not lead to this false claim:
    | there is a unit fraction ⅟k such that
    | for each unit.fraction ⅟jₖ
    | ⅟k < ⅟jₖ
    |
    | Assume it's true.
    | Counter.example: Consider jₖ+1 = k
    | It's a false claim.

    Instead,
    the first claim, a true claim,
    leads to the _negation_ of
    what you (WM) want to be a true claim.

    It's negation is a true claim:
    | there _isn't_ a unit fraction ⅟k such that
    | for each unit.fraction ⅟jₖ
    | ⅟k < ⅟jₖ
    |
    | Assume otherwise.
    | However: Consider jₖ+1 = k
    | Contradiction.
    | Otherwise is false.

    Does NUF(x) increase
    from NUF(0) = 0 to NUF(x>0) > 0
    in steps [of] height 1 or not?

    No.

    ----
    That particular quantifier shift,
    for any predicate P(⅟k,⅟j)
    from
    | ∀⅟j ∃⅟kⱼ: P(⅟kⱼ,⅟j)
    to
    | ∃⅟k ∀⅟jₖ: P(⅟k,⅟jₖ)
    _never_ works as an argument, because it
    _doesn't always_ work as an argument.

    Perhaps there is another route to
    arrive at the same claim.
    Perhaps there isn't another route.
    Either way, that isn't a route.

    On the other hand,
    shifting quantifiers in the other direction
    always works, and thus can be used to justify
    claims as being not-first-false.

    True claim
    | there is a real x such that
    | for each unit.fraction ⅟jₓ
    | x < ⅟jₓ
    |
    | Example: Consider x = 0
    |
    leads to true claim
    | for each unit.fraction ⅟j
    | there is a real xⱼ such that
    | xⱼ < ⅟j
    |
    | Example: Consider xⱼ = 0

    ∀⅟j,∃xⱼ:P(xⱼ,⅟j)
    is always not.first.false in
    ⟨ ∃x,∀⅟jₓ:P(x,⅟jₓ) ∀⅟j,∃xⱼ:P(xⱼ,⅟j) ⟩

    ∃⅟k,∀⅟jₖ:P(⅟k,⅟jₖ)
    is _not_ always not.first.false in
    ⟨ ∀⅟j,∃⅟kⱼ:P(⅟kⱼ,⅟j) ∃⅟k,∀⅟jₖ:P(⅟k,⅟jₖ) ⟩

    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)
  • From WM@21:1/5 to All on Mon Dec 25 19:29:08 2023
    Le 24/12/2023 à 18:11, Richard Damon a écrit :
    On 12/24/23 3:31 AM, WM wrote:
    Le 23/12/2023 à 14:44, Richard Damon a écrit :

    You don't seem to understand basic set theory.

    Teach me. Does NUF(x) increase from NUF(0) = 0 to NUF(x>0) > 0 in steps
    of height 1 or not?

    Not at finite numbers x.

    It can increase only at unit fractions. They all are finite numbers.

    The "points" where NUF(x) might increase in units of height of 1, would
    be at the infintesimals, with a proper definition of NUF to handle them.

    Unit fractions are not infinitesimals because natnumbers are not infinite.

    Otherwise, it is just discontinuous,

    Not ain agreement with mathematics: ∀n ∈ ℕ: 1/n - 1/(n+1) = d_n > 0.

    because it doesn't take into
    account unbounded nature of the unit fractions.

    Mathjematics shows that the unit fractions appear unbounded only because
    the dark unit fractions cannot ne seen.

    your logic of NUF(x) being 1 at the "lowest" Unit Fraction is a
    flawed assumption,

    It is simple and basic mathematics, valid for all unit fractions: n ∈
    ℕ: 1/n - 1/(n+1) = d_n > 0.

    You are just showing your ignorance of how mathematics actually works,

    Not like this? n ∈ ℕ: 1/n - 1/(n+1) = d_n > 0.

    Regards, WM

    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)
  • From Richard Damon@21:1/5 to All on Mon Dec 25 14:47:59 2023
    On 12/25/23 2:29 PM, WM wrote:
    Le 24/12/2023 à 18:11, Richard Damon a écrit :
    On 12/24/23 3:31 AM, WM wrote:
    Le 23/12/2023 à 14:44, Richard Damon a écrit :

    You don't seem to understand basic set theory.

    Teach me. Does NUF(x) increase from NUF(0) = 0 to NUF(x>0) > 0 in
    steps of height 1 or not?

    Not at finite numbers x.

    It can increase only at unit fractions. They all are finite numbers.

    And, since there is no minimum unit fraction, it can never have a finite
    value that is greater than 0.


    The "points" where NUF(x) might increase in units of height of 1,
    would be at the infintesimals, with a proper definition of NUF to
    handle them.

    Unit fractions are not infinitesimals because natnumbers are not infinite.

    Right, but that is the only points that you could try to define a NUF
    that has a finite value (other than 0)


    Otherwise, it is just discontinuous,

    Not ain agreement with mathematics: ∀n ∈ ℕ: 1/n - 1/(n+1) = d_n > 0.

    Where does that say your NUF can't be discontinous at 0?


    because it doesn't take into account unbounded nature of the unit
    fractions.

    Mathjematics shows that the unit fractions appear unbounded only because
    the dark unit fractions cannot ne seen.

    Nope. They ARE Unbounded as there is no bound to how small a Unit
    Fraction can get.


    your logic of NUF(x) being 1 at the "lowest" Unit Fraction is a flawed
    assumption,

    It is simple and basic mathematics, valid for all unit fractions: n ∈ ℕ: 1/n - 1/(n+1) = d_n > 0.

    Which does imply what you are trying to claim it does.

    Your logic is just invalid.


    You are just showing your ignorance of how mathematics actually works,

    Not like this? n ∈ ℕ: 1/n - 1/(n+1) = d_n > 0.
    Regards, WM


    WHich doesn't show what you are claiming it does.

    THat says from one unit fraction to another, it can only change by 1.

    Since 0 isn't a unit fration, is says nothing about how it increase from
    there to any unit fraction, except maybe the smallest, which doesn't
    exist, so doesn't apply.

    Your argument is just based on there needing to be a smallest Unit
    fraction, and thus a highest Natural Number, but such a thing doesn't
    actually exist.

    So, your logic is based on false assumptions,

    And you are showing your stupidity by repeating the debunked claim.

    You seem to be as dumb at Peter Olcott.

    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)
  • From Jim Burns@21:1/5 to All on Mon Dec 25 16:59:02 2023
    On 12/25/2023 2:29 PM, WM wrote:
    Le 24/12/2023 à 18:11, Richard Damon a écrit :

    [...]

    It can increase only at unit fractions.

    It = |{⅟n|⅟n<x}|

    It can increase only _near_ unit fractions.

    |{⅟n|⅟n<x}| does not increase at _point_ x

    |{⅟n|⅟n<x}| increases between _points_ x x′


    |{⅟n|⅟n<x}| increases _near_ 0
    For any positive x
    |{⅟n|⅟n<0}| < |{⅟n|⅟n<x}|

    0 is is not a unit fraction.
    However, 0 is near unit fractions.

    There is no positive point x near 0
    which does not have two unit fractions
    between it and 0
    0 < ⅟mₓ⁺¹ < ⅟mₓ < x

    Thus,
    |{⅟n|⅟n<x}| increases by more than 1 _near_ 0

    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)
  • From WM@21:1/5 to All on Mon Dec 25 22:26:51 2023
    Le 24/12/2023 à 21:44, Jim Burns a écrit :
    On 12/24/2023 3:30 AM, WM wrote:
    Le 23/12/2023 à 18:47, Jim Burns a écrit :

    Each claim in such a sequence is true.

    Does NUF(x) increase
    from NUF(0) = 0 to NUF(x>0) > 0
    in steps [of] height 1 or not?

    You (WM) _want_ to only be told a claim
    without an argument for it.
    You are _demanding_ no.argument.

    I am interested only whether you agree with mathematics or not.

    The following isn't what you're demanding:
    ----
    This true claim:
    | for each unit.fraction ⅟j
    | there is a unit fraction ⅟kⱼ such that
    | ⅟kⱼ < ⅟j
    |
    | Example: Consider kⱼ = j+1
    | It's a true claim.

    That is true if for each j there is j+1.

    But you seem to be convinced that ∀n ∈ ℕ: 1/n - 1/(n+1) = d_n > 0
    should be disregarded. Why?

    Does NUF(x) increase
    from NUF(0) = 0 to NUF(x>0) > 0
    in steps [of] height 1 or not?

    No.

    When going over points with gaps in linear order you will encounter a
    first one.

    That particular quantifier shift,

    Nonsense. If every point of (0, 1] has the property to have ℵ₀ unit fractions at its left-hand side, then all points of (0. 1] have the
    property to have ℵ₀ unit fractions at their left-hand side, then the interval (0, 1] has the property to have ℵ₀ unit fractions at its
    left-hand side. That is simplest logic and mathematics. Because if the
    latter was wrong, then this wrongness had to be proved. This could be done
    only by showing a point og (0, 1] for which it was wrong.

    Regards, WM

    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)
  • From WM@21:1/5 to All on Mon Dec 25 22:34:04 2023
    Le 25/12/2023 à 22:59, Jim Burns a écrit :
    On 12/25/2023 2:29 PM, WM wrote:
    Le 24/12/2023 à 18:11, Richard Damon a écrit :

    [...]

    It can increase only at unit fractions.

    It = |{⅟n|⅟n<x}|

    It can increase only _near_ unit fractions.

    |{⅟n|⅟n<x}| does not increase at _point_ x

    |{⅟n|⅟n<x}| increases between _points_ x x′


    |{⅟n|⅟n<x}| increases _near_ 0
    For any positive x
    |{⅟n|⅟n<0}| < |{⅟n|⅟n<x}|

    0 is is not a unit fraction.
    However, 0 is near unit fractions.

    There is no positive point x near 0
    which does not have two unit fractions
    between it and 0
    0 < ⅟mₓ⁺¹ < ⅟mₓ < x

    No such point can be found. But if no such point existed, then never the complete set could be applied for enumerating purposes.

    Thus,
    |{⅟n|⅟n<x}| increases by more than 1 _near_ 0

    That contradicts that gaps between all unit fractions exist. If all unit fractions have distances, then the distance after each one causes a halt
    in the increase of NUF(x). That is true for all and therefore also for the first unit fraction.

    Regards, WM

    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)
  • From WM@21:1/5 to All on Mon Dec 25 22:47:14 2023
    Le 25/12/2023 à 20:47, Richard Damon a écrit :
    On 12/25/23 2:29 PM, WM wrote:
    Le 24/12/2023 à 18:11, Richard Damon a écrit :
    On 12/24/23 3:31 AM, WM wrote:
    Le 23/12/2023 à 14:44, Richard Damon a écrit :

    You don't seem to understand basic set theory.

    Teach me. Does NUF(x) increase from NUF(0) = 0 to NUF(x>0) > 0 in
    steps of height 1 or not?

    Not at finite numbers x.

    It can increase only at unit fractions. They all are finite numbers.

    And, since there is no minimum unit fraction, it can never have a finite value that is greater than 0.

    The solution is dark unit fractions.


    The "points" where NUF(x) might increase in units of height of 1,
    would be at the infintesimals, with a proper definition of NUF to
    handle them.

    Unit fractions are not infinitesimals because natnumbers are not infinite.

    Right, but that is the only points that you could try to define a NUF
    that has a finite value (other than 0)

    The solution is dark unit fractions.


    Otherwise, it is just discontinuous,

    Not in agreement with mathematics: ∀n ∈ ℕ: 1/n - 1/(n+1) = d_n > 0.

    Where does that say your NUF can't be discontinous at 0?

    After every unit fractions there is a gap. So NUF cannot increase by more
    than 1 at every point.


    because it doesn't take into account unbounded nature of the unit
    fractions.

    Mathjematics shows that the unit fractions appear unbounded only because
    the dark unit fractions cannot ne seen.

    Nope. They ARE Unbounded as there is no bound to how small a Unit
    Fraction can get.

    After every unit fractions there is a gap. So NUF cannot increase by more
    than 1 at every point.


    your logic of NUF(x) being 1 at the "lowest" Unit Fraction is a flawed
    assumption,

    It is simple and basic mathematics, valid for all unit fractions: n ∈ ℕ: >> 1/n - 1/(n+1) = d_n > 0.

    Which does imply what you are trying to claim it does.

    Of course.

    THat says from one unit fraction to another, it can only change by 1.

    Since 0 isn't a unit fration, is says nothing about how it increase from there to any unit fraction,


    but from one unit fraction to another, it can only change by 1.

    Your argument is just based on

    from one unit fraction to another, it can only change by 1.

    there needing to be a smallest Unit
    fraction, and thus a highest Natural Number, but such a thing doesn't actually exist.

    Nevertheless, the increase by more than one from 0 to ℵo cannot be
    discerned. These unit fractions are not available as individuals.

    Regards, WM

    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)
  • From Richard Damon@21:1/5 to All on Mon Dec 25 19:38:45 2023
    On 12/25/23 5:47 PM, WM wrote:
    Le 25/12/2023 à 20:47, Richard Damon a écrit :
    On 12/25/23 2:29 PM, WM wrote:
    Le 24/12/2023 à 18:11, Richard Damon a écrit :
    On 12/24/23 3:31 AM, WM wrote:
    Le 23/12/2023 à 14:44, Richard Damon a écrit :

    You don't seem to understand basic set theory.

    Teach me. Does NUF(x) increase from NUF(0) = 0 to NUF(x>0) > 0 in
    steps of height 1 or not?

    Not at finite numbers x.

    It can increase only at unit fractions. They all are finite numbers.

    And, since there is no minimum unit fraction, it can never have a
    finite value that is greater than 0.

    The solution is dark unit fractions.

    A solution looking for a problem to solve it seems.

    And based on the assumption of the impossible.



    The "points" where NUF(x) might increase in units of height of 1,
    would be at the infintesimals, with a proper definition of NUF to
    handle them.

    Unit fractions are not infinitesimals because natnumbers are not
    infinite.

    Right, but that is the only points that you could try to define a NUF
    that has a finite value (other than 0)

    The solution is dark unit fractions.

    Again, a solution looking for a problem. And doing it by assuming the impossible happens.



    Otherwise, it is just discontinuous,

    Not in agreement with mathematics: ∀n ∈ ℕ: 1/n - 1/(n+1) = d_n > 0. >>
    Where does that say your NUF can't be discontinous at 0?

    After every unit fractions there is a gap. So NUF cannot increase by
    more than 1 at every point.

    So. That only applies at unit fractions. Since 0 isn't one, that doesn't
    apply there.



    because it doesn't take into account unbounded nature of the unit
    fractions.

    Mathjematics shows that the unit fractions appear unbounded only
    because the dark unit fractions cannot ne seen.

    Nope. They ARE Unbounded as there is no bound to how small a Unit
    Fraction can get.

    After every unit fractions there is a gap. So NUF cannot increase by
    more than 1 at every point.

    But that doesn't constrain it increasing after 0, since it isn't a unit fraction, there is no need for a finte space after it. It can just the
    infinite amount in the infintesimal space after 0.



    your logic of NUF(x) being 1 at the "lowest" Unit Fraction is a
    flawed assumption,

    It is simple and basic mathematics, valid for all unit fractions: n ∈
    ℕ: 1/n - 1/(n+1) = d_n > 0.

    Which does imply what you are trying to claim it does.

    Of course.

    Typo.

    You are just showing you lack of ability to actually reason.


    THat says from one unit fraction to another, it can only change by 1.

    Since 0 isn't a unit fration, is says nothing about how it increase
    from there to any unit fraction,


    but from one unit fraction to another, it can only change by 1.

    but 0 isn't a unit fraction, so the "gap" after it isn't constrained.


    Your argument is just based on

    from one unit fraction to another, it can only change by 1.

    And 0 isn't a unit fraction, so it isn't constrained.


    there needing to be a smallest Unit fraction, and thus a highest
    Natural Number, but such a thing doesn't actually exist.

    Nevertheless, the increase by more than one from 0 to ℵo cannot be discerned. These unit fractions are not available as individuals.

    Regards, WM



    YOU may not discern it, but it does happen.

    You are just showing your ignorance.

    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)
  • From Richard Damon@21:1/5 to All on Mon Dec 25 22:59:38 2023
    On 12/25/23 5:26 PM, WM wrote:
    Le 24/12/2023 à 21:44, Jim Burns a écrit :
    On 12/24/2023 3:30 AM, WM wrote:
    Le 23/12/2023 à 18:47, Jim Burns a écrit :

    Each claim in such a sequence is true.

    Does NUF(x) increase
    from NUF(0) = 0 to NUF(x>0) > 0
    in steps [of] height 1 or not?

    You (WM) _want_ to only be told a claim
    without an argument for it.
    You are _demanding_ no.argument.

    I am interested only whether you agree with mathematics or not.

    The following isn't what you're demanding:
    ----
    This true claim:
    | for each unit.fraction ⅟j
    | there is a unit fraction ⅟kⱼ such that
    | ⅟kⱼ < ⅟j
    |
    | Example: Consider kⱼ = j+1
    | It's a true claim.

    That is true if for each j there is j+1.

    But you seem to be convinced that ∀n ∈ ℕ: 1/n - 1/(n+1) = d_n > 0 should
    be disregarded. Why?

    Does NUF(x) increase
    from NUF(0) = 0 to NUF(x>0) > 0
    in steps [of] height 1 or not?

    No.

    When going over points with gaps in linear order you will encounter a
    first one.

    Nope, not when the gaps keep on shrinking as fast as they do.

    At the point 1/n, the size of the gap is such that you would expect to
    be able to put at least another n gaps between you and the origin, so
    that gaps shrink faster than they approach 0.



    That particular quantifier shift,

    Nonsense. If every point of (0, 1] has the property to have ℵ₀ unit fractions at its left-hand side, then all points of (0. 1] have the
    property to have ℵ₀ unit fractions at their left-hand side, then the interval (0, 1] has the property to have ℵ₀ unit fractions at its left-hand side. That is simplest logic and mathematics. Because if the
    latter was wrong, then this wrongness had to be proved. This could be
    done only by showing a point og (0, 1] for which it was wrong.

    Regards, WM

    And what is WRONG with th concept that ALL finite number x > 0 (or unit fractions) have an infinite number between them and zero?

    That is just another way to state the density property for the numbers aproaching 0.

    You seem to have a big problem with sets being infinite.

    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)
  • From Richard Damon@21:1/5 to All on Mon Dec 25 22:52:15 2023
    On 12/25/23 5:34 PM, WM wrote:
    Le 25/12/2023 à 22:59, Jim Burns a écrit :
    On 12/25/2023 2:29 PM, WM wrote:
    Le 24/12/2023 à 18:11, Richard Damon a écrit :

    [...]

    It can increase only at unit fractions.

    It = |{⅟n|⅟n<x}|

    It can increase only _near_ unit fractions.

    |{⅟n|⅟n<x}| does not increase at _point_ x

    |{⅟n|⅟n<x}| increases between _points_ x x′


    |{⅟n|⅟n<x}| increases _near_ 0
    For any positive x
    |{⅟n|⅟n<0}| < |{⅟n|⅟n<x}|

    0 is is not a unit fraction.
    However, 0 is near unit fractions.

    There is no positive point x near 0
    which does not have two unit fractions
    between it and 0
    0 < ⅟mₓ⁺¹ < ⅟mₓ < x

    No such point can be found. But if no such point existed, then never the complete set could be applied for enumerating purposes.

    Depends on how you define "applied".

    Your definition seems to say that the whole set of Natural Numbers can
    EVER be applied, which just says that your logic can't handle tha
    Natural numbers.

    Normal logic has no problem going FORWRD from the small natural numbers
    up to infinity. As long as you START at a defined number, you can
    continue on. Trying to start at the (non-existant) top gets you in
    trouble as you can't actually start.


    Thus,
    |{⅟n|⅟n<x}| increases by more than 1 _near_ 0

    That contradicts that gaps between all unit fractions exist. If all unit fractions have distances, then the distance after each one causes a halt
    in the increase of NUF(x). That is true for all and therefore also for
    the first unit fraction.

    Nope. What two unit fractions didn't have a gap?

    Remember, 0 isn't a unit fraction, so it doesn't count.

    It isn't "True for the first unit fraction" since the "First Unit
    Fraction" doesn't exist for it to be true at. (Try to name it).

    Your logic is just broken, as it is based on the non-existant magical
    unicorn to power it.


    Regards, WM



    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)
  • From WM@21:1/5 to All on Tue Dec 26 10:21:30 2023
    Le 26/12/2023 à 01:38, Richard Damon a écrit :
    On 12/25/23 5:47 PM, WM wrote:

    Your argument is just based on

    from one unit fraction to another, it can only change by 1.

    And 0 isn't a unit fraction, so it isn't constrained.

    But as soon as a unit fraction is encountered, it is followed by a gap.

    Nevertheless, the increase by more than one from 0 to ℵo cannot be
    discerned. These unit fractions are not available as individuals.

    YOU may not discern it, but it does happen.

    Can you discern it?

    Regards, WM

    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)
  • From WM@21:1/5 to All on Tue Dec 26 10:26:15 2023
    Le 26/12/2023 à 04:59, Richard Damon a écrit :
    On 12/25/23 5:26 PM, WM wrote:

    When going over points with gaps in linear order you will encounter a
    first one.

    Nope, not when the gaps keep on shrinking as fast as they do.

    Non-empty gaps remain non-empty.

    At the point 1/n, the size of the gap is such that you would expect to
    be able to put at least another n gaps between you and the origin, so
    that gaps shrink faster than they approach 0.

    All gaps are containing more than one point.

    And what is WRONG with the concept that ALL finite number x > 0 (or unit fractions) have an infinite number between them and zero?

    It clashes with ∀n ∈ ℕ: 1/n - 1/(n+1) = d_n > 0 where d_n consists
    of uncountably many points.

    Regards, WM

    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)
  • From Richard Damon@21:1/5 to All on Tue Dec 26 07:48:04 2023
    On 12/26/23 5:21 AM, WM wrote:
    Le 26/12/2023 à 01:38, Richard Damon a écrit :
    On 12/25/23 5:47 PM, WM wrote:

    Your argument is just based on

    from one unit fraction to another, it can only change by 1.

    And 0 isn't a unit fraction, so it isn't constrained.

    But as soon as a unit fraction is encountered, it is followed by a gap.

    But before that unit fraction, there are more unit fractions, so there
    isn't a first.


    Nevertheless, the increase by more than one from 0 to ℵo cannot be
    discerned. These unit fractions are not available as individuals.

    YOU may not discern it, but it does happen.

    Can you discern it?

    Yes, I can discern that the numbers get unboundedly dense towards 0, so
    the density goes towards infinity and thus the delta of your "NUF(x)"
    becomes infinite.

    The fact that YOU can't see that shows a lack of YOUR understanding.

    You keep on thinking that there is a first unit fraction (at the small
    value end) when there is no such thing, so you seem to have blinders on.


    Regards, WM


    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)
  • From Richard Damon@21:1/5 to All on Tue Dec 26 07:53:24 2023
    On 12/26/23 5:26 AM, WM wrote:
    Le 26/12/2023 à 04:59, Richard Damon a écrit :
    On 12/25/23 5:26 PM, WM wrote:

    When going over points with gaps in linear order you will encounter a
    first one.

    Nope, not when the gaps keep on shrinking as fast as they do.

    Non-empty gaps remain non-empty.

    Right, and NEVER reach to 0.


    At the point 1/n, the size of the gap is such that you would expect to
    be able to put at least another n gaps between you and the origin, so
    that gaps shrink faster than they approach 0.

    All gaps are containing more than one point.

    But since none of the gaps go to 0, so your argument doesn't apply


    And what is WRONG with the concept that ALL finite number x > 0 (or
    unit fractions) have an infinite number between them and zero?

    It clashes with ∀n ∈ ℕ: 1/n - 1/(n+1) = d_n > 0 where d_n consists of uncountably many points.


    How? That says that BETWEEN any TWO numbers there is a finite distance,
    but that distance can get unboundedly small, so we can fit an unbounded
    number of them in the finite space.

    It says NOTHING about the gap between 0 and any unit fraction, as any
    such gap will have space for an infinite number of these unit fractions.

    I guess you can't figure out how Achilless can pass the tortoise even if
    it needs an infinite number of cycles of catching up to where the
    tortoise was when he started that step.

    Your mind just can't seem to handle unbounded sets.

    Regards, WM



    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)
  • From WM@21:1/5 to All on Tue Dec 26 18:17:25 2023
    Le 26/12/2023 à 13:48, Richard Damon a écrit :
    On 12/26/23 5:21 AM, WM wrote:

    But as soon as a unit fraction is encountered, it is followed by a gap.

    But before that unit fraction, there are more unit fractions,

    Before the first, there are more? The they are dark.

    Nevertheless, the increase by more than one from 0 to ℵo cannot be
    discerned. These unit fractions are not available as individuals.

    YOU may not discern it, but it does happen.

    Can you discern it?

    Yes, I can discern that the numbers get unboundedly dense towards 0,

    But since you cannot distinguish these dense unit fractions, they are
    dark.

    Regards, WM

    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)
  • From WM@21:1/5 to All on Tue Dec 26 18:20:36 2023
    Le 26/12/2023 à 13:53, Richard Damon a écrit :

    How? That says that BETWEEN any TWO numbers there is a finite distance,
    but that distance can get unboundedly small, so we can fit an unbounded number of them in the finite space.

    Each gap contains an uncountably many points.

    Regards, WM

    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)
  • From WM@21:1/5 to All on Tue Dec 26 19:09:08 2023
    Le 26/12/2023 à 19:26, Jim Burns a écrit :
    On 12/26/2023 5:26 AM, WM wrote:

    What makes a gap a gap

    Non-empty means that the aps contain points, even uncountably many, none
    of which is a unit fraction.

    Instead of "points with gaps",
    consider "points next to each other"

    There are no unit fractions next to each other.

    Two points are _next_ to each other
    if there is no point between them.
    Between them is empty.

    Such points exist only in the form that one of them is dark.

    ⅟1 is next to ⅟2
    ⅟n is next to ⅟n⁺¹ and next to ⅟n⁻¹

    No. There are uncoauntably many points between them.

    If the first unit fraction existed,
    it would be next to 0

    Not necessarily. But we will never know.

    No unit fraction is next to 0
    No unit fraction is the first unit fraction.

    There is a point next to 0, because otherwise there would be nothing next
    to 0: A gapnot contauning points.

    When going over points with
    gaps in linear order
    you will encounter a first one.

    No.
    The first unit fraction not.exists.

    The alternative is more than one. But it can be excluded because of the
    gap after every unit fraction.

    Regards, WM

    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)
  • From Jim Burns@21:1/5 to All on Tue Dec 26 13:26:39 2023
    On 12/26/2023 5:26 AM, WM wrote:
    Le 26/12/2023 à 04:59, Richard Damon a écrit :
    On 12/25/23 5:26 PM, WM wrote:

    When going over points with
    gaps in linear order
    you will encounter a first one.

    Nope,
    not when the gaps keep on shrinking
    as fast as they do.

    Non-empty gaps remain non-empty.

    What makes a gap a gap
    is it being empty.
    Non-empty gaps are non-gaps.

    So, you (WM) have some private meaning of "gap"
    We need other words to communicate.
    Instead of "points with gaps",
    consider "points next to each other".

    Two points are _next_ to each other
    if there is no point between them.
    Between them is empty.

    ⅟1 is next to ⅟2
    ⅟n is next to ⅟n⁺¹ and next to ⅟n⁻¹

    If the first unit fraction existed,
    it would be next to 0
    No unit fraction is next to 0
    No unit fraction is the first unit fraction.

    When going over points with
    gaps in linear order
    you will encounter a first one.

    No.
    The first unit fraction not.exists.

    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)
  • From Jim Burns@21:1/5 to All on Tue Dec 26 15:23:39 2023
    On 12/26/2023 2:09 PM, WM wrote:
    Le 26/12/2023 à 19:26, Jim Burns a écrit :
    On 12/26/2023 5:26 AM, WM wrote:

    Non-empty gaps remain non-empty.

    What makes a gap a gap

    Non-empty means that the aps contain points,
    even uncountably many,
    none of which is a unit fraction.

    "none of which is a unit fraction" is
    what makes it a gap _in the unit fractions_

    Instead of "points with gaps",
    consider "points next to each other"

    There are no
    unit fractions next to each other.

    Fascinating.

    https://en.wikipedia.org/wiki/Programme_for_International_Student_Assessment

    How would you do, in an assessment?

    Two points are _next_ to each other
    if there is no point between them.
    Between them is empty.

    Such points exist only in the form that
    one of them is dark.

    ⅟1 is next to ⅟2
    ⅟n is next to ⅟n⁺¹ and next to ⅟n⁻¹

    No.
    There are uncountably many points
    between them.

    There are no unit fractions between ⅟1 and ⅟2
    between ⅟n⁺¹ and ⅟n
    between ⅟n and ⅟n⁻¹
    NOT "we can't seeᵂᴹ unit fractions between"
    Unit fractions between not.exist.

    We begin our discussion of unit fractions
    with claims true of each unit fraction,
    seenᵂᴹ or unseenᵂᴹ

    We continue our discussion with claims
    not.first.false of each unit fraction,
    seenᵂᴹ or unseenᵂᴹ

    Because it's a finite sequence of claims,
    each not.first.false of seenᵂᴹ and unseenᵂᴹ
    is true of seenᵂᴹ and unseenᵂᴹ

    That's how we know these not.seenᵂᴹ
    are not.existing,
    by augmenting description not.first.false.ly.

    If the first unit fraction existed,
    it would be next to 0

    Not necessarily.

    Necessarily.
    Because of what "first" means.

    But we will never know.

    People who know what "first" means know.

    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)
  • From Richard Damon@21:1/5 to All on Tue Dec 26 21:29:53 2023
    On 12/26/23 1:20 PM, WM wrote:
    Le 26/12/2023 à 13:53, Richard Damon a écrit :

    How? That says that BETWEEN any TWO numbers there is a finite
    distance, but that distance can get unboundedly small, so we can fit
    an unbounded number of them in the finite space.

    Each gap contains an uncountably many points.

    Regards, WM


    So?

    That just says that finite numbers are dense.

    And, the number of points depends on what number system you are using.

    If the Rationals, then there are a COUNTABLE INFINITE number of points
    between two points. You need to move up to the Reals to get uncountably
    many points.

    This doesn't say anything about the need for there to be a "first" Unit Fraction for NUF(x) to be 1 at.

    So, you are just showing you are out of ideas.

    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)
  • From WM@21:1/5 to All on Wed Dec 27 08:37:52 2023
    Le 26/12/2023 à 21:23, Jim Burns a écrit :

    We begin our discussion of unit fractions
    with claims true of each unit fraction,
    seenᵂᴹ or unseenᵂᴹ

    Yes, in particular this one:
    ∀n ∈ ℕ: 1/n - 1/(n+1) = d_n > 0

    If the first unit fraction existed,
    it would be next to 0

    Not necessarily.

    Necessarily.
    Because of what "first" means.

    "Next to" means that there is no point in between. First means there is no
    unit fraction in between.

    Regards, WM

    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)
  • From WM@21:1/5 to All on Wed Dec 27 09:35:54 2023
    Le 27/12/2023 à 03:29, Richard Damon a écrit :
    On 12/26/23 1:17 PM, WM wrote:

    Yes, I can discern that the numbers get unboundedly dense towards 0,

    But since you cannot distinguish these dense unit fractions, they are dark.

    Who says we can not distinguish them.

    You. All you can discern is not "unboundedly dense".

    What we can't do is to try to name them all individually at once, since
    the set has an infinite membership.

    All you can name, at onece or later, belong to a finite initial segment
    much smaller than the whole set. Almost all cannot be named.

    Regards, WM

    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)
  • From WM@21:1/5 to All on Wed Dec 27 09:40:47 2023
    Le 27/12/2023 à 03:29, Richard Damon a écrit :
    On 12/26/23 1:20 PM, WM wrote:
    Le 26/12/2023 à 13:53, Richard Damon a écrit :

    How? That says that BETWEEN any TWO numbers there is a finite
    distance, but that distance can get unboundedly small, so we can fit
    an unbounded number of them in the finite space.

    Each gap contains an uncountably many points.

    This doesn't say anything about the need for there to be a "first" Unit Fraction for NUF(x) to be 1 at.

    For uncountably many points, however small, NUF is constant before and
    after every unit fraction. Hence there cannot be more than one at first.

    Regards, WM

    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)
  • From Richard Damon@21:1/5 to All on Wed Dec 27 08:37:40 2023
    On 12/27/23 4:35 AM, WM wrote:
    Le 27/12/2023 à 03:29, Richard Damon a écrit :
    On 12/26/23 1:17 PM, WM wrote:

    Yes, I can discern that the numbers get unboundedly dense towards 0,

    But since you cannot distinguish these dense unit fractions, they are
    dark.

    Who says we can not distinguish them.

    You. All you can discern is not "unboundedly dense".

    You can discern ANY of the number in that unboundedly dense set.

    You are confusing ANY with ALL.

    Yes, we can not finitely discern the whole set individually, as that
    would be a contradiction, as the set is infinite. We can discern any of
    the elements individually.


    What we can't do is to try to name them all individually at once,
    since the set has an infinite membership.

    All you can name, at onece or later, belong to a finite initial segment
    much smaller than the whole set. Almost all cannot be named.

    Regards, WM



    And all Natural Numbers are finite, and thus can be use to define a
    finite initial segment, so ALL can used.

    Countable unboundedness is a "bridge" between the finite and the infinite.

    It creates an infinite set where every element is finite and
    definable/usable.

    If your logic can't handle it, it isn't suitable for the Natural Numbers.

    If you continue to try, you just prove your stupidity.

    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)
  • From Jim Burns@21:1/5 to All on Wed Dec 27 11:45:29 2023
    On 12/27/2023 3:37 AM, WM wrote:
    Le 26/12/2023 à 21:23, Jim Burns a écrit :

    We begin our discussion of unit fractions
    with claims true of each unit fraction,
    seenᵂᴹ or unseenᵂᴹ

    Yes, in particular this one:
    ∀n ∈ ℕ: 1/n - 1/(n+1) = d_n > 0

    Yes, including that one.
    And there are background claims,
    often unstated, for ℕ and '/' and so on.

    Context is essential to
    correctly interpreting these claims.
    ∀n,m ∈ ℕ: n+m = m+n
    ∀n,m ∈ ℝ: n+m = m+n
    are different claims,
    because they're about different objects.

    ∀n ∈ ℕ: 1/n - 1/(n+1) = d_n > 0
    makes no claim for 0
    because for no n ∈ ℕ is 1/n = 0

    If the first unit fraction existed,
    it would be next to 0

    Not necessarily.

    Necessarily.
    Because of what "first" means.

    "Next to" means that
    there is no point in between.

    "Next to" in the unit.fractions means that
    there is no unit.fraction between.

    Context.
    1 is next to 2 in the integers.
    1 isn't next to 2 in the rationals.

    First means there is
    no unit fraction in between.

    ⅟ℕ₁ is the set of unit fractions.
    ⅟ℕ₁∪{-1,2} is a slightly expanded context.

    ⅟1 exists next to 2 in ⅟ℕ₁∪{-1,2}
    No point in ⅟ℕ₁∪{-1,2} is between ⅟1 and 2
    ⅟1 is largest in ⅟ℕ₁

    For each ⅟n, -1 < ⅟n⁺¹ < ⅟n
    and ⅟n isn't next to -1
    No ⅟n exists in ⅟ℕ₁∪{-1,2} next to -1
    Nothing is smallest in ⅟ℕ₁

    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)
  • From WM@21:1/5 to All on Wed Dec 27 19:24:16 2023
    Le 27/12/2023 à 17:45, Jim Burns a écrit :

    "Next to" means that
    there is no point in between.

    "Next to" in the unit.fractions means that
    there is no unit.fraction between.

    One could use that meaning but it may be misleading. Therefore I don't use
    it.

    Regards, WM

    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)
  • From Richard Damon@21:1/5 to All on Wed Dec 27 15:06:43 2023
    On 12/27/23 2:24 PM, WM wrote:
    Le 27/12/2023 à 17:45, Jim Burns a écrit :

    "Next to" means that
    there is no point in between.

    "Next to" in the unit.fractions means that
    there is no unit.fraction between.

    One could use that meaning but it may be misleading. Therefore I don't
    use it.

    Regards, WM




    But, when our domain of discussion in the "Unit Fractions" then 1/2 is
    "next to" 1/3, as there is nothing IN THE SET between them.

    Just as 1 is next to 2 in the Natural Numbers.

    You seem to be having an issue with keeping defined what number system
    you are working in, which does cause problems knowing what rules are applicable.

    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)
  • From WM@21:1/5 to All on Wed Dec 27 22:15:15 2023
    Le 27/12/2023 à 21:06, Richard Damon a écrit :
    On 12/27/23 2:24 PM, WM wrote:
    Le 27/12/2023 à 17:45, Jim Burns a écrit :

    "Next to" means that
    there is no point in between.

    "Next to" in the unit.fractions means that
    there is no unit.fraction between.

    One could use that meaning but it may be misleading. Therefore I don't
    use it.


    But, when our domain of discussion in the "Unit Fractions" then 1/2 is
    "next to" 1/3, as there is nothing IN THE SET between them.

    But that use encourages you to forget that uncountably many points lie
    between any two unit fractions. Therefore the function NUF(x) has always a level between any two unit fractions. Therefoe never infinitely many can
    be at the start together.

    Regards, WM

    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)
  • From WM@21:1/5 to All on Wed Dec 27 22:47:28 2023
    Le 27/12/2023 à 14:37, Richard Damon a écrit :
    On 12/27/23 4:35 AM, WM wrote:
    Le 27/12/2023 à 03:29, Richard Damon a écrit :
    On 12/26/23 1:17 PM, WM wrote:

    Yes, I can discern that the numbers get unboundedly dense towards 0,

    But since you cannot distinguish these dense unit fractions, they are
    dark.

    Who says we can not distinguish them.

    You. All you can discern is not "unboundedly dense".

    You can discern ANY of the number in that unboundedly dense set.

    Each discerned unit fraction has a distance from its next unit fractions
    and hence is not unboundedly dense.

    And all Natural Numbers are finite, and thus can be use to define a
    finite initial segment, so ALL can used.

    All have distances and therefore NUF(x) can never grow from 0 to infinity without passing a first unit fraction and being constant afterwards for a while.

    Regards, WM

    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)
  • From Richard Damon@21:1/5 to All on Wed Dec 27 21:57:51 2023
    On 12/27/23 5:15 PM, WM wrote:
    Le 27/12/2023 à 21:06, Richard Damon a écrit :
    On 12/27/23 2:24 PM, WM wrote:
    Le 27/12/2023 à 17:45, Jim Burns a écrit :

    "Next to" means that
    there is no point in between.

    "Next to" in the unit.fractions means that
    there is no unit.fraction between.

    One could use that meaning but it may be misleading. Therefore I
    don't use it.


    But, when our domain of discussion in the "Unit Fractions" then 1/2 is
    "next to" 1/3, as there is nothing IN THE SET between them.

    But that use encourages you to forget that uncountably many points lie between any two unit fractions. Therefore the function NUF(x) has always
    a level between any two unit fractions. Therefoe never infinitely many
    can be at the start together.

    Regards, WM


    But there can be infinitely many before any unit fraction, as there
    isn't a "Start" that is a unit fraction.

    Again, you aren't making clear what type of "points" you are taling about!

    Are we talking Rational, Real, of Transfinite points?

    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)
  • From Richard Damon@21:1/5 to All on Wed Dec 27 21:58:16 2023
    On 12/27/23 5:47 PM, WM wrote:
    Le 27/12/2023 à 14:37, Richard Damon a écrit :
    On 12/27/23 4:35 AM, WM wrote:
    Le 27/12/2023 à 03:29, Richard Damon a écrit :
    On 12/26/23 1:17 PM, WM wrote:

    Yes, I can discern that the numbers get unboundedly dense towards 0, >>>>>
    But since you cannot distinguish these dense unit fractions, they
    are dark.

    Who says we can not distinguish them.

    You. All you can discern is not "unboundedly dense".

    You can discern ANY of the number in that unboundedly dense set.

    Each discerned unit fraction has a distance from its next unit fractions
    and hence is not unboundedly dense.

    Then what is the BOUND?

    Remember, that distance reduces without any limit to its size, so you
    claim of not unboundedly dense is proved to be untrue.

    If you disagree, name the bound.


    And all Natural Numbers are finite, and thus can be use to define a
    finite initial segment, so ALL can used.

    All have distances and therefore NUF(x) can never grow from 0 to
    infinity without passing a first unit fraction and being constant
    afterwards for a while.

    Regards, WM


    Which only applies *IF* there IS a "first" Unit Fraction, which there isn't.

    THe problem is your logic only handles "bounded' sets, which isn't what
    we have. You are just using logic which can't handle the numbers you are
    trying to process.

    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)
  • From WM@21:1/5 to All on Thu Dec 28 10:44:11 2023
    Le 28/12/2023 à 03:57, Richard Damon a écrit :
    On 12/27/23 5:15 PM, WM wrote:
    Therefore the function NUF(x) has always
    a level between any two unit fractions. Therefore never infinitely many
    can be at the start together.

    But there can be infinitely many before any unit fraction,

    My logic does not allow that.


    Regards, WM

    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)
  • From Jim Burns@21:1/5 to All on Thu Dec 28 05:39:53 2023
    On 12/27/2023 2:24 PM, WM wrote:
    Le 27/12/2023 à 17:45, Jim Burns a écrit :
    On 12/27/2023 3:37 AM, WM wrote:
    Le 26/12/2023 à 21:23, Jim Burns a écrit :

    If the first unit fraction existed,
    it would be next to 0

    Not necessarily.

    Necessarily.
    Because of what "first" means.

    "Next to" means that
    there is no point in between.

    "Next to" in the unit.fractions means that
    there is no unit.fraction between.

    One could use that meaning
    but it may be misleading.

    Points exist in ℝ between ⅟1 and 2 but
    no point exists in ⅟ℕ₁ between ⅟1 and 2
    ⅟1 exists next to 2 in ⅟ℕ₁∪{-1,2}
    ⅟1 is largest in ⅟ℕ₁

    For each ⅟n in ⅟ℕ₁
    Points exist in ℝ between ⅟n and -1 and
    ⅟n⁺¹ exists in ⅟ℕ₁ between ⅟n and -1
    ⅟n exists not.next to -1 in ⅟ℕ₁∪{-1,2}
    ⅟n is not.smallest in ⅟ℕ₁

    For each ⅟n in ⅟ℕ₁
    ⅟n is not.smallest in ⅟ℕ₁

    Therefore I don't use it.

    Context is essential to
    correctly interpreting these claims.
    ∀n,m ∈ ℕ: n+m = m+n
    ∀n,m ∈ ℝ: n+m = m+n
    are different claims,
    because they're about different objects.

    | ∀n ∈ ℕ: 1/n - 1/(n+1) = d_n > 0
    makes no claim for 0
    because for no n ∈ ℕ is 1/n = 0

    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)
  • From WM@21:1/5 to All on Thu Dec 28 10:52:21 2023
    Le 28/12/2023 à 03:58, Richard Damon a écrit :
    On 12/27/23 5:47 PM, WM wrote:

    Each discerned unit fraction has a distance from its next unit fractions
    and hence is not unboundedly dense.

    Then what is the BOUND?

    The bound is zero. The accumulation point contains dark unit fractions.

    Remember, that distance reduces without any limit to its size, so you
    claim of not unboundedly dense is proved to be untrue.

    Every chosen unit fraction and its neighbours are not unboundedly dense.
    The dark unit fractions appear unboundedly dense.

    All have distances and therefore NUF(x) can never grow from 0 to
    infinity without passing a first unit fraction and being constant
    afterwards for a while.


    Which only applies *IF* there IS a "first" Unit Fraction, which there isn't.

    If there are any unit fractions in separated oder, then there is a first
    one.

    Regards, WM

    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)
  • From WM@21:1/5 to All on Thu Dec 28 11:02:12 2023
    Le 28/12/2023 à 11:39, Jim Burns a écrit :

    | ∀n ∈ ℕ: 1/n - 1/(n+1) = d_n > 0
    makes no claim for 0
    because for no n ∈ ℕ is 1/n = 0

    That is true. But as soon as one unit fraction is encountered, it is
    followed by a gap consisting of points which are not unit fractions. Hence NUF(x) is a step function.

    According to ZFC the number of unit fractions grows between 0 and (0, 1]
    by more than 2.
    This is impossible because between any two unit fractions there are "uncountably" many points which do not fit between 0 and (0, 1]. This
    proves that ZFC does not deliver correct mathematics in this case, like in several others: https://www.researchgate.net/publication/376587391_The_seven_deadly_sins_of_set_theory

    Regards, WM

    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)
  • From Richard Damon@21:1/5 to All on Thu Dec 28 08:17:47 2023
    On 12/28/23 5:44 AM, WM wrote:
    Le 28/12/2023 à 03:57, Richard Damon a écrit :
    On 12/27/23 5:15 PM, WM wrote:
    Therefore the function NUF(x) has always a level between any two unit
    fractions. Therefore never infinitely many can be at the start together.

    But there can be infinitely many before any unit fraction,

    My logic does not allow that.


    Regards, WM



    Then you logic can't handle the problem, because they exist.

    You are just proving your logic isn't sufficient for the problem.

    It could be that you are just stuck on ZFC, and don't understand the
    limits of the number system it generates.

    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)
  • From Richard Damon@21:1/5 to All on Thu Dec 28 08:18:04 2023
    On 12/28/23 5:52 AM, WM wrote:
    Le 28/12/2023 à 03:58, Richard Damon a écrit :
    On 12/27/23 5:47 PM, WM wrote:

    Each discerned unit fraction has a distance from its next unit
    fractions and hence is not unboundedly dense.

    Then what is the BOUND?

    The bound is zero. The accumulation point contains dark unit fractions.

    But 0 is not a member of the set, so there is no bound in the set.

    If you accept 0 as the bound of the distance, then you accept that the
    points ARE allowed to "pile up" on each other, as there distance can go
    to 0.

    You can't use two different definitions of Bound in the same


    Remember, that distance reduces without any limit to its size, so you
    claim of not unboundedly dense is proved to be untrue.

    Every chosen unit fraction and its neighbours are not unboundedly dense.
    The dark unit fractions appear unboundedly dense.

    But Bounddness isn't a property of an individual number, but a
    collection of them. The Unboundedness of Unit Fractions approaching Zero
    means that there is no first one. Your "Darkness" is just an artifact of
    using improper logic that can't handle your set.


    All have distances and therefore NUF(x) can never grow from 0 to
    infinity without passing a first unit fraction and being constant
    afterwards for a while.


    Which only applies *IF* there IS a "first" Unit Fraction, which there
    isn't.

    If there are any unit fractions in separated oder, then there is a first
    one.

    Regards, WM

    Nope. Bounded Logic on an Unbounded set.

    Proves you to be too stupid for the problem.

    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)
  • From Richard Damon@21:1/5 to All on Thu Dec 28 08:21:18 2023
    On 12/28/23 6:02 AM, WM wrote:
    Le 28/12/2023 à 11:39, Jim Burns a écrit :

    | ∀n ∈ ℕ: 1/n - 1/(n+1) = d_n > 0
    makes no claim for 0
    because for no n ∈ ℕ is 1/n = 0

    That is true. But as soon as one unit fraction is encountered, it is
    followed by a gap consisting of points which are not unit fractions.
    Hence NUF(x) is a step function.

    According to ZFC the number of unit fractions grows between 0 and (0, 1]
    by more than 2.
    This is impossible because between any two unit fractions there are "uncountably" many points which do not fit between 0 and (0, 1]. This
    proves that ZFC does not deliver correct mathematics in this case, like
    in several others: https://www.researchgate.net/publication/376587391_The_seven_deadly_sins_of_set_theory

    Regards, WM




    But that logic presumse that there is a first unit fraction, which there
    isn't.

    Note, ZFC acknoleges that it is not a complete theory of even the Natual Numbers, but an attempt to get as close a possible with a "complete" theory.

    Your not recognizing this, shows your own misunderstanding.

    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)
  • From Jim Burns@21:1/5 to All on Thu Dec 28 13:17:03 2023
    On 12/28/2023 6:02 AM, WM wrote:
    Le 28/12/2023 à 11:39, Jim Burns a écrit :

    [...]

    According to ZFC

    No, ZFC doesn't say this:

    the number of unit fractions grows
    between 0 and (0, 1]
    by more than 2.

    According to ZFC (meaning "standardly")
    nothing visibleᵂᴹ and nothing darkᵂᴹ
    is between 0 and (0,1]

    Something between 0 and (0,1] is
    below (0,1] and also
    between 0 and 1, thus in (0,1]

    Below and in (0,1] is contradictory.
    ZFC says
    we don't need to seeᵂᴹ a contradictory thing
    in order to know it not.exists.

    ----
    | ∀n ∈ ℕ: 1/n - 1/(n+1) = d_n > 0
    makes no claim for 0
    because for no n ∈ ℕ is 1/n = 0

    That is true.
    But as soon as one unit fraction is encountered,
    it is followed by a gap consisting of
    points which are not unit fractions.
    Hence NUF(x) is a step function.

    A cardinality which can change by 1
    is
    a finite cardinality.

    A cardinality which can change by 1
    is followed by
    another cardinality which can change by 1.

    In the context of
    cardinalities which can change by 1,
    the last cardinality which can change by 1
    not.exists.


    A set with
    a (finite) cardinality which can change by 1
    can be ordered such that,
    for each of its non-empty splits F,H
    some last|first element exists in F|H
    and
    some first|last element exists in the set

    The least upper bound (ℕ) of
    sets (FISONs) orderable by successor with
    cardinalities which can change by 1
    does not contain
    a last element.

    The least upper bound (ℕ) of
    sets (FISONs) orderable by successor with
    cardinalities which can change by 1
    is not
    a set orderable by successor with
    a cardinality which can change by 1

    ℕ is a set orderable by successor.
    Its cardinality ℵ₀ can't change by 1.

    | ∀n ∈ ℕ: 1/n - 1/(n+1) = d_n > 0
    makes no claim for 0
    because for no n ∈ ℕ is 1/n = 0

    That is true.
    But as soon as one unit fraction is encountered,
    it is followed by a gap consisting of
    points which are not unit fractions.
    Hence NUF(x) is a step function.

    NUF(x) = |(-∞,x]∩⅟ℕ₁|

    Each element of (-∞,x]∩⅟ℕ₁ is not
    its lower end.
    Its lower end, visibleᵂᴹ or darkᵂᴹ
    not.exists.

    (-∞,x]∩⅟ℕ₁ cannot be ordered such that,
    for each of its non-empty splits F,H
    some last|first element exists in F|H
    and
    some first|last element exists in the set

    For x > 0
    NUF(x) = |(-∞,x]∩⅟ℕ₁| = ℵ₀ is infinite.
    ℵ₀ can't change by 1

    ----
    According to ZFC
    the number of unit fractions grows
    between 0 and (0, 1]
    by more than 2.
    This is impossible because
    between any two unit fractions there are
    "uncountably" many points which do not fit
    between 0 and (0, 1].
    This proves that ZFC does not deliver
    correct mathematics in this case,
    like in several others:

    What you (WM) tell people ZFC says
    is incorrect mathematics.
    However, you are an unreliable source
    for what ZFC says.

    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)
  • From Jim Burns@21:1/5 to All on Fri Dec 29 17:23:47 2023
    On 12/28/2023 6:02 AM, WM wrote:
    Le 28/12/2023 à 11:39, Jim Burns a écrit :

    [...]

    Hence NUF(x) is a step function.

    A cardinality.which.can.change.by.1
    is
    a finite cardinality.

    Changing (augmenting) by 1
    a cardinality.which.can.change.by.1
    yields a different (larger)
    cardinality.which.can.change.by.1.

    An upper.bound.which.can.change.by.1
    of cardinalities.which.can.change.by.1
    does not bound all
    cardinalities.which.can.change.by.1.
    It is a bound which is not a bound.
    It not.exists.

    An upper.bound
    of cardinalities.which.can.change.by.1
    is not
    an upper.bound.which.can.change.by.1
    of cardinalities.which.can.change.by.1.
    It cannot change by 1.

    Hence NUF(x) is a step function.

    For x > 0
    NUF(x) = |(-∞,x]∩⅟ℕ₁| = ℵ₀
    is an upper bound of
    cardinalities.which.can.change.by.1.
    NUF(x) cannot change by 1.

    Hence NUF(x) is not a step function.

    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)
  • From WM@21:1/5 to All on Sat Dec 30 12:18:03 2023
    Le 28/12/2023 à 19:17, Jim Burns a écrit :
    On 12/28/2023 6:02 AM, WM wrote:

    No, ZFC doesn't say this:

    the number of unit fractions grows
    between 0 and (0, 1]
    by more than 2.

    According to ZFC (meaning "standardly")
    nothing visibleᵂᴹ and nothing darkᵂᴹ
    is between 0 and (0,1]

    But NUF has collected ℵo unit fractions before any positive point. They
    do not come from Santa Claus.

    Something between 0 and (0,1] is
    below (0,1] and also
    between 0 and 1, thus in (0,1]

    Below and in (0,1] is contradictory.

    Therefore ℵo unit fractions come into being only after 0 within (0, 1],
    but not at a single point. Therefore ∀x ∈ (0, 1]: NUF(x) = ℵo is
    wrong.

    Regards, WM

    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)
  • From WM@21:1/5 to All on Sat Dec 30 14:03:35 2023
    Le 28/12/2023 à 14:17, Richard Damon a écrit :
    On 12/28/23 5:44 AM, WM wrote:
    Le 28/12/2023 à 03:57, Richard Damon a écrit :
    On 12/27/23 5:15 PM, WM wrote:
    Therefore the function NUF(x) has always a level between any two unit
    fractions. Therefore never infinitely many can be at the start together. >>
    But there can be infinitely many before any unit fraction,

    My logic does not allow that.

    It could be that you are just stuck on ZFC, and don't understand the
    limits of the number system it generates.

    I have set out to prove that ZFC can't handle this problem.

    Regards, WM

    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)
  • From Richard Damon@21:1/5 to All on Sat Dec 30 11:43:18 2023
    On 12/30/23 9:03 AM, WM wrote:
    Le 28/12/2023 à 14:17, Richard Damon a écrit :
    On 12/28/23 5:44 AM, WM wrote:
    Le 28/12/2023 à 03:57, Richard Damon a écrit :
    On 12/27/23 5:15 PM, WM wrote:
    Therefore the function NUF(x) has always a level between any two
    unit fractions. Therefore never infinitely many can be at the start
    together.

    But there can be infinitely many before any unit fraction,

    My logic does not allow that.

    It could be that you are just stuck on ZFC, and don't understand the
    limits of the number system it generates.

    I have set out to prove that ZFC can't handle this problem.

    Regards, WM



    Have you first shown that ZFC claims to handle this problem?

    From the ZFC generated Naturals Numbers, we just get the fact that
    there is no smallest unit fraction, and thus it would be expected that
    there would be infinitely many unit fractions below every unit fraction,
    and thus no point that NUF(x) would be a finite value.

    So, of course ZFC won't let you define an x that NUF(x) is 1.

    It seems the problem is realy with your problem, and it asking for
    something that is just impossible.

    As I have mentioned, there ARE number systems beyond the Natual /
    Rational / Real sets that you want to be in, that have numbers with some
    of the properties you are looking for. You even use one of the names for
    them, the "Transfinites", and in particular, the infintesimals, can have
    values sort of like what you are looking for in NUF(x), where there does
    exist a number x, such that NUF(x) could be 1 at it.

    I would need to look into more of the set theory to see if ZFC can
    generate this set with the appropriate base set of construction rules, I
    will admit that is beyond my studies.

    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)
  • From WM@21:1/5 to All on Sat Dec 30 17:00:03 2023
    Le 18/12/2023 à 16:00, Richard Damon a écrit :
    On 12/18/23 5:47 AM, WM wrote:

    You aren't using the right mapping, showing your ignorance.

    Where does my mapping deviate from Cantor's
    k = (m + n - 1)(m + n - 2)/2 + m
    with the result
    1/1, 1/2, 2/1, 1/3, 2/2, 3/1, 1/4, 2/3, 3/2, 4/1, 1/5, 2/4, 3/3, 4/2,
    5/1, 1/6, 2/5, 3/4, 4/3, 5/2, 6/1,  ...
    for the first time?


    Which shows that your matrix B having values like

    1/1 2/1 3/1 ...
    1/2 2/2 3/2 ...
    1/3 2/3 3/3 ...
    .
    .
    .

    is exactly indexed by a matrix form like:

    1 3 6 10
    2 5 9
    4 8 13
    7 12
    11


    This is a triangle, never a matrix.

    Regards, WM

    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)
  • From Richard Damon@21:1/5 to All on Sat Dec 30 12:33:53 2023
    On 12/30/23 12:00 PM, WM wrote:
    Le 18/12/2023 à 16:00, Richard Damon a écrit :
    On 12/18/23 5:47 AM, WM wrote:

    You aren't using the right mapping, showing your ignorance.

    Where does my mapping deviate from Cantor's
    k = (m + n - 1)(m + n - 2)/2 + m
    with the result
    1/1, 1/2, 2/1, 1/3, 2/2, 3/1, 1/4, 2/3, 3/2, 4/1, 1/5, 2/4, 3/3, 4/2,
    5/1, 1/6, 2/5, 3/4, 4/3, 5/2, 6/1,  ...
    for the first time?


    Which shows that your matrix B having values like

    1/1 2/1 3/1 ...
    1/2 2/2 3/2 ...
    1/3 2/3 3/3 ...
    .
    .
    .

    is exactly indexed by a matrix form like:

    1  3   6  10
    2  5   9
    4  8  13
    7  12
    11


    This is a triangle, never a matrix.

    WHat point in the matrix does it never cover?

    You just don't understand unbounded sets.



    Regards, WM


    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)
  • From WM@21:1/5 to All on Sat Dec 30 17:45:31 2023
    Le 30/12/2023 à 18:33, Richard Damon a écrit :
    On 12/30/23 12:00 PM, WM wrote:
    Le 18/12/2023 à 16:00, Richard Damon a écrit :
    On 12/18/23 5:47 AM, WM wrote:

    You aren't using the right mapping, showing your ignorance.

    Where does my mapping deviate from Cantor's
    k = (m + n - 1)(m + n - 2)/2 + m
    with the result
    1/1, 1/2, 2/1, 1/3, 2/2, 3/1, 1/4, 2/3, 3/2, 4/1, 1/5, 2/4, 3/3, 4/2,
    5/1, 1/6, 2/5, 3/4, 4/3, 5/2, 6/1,  ...
    for the first time?


    Which shows that your matrix B having values like

    1/1 2/1 3/1 ...
    1/2 2/2 3/2 ...
    1/3 2/3 3/3 ...
    .
    .
    .

    is exactly indexed by a matrix form like:

    1  3   6  10
    2  5   9
    4  8  13
    7  12
    11


    This is a triangle, never a matrix.

    WHat point in the matrix does it never cover?

    Almost all.

    XOOO...
    XOOO...
    XOOO...
    XOOO...
    ..

    Regards, WM

    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)
  • From Richard Damon@21:1/5 to All on Sat Dec 30 13:07:58 2023
    On 12/30/23 12:45 PM, WM wrote:
    Le 30/12/2023 à 18:33, Richard Damon a écrit :
    On 12/30/23 12:00 PM, WM wrote:
    Le 18/12/2023 à 16:00, Richard Damon a écrit :
    On 12/18/23 5:47 AM, WM wrote:

    You aren't using the right mapping, showing your ignorance.

    Where does my mapping deviate from Cantor's
    k = (m + n - 1)(m + n - 2)/2 + m
    with the result
    1/1, 1/2, 2/1, 1/3, 2/2, 3/1, 1/4, 2/3, 3/2, 4/1, 1/5, 2/4, 3/3,
    4/2, 5/1, 1/6, 2/5, 3/4, 4/3, 5/2, 6/1,  ...
    for the first time?


    Which shows that your matrix B having values like

    1/1 2/1 3/1 ...
    1/2 2/2 3/2 ...
    1/3 2/3 3/3 ...
    .
    .
    .

    is exactly indexed by a matrix form like:

    1  3   6  10
    2  5   9
    4  8  13
    7  12
    11


    This is a triangle, never a matrix.

    WHat point in the matrix does it never cover?

    Almost all.
    XOOO...
    XOOO...
    XOOO...
    XOOO...
    ..

    Regards, WM


    But your x's aren't the triangle, so your brain is obviously off.

    You are just showing that your logic just can't handle what you are
    trying to do, likely because you don't understand what it is actually doing.

    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)
  • From Jim Burns@21:1/5 to All on Sat Dec 30 23:42:01 2023
    On 12/30/2023 7:18 AM, WM wrote:
    Le 28/12/2023 à 19:17, Jim Burns a écrit :
    On 12/28/2023 6:02 AM, WM wrote:

    [...]

    No, ZFC doesn't say this:

    the number of unit fractions grows
    between 0 and (0, 1]
    by more than 2.

    According to ZFC (meaning "standardly")
    nothing visibleᵂᴹ and nothing darkᵂᴹ
    is between 0 and (0,1]

    But NUF has collected ℵo unit fractions
    before any positive point.
    They do not come from Santa Claus.

    A claim true of each of ℵ₀.many was
    a claim true of each of ℵ₀.many
    when that claim was made.

    We know that claims are
    true of each of ℵ₀.many
    when we both
    know what we describe is
    true of each of those ℵ₀.many
    and
    can see that each augmenting claim is
    not.first.false of each of those ℵ₀.many.

    Truly descriptive and visibly not.first.false
    are
    truly descriptive and visibly not.first.false
    whenever those claims are made.

    They do not come from Santa Claus.

    ℵ₀.many unit.fractions do not
    come from Santa Claus.
    They do not come from anyone or anywhere.
    They have already been here.

    But NUF has collected ℵo unit fractions
    before any positive point.

    A cardinal which.can.grow.by.1
    is a finite cardinal.

    For each cardinal which.can.grow.by.1
    a larger cardinal exists which.can.grow.by.1.

    Each cardinal which.can.grow.by.1
    is not an upper.bound of
    cardinals which.can.grow.by.1

    An upper.bound of
    cardinals which.can.grow.by.1
    is not
    a cardinal which.can.grow.by.q

    NUF(x) for x > 0
    is an upper.bound of
    cardinals which.can.grow.by.1
    which cannot grow by 1

    NUF(x) for x > 0
    cannot grow by 1

    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)
  • From WM@21:1/5 to All on Sun Dec 31 09:51:41 2023
    Le 31/12/2023 à 05:42, Jim Burns a écrit :

    ℵ₀.many unit.fractions do not
    come from Santa Claus.
    They do not come from anyone or anywhere.
    They have already been here.

    They are before any positive x but after 0 without any internal distances? Goodbye.

    Regards, WM

    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)
  • From WM@21:1/5 to All on Sun Dec 31 09:35:38 2023
    Le 30/12/2023 à 19:07, Richard Damon a écrit :
    On 12/30/23 12:45 PM, WM wrote:
    Le 30/12/2023 à 18:33, Richard Damon a écrit :
    On 12/30/23 12:00 PM, WM wrote:
    Le 18/12/2023 à 16:00, Richard Damon a écrit :
    On 12/18/23 5:47 AM, WM wrote:

    You aren't using the right mapping, showing your ignorance.

    Where does my mapping deviate from Cantor's
    k = (m + n - 1)(m + n - 2)/2 + m
    with the result
    1/1, 1/2, 2/1, 1/3, 2/2, 3/1, 1/4, 2/3, 3/2, 4/1, 1/5, 2/4, 3/3,
    4/2, 5/1, 1/6, 2/5, 3/4, 4/3, 5/2, 6/1,  ...
    for the first time?


    Which shows that your matrix B having values like

    1/1 2/1 3/1 ...
    1/2 2/2 3/2 ...
    1/3 2/3 3/3 ...
    .
    .
    .

    is exactly indexed by a matrix form like:

    1  3   6  10
    2  5   9
    4  8  13
    7  12
    11


    This is a triangle, never a matrix.

    WHat point in the matrix does it never cover?

    Almost all.
    XOOO...
    XOOO...
    XOOO...
    XOOO...
    ...

    But your x's aren't the triangle,

    They become the triangle, never more.

    Regards, WM

    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)
  • From Jim Burns@21:1/5 to All on Sun Dec 31 07:04:17 2023
    On 12/31/2023 4:51 AM, WM wrote:
    Le 31/12/2023 à 05:42, Jim Burns a écrit :

    ℵ₀.many unit.fractions do not
    come from Santa Claus.
    They do not come from anyone or anywhere.
    They have already been here.

    They are
    before any positive x but
    after 0
    without any internal distances?

    No.
    None is before each positive x.
    Each two have a positive distance.

    Your (WM's) quantifier magic is bad.


    For each cardinal n which.can.grow.by.1
    more than n.many unit.fractions are in (0,x]

    mₓ < ⅟x ≤ mₓ⁺¹ < ... < mₓ⁺ⁿ⁺¹

    0 < ⅟mₓ⁺ⁿ⁺¹ < ... < ⅟mₓ⁺¹ ≤ x

    For each cardinal n which.can.grow.by.1
    NUF(x) = |(-∞,x]∩⅟ℕ₁| is not n

    NUF(x) is
    not a cardinal which.can.grow.by.1

    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)
  • From Richard Damon@21:1/5 to All on Sun Dec 31 07:46:17 2023
    On 12/31/23 4:35 AM, WM wrote:
    Le 30/12/2023 à 19:07, Richard Damon a écrit :
    On 12/30/23 12:45 PM, WM wrote:
    Le 30/12/2023 à 18:33, Richard Damon a écrit :
    On 12/30/23 12:00 PM, WM wrote:
    Le 18/12/2023 à 16:00, Richard Damon a écrit :
    On 12/18/23 5:47 AM, WM wrote:

    You aren't using the right mapping, showing your ignorance.

    Where does my mapping deviate from Cantor's
    k = (m + n - 1)(m + n - 2)/2 + m
    with the result
    1/1, 1/2, 2/1, 1/3, 2/2, 3/1, 1/4, 2/3, 3/2, 4/1, 1/5, 2/4, 3/3, >>>>>>> 4/2, 5/1, 1/6, 2/5, 3/4, 4/3, 5/2, 6/1,  ...
    for the first time?


    Which shows that your matrix B having values like

    1/1 2/1 3/1 ...
    1/2 2/2 3/2 ...
    1/3 2/3 3/3 ...
    .
    .
    .

    is exactly indexed by a matrix form like:

    1  3   6  10
    2  5   9
    4  8  13
    7  12
    11


    This is a triangle, never a matrix.

    WHat point in the matrix does it never cover?

    Almost all.
    XOOO...
    XOOO...
    XOOO...
    XOOO...
    ...

    But your x's aren't the triangle,

    They become the triangle, never more.

    Regards, WM



    WHat point in the matrix can't they reach?

    You seem to like making claims you can't actually back up, because you
    don't understand the basic of infinite sets.


    Note, your citing you column and calling it the triangle shows your
    utter lack of understanding and need to use strawmen.

    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)
  • From Richard Damon@21:1/5 to All on Sun Dec 31 07:48:04 2023
    On 12/31/23 4:51 AM, WM wrote:
    Le 31/12/2023 à 05:42, Jim Burns a écrit :

    ℵ₀.many unit.fractions do not
    come from Santa Claus.
    They do not come from anyone or anywhere.
    They have already been here.

    They are before any positive x but after 0 without any internal distances? Goodbye.

    Regards, WM


    They are "bright", they can be used individually, but not as a
    collective whole. At least not when doing operations that must be finite
    (since the collective whole is not finite).

    This is as opposed to you, who is "dark" as you can not see what is true.

    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)
  • From WM@21:1/5 to All on Mon Jan 1 15:34:39 2024
    Le 31/12/2023 à 13:04, Jim Burns a écrit :
    On 12/31/2023 4:51 AM, WM wrote:
    Le 31/12/2023 à 05:42, Jim Burns a écrit :

    ℵ₀.many unit.fractions do not
    come from Santa Claus.
    They do not come from anyone or anywhere.
    They have already been here.

    They are
    before any positive x but
    after 0
    without any internal distances?

    No.
    None is before each positive x.

    Fine. The first unit fraction is a positive x itself. But you cannot find
    an x with less than ℵ₀ smaller unit fractions. They are dark.

    Regards, WM

    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)
  • From Jim Burns@21:1/5 to All on Mon Jan 1 12:11:13 2024
    On 1/1/2024 10:34 AM, WM wrote:
    Le 31/12/2023 à 13:04, Jim Burns a écrit :
    On 12/31/2023 4:51 AM, WM wrote:
    Le 31/12/2023 à 05:42, Jim Burns a écrit :

    ℵ₀.many unit.fractions do not
    come from Santa Claus.
    They do not come from anyone or anywhere.
    They have already been here.

    They are
    before any positive x but
    after 0
    without any internal distances?

    No.
    None is before each positive x.

    Fine.
    The first unit fraction is a positive x itself.

    The first unit fraction not.exists.
    A unit fraction before each x > 0 not.exists.

    You (WM) are using bad quantifier magic.

    But you cannot find

    We know not.exists...

    an x with less than ℵ₀ smaller unit fractions.
    They are dark.

    Be they visibleᵂᴹ or be they darkᵂᴹ,
    we know that
    "Each of those we discuss is like this"
    is true of each of those we discuss.

    Proof:
    _We can hear ourselves discuss them_

    We know that
    augmenting description with only
    not.first.false about each of those we discuss
    is
    augmenting with only
    true about each of those we discuss.

    Proof:
    A finite sequence of claims with
    a false claim
    has a first.false claim.

    Even if what we discuss is darkᵂᴹ,
    our discussion is visible.
    We know what it is we are saying.

    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)
  • From Jim Burns@21:1/5 to Jim Burns on Mon Jan 1 12:11:02 2024
    On 12/31/2023 7:04 AM, Jim Burns wrote:
    On 12/31/2023 4:51 AM, WM wrote:
    Le 31/12/2023 à 05:42, Jim Burns a écrit :

    ℵ₀.many unit.fractions do not
    come from Santa Claus.
    They do not come from anyone or anywhere.
    They have alreadcy been here.

    They are
    before any positive x but
    after 0
    without any internal distances?

    Santa Claus came to town.
    He brought shiny, new vocabulary.
    Let's play.

    A cardinal n which.can.change.by.1
    is
    a finite cardinal.

    A cardinal n which.can.change.by.1
    which changes by 1
    changes to
    another cardinal n⁺¹ or n⁻¹ which.
    .can.change.by.1

    A cardinal φ which, for any
    cardinal n which.can.change.by.1,
    is larger than n,
    is not n,
    is not any cardinal which.
    .can.change.by.1.

    φ is
    not one of those which.can.change.by.1.
    φ is infinite.

    For any cardinal |1,…,n| which.
    .can.change.by.1, that is,
    for which |1,…,n| < |1,…,n⁺¹|
    |1,…,n| < |1,…,n⁺¹,…|
    |1,…,n| ≠ |1,…,n⁺¹,…|

    |1,…,n⁺¹,…| = |ℕ| is
    not one of those which.can.change.by.1.
    |ℕ| is infinite.

    For any cardinal |m⁺¹,…,m⁺ⁿ| which.
    .can.change.by.1, that is,
    for which |m⁺¹,…,m⁺ⁿ| < |m⁺¹,…,m⁺ⁿ⁺¹|
    |m⁺¹,…,m⁺ⁿ| < |m⁺¹,…,m⁺ⁿ⁺¹,…|
    |m⁺¹,…,m⁺ⁿ| ≠ |m⁺¹,…,m⁺ⁿ⁺¹,…|

    |m⁺¹,…,m⁺ⁿ⁺¹,…| = |E(m⁺¹)| is
    not one of those which.can.change.by.1.
    |E(m⁺¹)| is infinite

    They are
    before any positive x but
    after 0
    without any internal distances?

    Their cardinal is larger than any of
    those which.can.change.by.1.

    Let mₓ < ⅟x ≤ mₓ⁺¹
    ⅟mₓ⁺¹ ≤ x < ⅟mₓ

    For any cardinal |⅟mₓ⁺ⁿ,…,⅟mₓ⁺¹| which.
    .can.change.by.1, that is,
    for which |⅟mₓ⁺ⁿ,…,⅟mₓ⁺¹| < |⅟mₓ⁺ⁿ⁺¹,…,⅟mₓ⁺¹|
    |⅟mₓ⁺ⁿ,…,⅟mₓ⁺¹| < |…,⅟mₓ⁺ⁿ⁺¹,…,⅟mₓ⁺¹| |⅟mₓ⁺ⁿ,…,⅟mₓ⁺¹| ≠ |…,⅟mₓ⁺ⁿ⁺¹,…,⅟mₓ⁺¹|

    |…,⅟mₓ⁺ⁿ⁺¹,…,⅟mₓ⁺¹| = |⅟ℕₓ| is
    not one of those which.can.change.by.1.
    |⅟ℕₓ| is infinite.

    No.
    None is before each positive x.
    Each two have a positive distance.

    Your (WM's) quantifier magic is bad.


    For each cardinal n which.can.grow.by.1
    more than n.many unit.fractions are in (0,x]

    mₓ < ⅟x ≤ mₓ⁺¹ < ... < mₓ⁺ⁿ⁺¹

    0 < ⅟mₓ⁺ⁿ⁺¹ < ... < ⅟mₓ⁺¹ ≤ x

    For each cardinal n which.can.grow.by.1
    NUF(x) = |(-∞,x]∩⅟ℕ₁| is not n

    NUF(x) is
    not a cardinal which.can.grow.by.1



    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)
  • From WM@21:1/5 to All on Mon Jan 1 17:21:20 2024
    Le 31/12/2023 à 13:46, Richard Damon a écrit :
    On 12/31/23 4:35 AM, WM wrote:
    Le 30/12/2023 à 19:07, Richard Damon a écrit :
    On 12/30/23 12:45 PM, WM wrote:
    Le 30/12/2023 à 18:33, Richard Damon a écrit :
    On 12/30/23 12:00 PM, WM wrote:
    Le 18/12/2023 à 16:00, Richard Damon a écrit :
    On 12/18/23 5:47 AM, WM wrote:

    You aren't using the right mapping, showing your ignorance.

    Where does my mapping deviate from Cantor's
    k = (m + n - 1)(m + n - 2)/2 + m
    with the result
    1/1, 1/2, 2/1, 1/3, 2/2, 3/1, 1/4, 2/3, 3/2, 4/1, 1/5, 2/4, 3/3, >>>>>>>> 4/2, 5/1, 1/6, 2/5, 3/4, 4/3, 5/2, 6/1,  ...
    for the first time?


    Which shows that your matrix B having values like

    1/1 2/1 3/1 ...
    1/2 2/2 3/2 ...
    1/3 2/3 3/3 ...
    .
    .
    .

    is exactly indexed by a matrix form like:

    1  3   6  10
    2  5   9
    4  8  13
    7  12
    11


    This is a triangle, never a matrix.

    WHat point in the matrix does it never cover?

    Almost all.
    XOOO...
    XOOO...
    XOOO...
    XOOO...
    ...

    But your x's aren't the triangle,

    They become the triangle, never more.

    WHat point in the matrix can't they reach?

    Such a point cannot be given. However, as you recognized yourself, the X
    from the first column cannot cover the whole matrix. The O will remain.

    You seem to like making claims you can't actually back up,

    I can prove them using mathematics: The o will remain. Therefore their positions exist. But they cannot be given. Conclusion: They are at dark positions.

    Regards, WM

    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)
  • From Richard Damon@21:1/5 to All on Mon Jan 1 13:19:24 2024
    On 1/1/24 12:21 PM, WM wrote:
    Le 31/12/2023 à 13:46, Richard Damon a écrit :
    On 12/31/23 4:35 AM, WM wrote:
    Le 30/12/2023 à 19:07, Richard Damon a écrit :
    On 12/30/23 12:45 PM, WM wrote:
    Le 30/12/2023 à 18:33, Richard Damon a écrit :
    On 12/30/23 12:00 PM, WM wrote:
    Le 18/12/2023 à 16:00, Richard Damon a écrit :
    On 12/18/23 5:47 AM, WM wrote:

    You aren't using the right mapping, showing your ignorance. >>>>>>>>>
    Where does my mapping deviate from Cantor's
    k = (m + n - 1)(m + n - 2)/2 + m
    with the result
    1/1, 1/2, 2/1, 1/3, 2/2, 3/1, 1/4, 2/3, 3/2, 4/1, 1/5, 2/4,
    3/3, 4/2, 5/1, 1/6, 2/5, 3/4, 4/3, 5/2, 6/1,  ...
    for the first time?


    Which shows that your matrix B having values like

    1/1 2/1 3/1 ...
    1/2 2/2 3/2 ...
    1/3 2/3 3/3 ...
    .
    .
    .

    is exactly indexed by a matrix form like:

    1  3   6  10
    2  5   9
    4  8  13
    7  12
    11


    This is a triangle, never a matrix.

    WHat point in the matrix does it never cover?

    Almost all.
    XOOO...
    XOOO...
    XOOO...
    XOOO...
    ...

    But your x's aren't the triangle,

    They become the triangle, never more.

    WHat point in the matrix can't they reach?

    Such a point cannot be given. However, as you recognized yourself, the X
    from the first column cannot cover the whole matrix. The O will remain.

    No O will remain, as some x from below will map to it


    You seem to like making claims you can't actually back up,

    I can prove them using mathematics: The o will remain. Therefore their positions exist. But they cannot be given. Conclusion: They are at dark positions.


    Then try to use MATHEMATICS, and not just broken logic

    All the locations of the O's can be given, they are just m,n and for
    every combination, a k exists that maps by the formula.

    Regards, WM


    You are just proving your stupidity.

    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)
  • From WM@21:1/5 to All on Tue Jan 2 19:35:52 2024
    Le 01/01/2024 à 19:19, Richard Damon a écrit :
    On 1/1/24 12:21 PM, WM wrote:

    WHat point in the matrix can't they reach?

    Such a point cannot be given. However, as you recognized yourself, the X
    from the first column cannot cover the whole matrix. The O will remain.

    No O will remain, as some x from below will map to it

    By definition of exchange of O and X never an O will leave the matrix.

    Regards, WM

    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)
  • From WM@21:1/5 to All on Tue Jan 2 19:30:14 2024
    Le 01/01/2024 à 18:11, Jim Burns a écrit :
    On 1/1/2024 10:34 AM, WM wrote:

    The first unit fraction is a positive x itself.

    The first unit fraction not.exists.

    The munit fractions are placed in linear order with distances and none is smaller than zero. That enforces a first one.

    Regards, WM

    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)
  • From Jim Burns@21:1/5 to All on Tue Jan 2 16:15:22 2024
    On 1/2/2024 2:30 PM, WM wrote:
    Le 01/01/2024 à 18:11, Jim Burns a écrit :
    On 1/1/2024 10:34 AM, WM wrote:

    The first unit fraction is a positive x itself.

    The first unit fraction not.exists.

    The [unit] fractions are placed in linear order
    with distances and none is smaller than zero.
    That enforces a first one.

    For each cardinal n which.can.change.by.1
    there are n⁺¹ unit.fractions
    |⅟mₓ⁺ⁿ⁺¹,…,⅟mₓ⁺¹|
    between 0 and x
    in linear order
    with only positive distances
    with none smaller than 0
    and
    which enforces a first one ⅟mₓ⁺ⁿ⁺¹

    In other words,
    for each cardinal n which.can.change.by.1
    the cardinal of unit fractions in (0,x] is
    more than n

    The cardinal of unit.fractions
    between 0 and x
    in linear order
    with only positive distances
    with none smaller than 0
    is
    not any cardinal which.can.change.by.1.
    and
    which does not enforce a first one.

    Each unit fraction ⅟n in (0,x] is
    not the first unit.fraction in (0,x]
    0 < ⅟n⁺¹ < ⅟n

    A first unit.fraction not.exists.

    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)
  • From Richard Damon@21:1/5 to All on Tue Jan 2 18:59:31 2024
    On 1/2/24 2:35 PM, WM wrote:
    Le 01/01/2024 à 19:19, Richard Damon a écrit :
    On 1/1/24 12:21 PM, WM wrote:

    WHat point in the matrix can't they reach?

    Such a point cannot be given. However, as you recognized yourself,
    the X from the first column cannot cover the whole matrix. The O will
    remain.

    No O will remain, as some x from below will map to it

    By definition of exchange of O and X never an O will leave the matrix.

    Regards, WM


    But we are not "swapping" them but mapping the X to its new position.

    Note also, every O that you move to the column with the X, will later be covered by an even later x, until the o's all disappear into infinity.

    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)
  • From WM@21:1/5 to All on Wed Jan 3 10:22:20 2024
    Le 02/01/2024 à 22:15, Jim Burns a écrit :
    On 1/2/2024 2:30 PM, WM wrote:

    The [unit] fractions are placed in linear order
    with distances and none is smaller than zero.
    That enforces a first one.

    For each cardinal n which.can.change.by.1

    For every unit fraction.

    Regards, WM

    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)
  • From WM@21:1/5 to All on Wed Jan 3 10:27:05 2024
    Le 03/01/2024 à 00:59, Richard Damon a écrit :
    On 1/2/24 2:35 PM, WM wrote:
    Le 01/01/2024 à 19:19, Richard Damon a écrit :
    On 1/1/24 12:21 PM, WM wrote:

    WHat point in the matrix can't they reach?

    Such a point cannot be given. However, as you recognized yourself,
    the X from the first column cannot cover the whole matrix. The O will
    remain.

    No O will remain, as some x from below will map to it

    By definition of exchange of O and X never an O will leave the matrix.

    But we are not "swapping" them but mapping the X to its new position

    and mapping the O to the position which the X has left.

    Note also, every O that you move to the column with the X, will later be covered by an even later x, until the o's all disappear into infinity.

    They do not disappear because by exchanging with an X from the matrix the
    O remains in the matrix.

    Regards, WM

    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)
  • From Richard Damon@21:1/5 to All on Wed Jan 3 07:08:21 2024
    On 1/3/24 5:27 AM, WM wrote:
    Le 03/01/2024 à 00:59, Richard Damon a écrit :
    On 1/2/24 2:35 PM, WM wrote:
    Le 01/01/2024 à 19:19, Richard Damon a écrit :
    On 1/1/24 12:21 PM, WM wrote:

    WHat point in the matrix can't they reach?

    Such a point cannot be given. However, as you recognized yourself,
    the X from the first column cannot cover the whole matrix. The O
    will remain.

    No O will remain, as some x from below will map to it

    By definition of exchange of O and X never an O will leave the matrix.

    But we are not "swapping" them but mapping the X to its new position

    and mapping the O to the position which the X has left.

    Why?


    Note also, every O that you move to the column with the X, will later
    be covered by an even later x, until the o's all disappear into infinity.

    They do not disappear because by exchanging with an X from the matrix
    the O remains in the matrix.

    Regards, WM



    And "exchanging" isn't the operation being performed, but a showing that
    there exists a correspondance between the members of the two sets.

    You are just showing that you don't understand what you are talking about.

    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)
  • From Jim Burns@21:1/5 to All on Wed Jan 3 06:28:08 2024
    On 1/3/2024 5:22 AM, WM wrote:
    Le 02/01/2024 à 22:15, Jim Burns a écrit :
    On 1/2/2024 2:30 PM, WM wrote:

    The [unit] fractions are placed in linear order
    with distances and none is smaller than zero.
    That enforces a first one.

    For each cardinal n which.can.change.by.1

    For every unit fraction.

    For each cardinal
    |⅟mₓ⁺ⁿ,…,⅟mₓ⁺¹| which.can.change.by.1
    there are unit.fractions
    |⅟mₓ⁺ⁿ⁺¹,…,⅟mₓ⁺¹|.many unit.fractions |⅟mₓ⁺ⁿ⁺¹,…,⅟mₓ⁺¹| = |⅟mₓ⁺ⁿ,…,⅟mₓ⁺¹|+1 between 0 and x
    in linear order
    with only positive distances
    with none smaller than 0
    and
    which enforces a first one ⅟mₓ⁺ⁿ⁺¹

    In other words,
    for each cardinal
    |⅟mₓ⁺ⁿ,…,⅟mₓ⁺¹| which.can.change.by.1
    the cardinal of unit fractions in (0,x] is
    more than |⅟mₓ⁺ⁿ,…,⅟mₓ⁺¹|

    The cardinal of unit.fractions
    between 0 and x
    in linear order
    with only positive distances
    with none smaller than 0
    is
    not any cardinal which.can.change.by.1.
    and
    which does not enforce a first one.

    Each unit fraction ⅟n in (0,x] is
    not the first unit.fraction in (0,x]
    0 < ⅟n⁺¹ < ⅟n

    A first unit.fraction not.exists.
    A first unit.fraction is not.enforced.

    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)
  • From WM@21:1/5 to All on Wed Jan 3 15:11:09 2024
    Le 03/01/2024 à 13:08, Richard Damon a écrit :
    On 1/3/24 5:27 AM, WM wrote:
    Le 03/01/2024 à 00:59, Richard Damon a écrit :
    On 1/2/24 2:35 PM, WM wrote:
    Le 01/01/2024 à 19:19, Richard Damon a écrit :
    On 1/1/24 12:21 PM, WM wrote:

    WHat point in the matrix can't they reach?

    Such a point cannot be given. However, as you recognized yourself, >>>>>> the X from the first column cannot cover the whole matrix. The O
    will remain.

    No O will remain, as some x from below will map to it

    By definition of exchange of O and X never an O will leave the matrix.

    But we are not "swapping" them but mapping the X to its new position

    and mapping the O to the position which the X has left.

    Why?

    Because there is no index remaining. This is indicated by O aka withOut
    indeX.

    And "exchanging" isn't the operation being performed,

    The game is defined so. You are not allowed to change it.

    Regards, WM

    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)
  • From WM@21:1/5 to All on Wed Jan 3 15:18:53 2024
    Le 03/01/2024 à 12:28, Jim Burns a écrit :
    On 1/3/2024 5:22 AM, WM wrote:
    Le 02/01/2024 à 22:15, Jim Burns a écrit :
    On 1/2/2024 2:30 PM, WM wrote:

    The [unit] fractions are placed in linear order
    with distances and none is smaller than zero.
    That enforces a first one.

    For each cardinal n which.can.change.by.1

    For every unit fraction.

    For each cardinal
    |⅟mₓ⁺ⁿ,…,⅟mₓ⁺¹| which.can.change.by.1

    For every unit fraction!

    EOD
    Regards, WM

    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)
  • From Jim Burns@21:1/5 to All on Wed Jan 3 12:59:00 2024
    On 1/3/2024 10:18 AM, WM wrote:
    Le 03/01/2024 à 12:28, Jim Burns a écrit :
    On 1/3/2024 5:22 AM, WM wrote:
    Le 02/01/2024 à 22:15, Jim Burns a écrit :
    On 1/2/2024 2:30 PM, WM wrote:

    The [unit] fractions are placed in linear order
    with distances and none is smaller than zero.
    That enforces a first one.

    For each cardinal n which.can.change.by.1

    For every unit fraction.

    For each cardinal
    |⅟mₓ⁺ⁿ,…,⅟mₓ⁺¹| which.can.change.by.1

    For every unit fraction!

    WM:
    The [unit] fractions are placed in linear order
    with distances and none is smaller than zero.
    That enforces a first one.

    A cardinal larger than
    each cardinal which.can.grow.by.1
    is not
    any cardinal which.can.grow.by.1.
    and is
    a cardinal which.CANNOT.grow.by.1.

    A cardinal which.CANNOT.grow.by.1
    DOES NOT enforce a first one.

    Yes.
    There exists a sequence of length n⁺¹
    of visibleᵂᴹ unit fractions in (0,x]
    in linear order
    with positive distances from each other
    each larger than 0
    enforcing a first one within that sequence,

    Such a sequence exists for
    each positive point x and
    each cardinal n which.can.change.by.1

    However.
    For each positive point x the cardinal of
    visibleᵂᴹ unit.fractions in (0,x]
    is not
    any cardinal n which.can.change.by.1
    and a first one is not enforced.

    It is not a contradiction to claim that,
    for each visibleᵂᴹ unit.fraction ⅟nₓ in (0,x]
    ⅟nₓ is not first visibleᵂᴹ unit.fraction in (0,x]
    It is not a contradiction to claim that
    0 < ⅟nₓ⁺¹ < ⅟nₓ ≤ x

    We see that the cardinal of
    visibleᵂᴹ unit.fractions in (0,x]
    does not enforce a first one.

    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)
  • From Richard Damon@21:1/5 to All on Wed Jan 3 19:35:19 2024
    On 1/3/24 10:11 AM, WM wrote:
    Le 03/01/2024 à 13:08, Richard Damon a écrit :
    On 1/3/24 5:27 AM, WM wrote:
    Le 03/01/2024 à 00:59, Richard Damon a écrit :
    On 1/2/24 2:35 PM, WM wrote:
    Le 01/01/2024 à 19:19, Richard Damon a écrit :
    On 1/1/24 12:21 PM, WM wrote:

    WHat point in the matrix can't they reach?

    Such a point cannot be given. However, as you recognized
    yourself, the X from the first column cannot cover the whole
    matrix. The O will remain.

    No O will remain, as some x from below will map to it

    By definition of exchange of O and X never an O will leave the matrix. >>>
    But we are not "swapping" them but mapping the X to its new position

    and mapping the O to the position which the X has left.

    Why?

    Because there is no index remaining. This is indicated by O aka withOut indeX.

    Why isn't the index remaining?

    What used it up?

    Remember, we are using the k's to mark off the m,n, that doesn't make
    the k go away.


    And "exchanging" isn't the operation being performed,

    The game is defined so. You are not allowed to change it.

    Regards, WM


    But you are playing the wrong game.

    It doesn't matter if you win the game that doesn't count.

    Cantor played a different game, which it seems you don't even know how
    to play.

    So, you lose.

    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)
  • From WM@21:1/5 to All on Thu Jan 4 10:39:42 2024
    Le 04/01/2024 à 01:35, Richard Damon a écrit :
    On 1/3/24 10:11 AM, WM wrote:
    Le 03/01/2024 à 13:08, Richard Damon a écrit :


    By definition of exchange of O and X never an O will leave the matrix. >>>>
    But we are not "swapping" them but mapping the X to its new position

    and mapping the O to the position which the X has left.

    Why?

    Because there is no index remaining. This is indicated by O aka withOut
    indeX.

    Why isn't the index remaining?

    The index is placed according to Cant6or's formula. It must leave its
    initial position which is now without index.

    What used it up?

    Remember, we are using the k's to mark off the m,n, that doesn't make
    the k go away.

    They go away from their initial positions in the first column.


    And "exchanging" isn't the operation being performed,

    The game is defined so. You are not allowed to change it.

    But you are playing the wrong game.

    No, you are not understanding it.

    Cantor played a different game, which it seems you don't even know how
    to play.

    I repeat exactly his game:
    k = (m + n - 1)(m + n - 2)/2 + m
    The k come from the first column

    1, 1/2, 1/3, 1/4, ...
    2, 2/2, 2/3, 2/4, ...
    3, 3/2, 3/3, 3/4, ...
    4, 4/2, 4/3, 4/4, ...
    5, 5/2, 5/3, 5/4, ...
    ..

    and go to
    1/1, 1/2, 2/1, 1/3, 2/2, 3/1, 1/4, 2/3, 3/2, 4/1, 1/5, 2/4, 3/3, 4/2, 5/1,
    1/6, 2/5, 3/4, 4/3, 5/2, 6/1, ...

    XOOO... XXOO... XXOO... XXXO... ... XXXX...
    XOOO... OOOO... XOOO... XOOO... ... XXXX...
    XOOO... XOOO... OOOO... OOOO... ... XXXX...
    XOOO... XOOO... XOOO... OOOO... ... XXXX... ..........................................

    Regards, WM

    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)
  • From Richard Damon@21:1/5 to All on Thu Jan 4 07:26:43 2024
    On 1/4/24 5:39 AM, WM wrote:
    Le 04/01/2024 à 01:35, Richard Damon a écrit :
    On 1/3/24 10:11 AM, WM wrote:
    Le 03/01/2024 à 13:08, Richard Damon a écrit :


    By definition of exchange of O and X never an O will leave the
    matrix.

    But we are not "swapping" them but mapping the X to its new position >>>>>
    and mapping the O to the position which the X has left.

    Why?

    Because there is no index remaining. This is indicated by O aka
    withOut indeX.

    Why isn't the index remaining?

    The index is placed according to Cant6or's formula. It must leave its
    initial position which is now without index.

    WHY?

    You don't understand what the operation defined is.


    What used it up?

    Remember, we are using the k's to mark off the m,n, that doesn't make
    the k go away.

    They go away from their initial positions in the first column.

    Nope.



    And "exchanging" isn't the operation being performed,

    The game is defined so. You are not allowed to change it.

    But you are playing the wrong game.

    No, you are not understanding it.

    You can't win the game if you aren't playing the right game.


    Cantor played a different game, which it seems you don't even know how
    to play.

    I repeat exactly his game:
    k = (m + n - 1)(m + n - 2)/2 + m
    The k come from the first column

    1, 1/2, 1/3, 1/4, ...
    2, 2/2, 2/3, 2/4, ...
    3, 3/2, 3/3, 3/4, ...
    4, 4/2, 4/3, 4/4, ...
    5, 5/2, 5/3, 5/4, ...
    ..

    and go to
    1/1, 1/2, 2/1, 1/3, 2/2, 3/1, 1/4, 2/3, 3/2, 4/1, 1/5, 2/4, 3/3, 4/2,
    5/1, 1/6, 2/5, 3/4, 4/3, 5/2, 6/1,  ...

    XOOO... XXOO... XXOO... XXXO... ... XXXX...
    XOOO... OOOO... XOOO... XOOO... ... XXXX...
    XOOO... XOOO... OOOO... OOOO... ... XXXX...
    XOOO... XOOO... XOOO... OOOO... ... XXXX... ..........................................

    Regards, WM






    And we show that every element of the "matrix" has a matching point in
    the list of Natural Numbers, (and vis-versa) therefore the sets are BY DEFINITION, the same size.

    PERIOD.

    You are just proving your stupidity.

    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)
  • From WM@21:1/5 to All on Thu Jan 4 14:45:52 2024
    Le 03/01/2024 à 18:59, Jim Burns a écrit :

    A cardinal larger than
    each cardinal which.can.grow.by.1
    is not
    any cardinal which.can.grow.by.1.
    and is
    a cardinal which.CANNOT.grow.by.1.

    Every unit fraction obeys maths. It is useless to talk to people who deny
    this. EOD

    Regards, WM

    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)
  • From WM@21:1/5 to All on Thu Jan 4 14:57:21 2024
    Le 04/01/2024 à 13:26, Richard Damon a écrit :
    On 1/4/24 5:39 AM, WM wrote:
    Le 04/01/2024 à 01:35, Richard Damon a écrit :
    On 1/3/24 10:11 AM, WM wrote:
    Le 03/01/2024 à 13:08, Richard Damon a écrit :


    By definition of exchange of O and X never an O will leave the >>>>>>>> matrix.

    But we are not "swapping" them but mapping the X to its new position >>>>>>
    and mapping the O to the position which the X has left.

    Why?

    Because there is no index remaining. This is indicated by O aka
    withOut indeX.

    Why isn't the index remaining?

    The index is placed according to Cant6or's formula. It must leave its
    initial position which is now without index.

    WHY?

    You don't understand what the operation defined is.


    What used it up?

    Remember, we are using the k's to mark off the m,n, that doesn't make
    the k go away.

    They go away from their initial positions in the first column.

    Nope.

    Yes, they do.

    And "exchanging" isn't the operation being performed,

    The game is defined so. You are not allowed to change it.

    But you are playing the wrong game.

    No, you are not understanding it.

    You can't win the game if you aren't playing the right game.

    I do the the game exactly according to the rules given by Cantor with one exception: First I use a bijection between ℕ and the first column of

    1/1, 1/2, 1/3, 1/4, ...
    2/1, 2/2, 2/3, 2/4, ...
    3/1, 3/2, 3/3, 3/4, ...
    4/1, 4/2, 4/3, 4/4, ...
    5/1, 5/2, 5/3, 5/4, ...
    ..

    to get

    1, 1/2, 1/3, 1/4, ...
    2, 2/2, 2/3, 2/4, ...
    3, 3/2, 3/3, 3/4, ...
    4, 4/2, 4/3, 4/4, ...
    5, 5/2, 5/3, 5/4, ...
    ..

    If however that destroys Cantor's game, then it has never worked.

    Cantor played a different game, which it seems you don't even know how
    to play.

    I repeat exactly his game:
    k = (m + n - 1)(m + n - 2)/2 + m
    The k come from the first column

    1, 1/2, 1/3, 1/4, ...
    2, 2/2, 2/3, 2/4, ...
    3, 3/2, 3/3, 3/4, ...
    4, 4/2, 4/3, 4/4, ...
    5, 5/2, 5/3, 5/4, ...
    ..

    and go to
    1/1, 1/2, 2/1, 1/3, 2/2, 3/1, 1/4, 2/3, 3/2, 4/1, 1/5, 2/4, 3/3, 4/2,
    5/1, 1/6, 2/5, 3/4, 4/3, 5/2, 6/1,  ...

    XOOO... XXOO... XXOO... XXXO... ... XXXX...
    XOOO... OOOO... XOOO... XOOO... ... XXXX...
    XOOO... XOOO... OOOO... OOOO... ... XXXX...
    XOOO... XOOO... XOOO... OOOO... ... XXXX...
    ..........................................


    And we show that every element of the "matrix" has a matching point in
    the list of Natural Numbers,

    No, You show that all known natural numbers match all known unit
    fractions. Therefore they appear to be in bijection. Only a very, very
    stupid mind can believe that there are as many natural numbers divisible
    by 10^10^1000 as fractions or even as algebraic numbers.

    But since you do not even understand the basics of this discussion, it is useless to continue. EOD

    Regards, WM

    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)
  • From Jim Burns@21:1/5 to All on Thu Jan 4 15:51:07 2024
    ⁺¹
    On 1/4/2024 9:45 AM, WM wrote:
    Le 03/01/2024 à 18:59, Jim Burns a écrit :

    A cardinal larger than
    each cardinal which.can.grow.by.1
    is not
    any cardinal which.can.grow.by.1.
    and is
    a cardinal which.CANNOT.grow.by.1.

    Every unit fraction obeys maths.

    No unit fraction obeys a contradiction.

    Math claims that
    a set larger than each ⁺¹.able.cardinal
    does not itself have a ⁺¹.able.cardinal.

    For each ⁺¹.able |⅟m,…,⅟1|
    |⅟m,…,⅟1| < |⅟n⁺ᵐ⁺¹,…,⅟n⁺¹| ≤ |…,⅟n⁺²,⅟n⁺¹| and
    |…,⅟n⁺²,⅟n⁺¹| is not.⁺¹.able

     not.⁺¹.able       ⁺¹.able |…,⅟n⁺²,⅟n⁺¹|   |⅟n,⅟n⁻¹,…,⅟1|        =               > |…,⅟n⁺²,⅟n⁺¹,⅟n|   |⅟n⁻¹,…,⅟1|

    Pseudo.induction (inductionᵂᴹ) claims that
    a set larger than each ⁺¹.able.cardinal
    itself has a ⁺¹.able.cardinal.

    Pseudo.induction (inductionᵂᴹ) is a contradiction.
    The unit fractions obey math,
    not pseudo.induction (inductionᵂᴹ).

    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)
  • From Richard Damon@21:1/5 to All on Thu Jan 4 19:58:20 2024
    On 1/4/24 9:57 AM, WM wrote:
    Le 04/01/2024 à 13:26, Richard Damon a écrit :
    On 1/4/24 5:39 AM, WM wrote:
    Le 04/01/2024 à 01:35, Richard Damon a écrit :
    On 1/3/24 10:11 AM, WM wrote:
    Le 03/01/2024 à 13:08, Richard Damon a écrit :


    By definition of exchange of O and X never an O will leave the >>>>>>>>> matrix.

    But we are not "swapping" them but mapping the X to its new
    position

    and mapping the O to the position which the X has left.

    Why?

    Because there is no index remaining. This is indicated by O aka
    withOut indeX.

    Why isn't the index remaining?

    The index is placed according to Cant6or's formula. It must leave its
    initial position which is now without index.

    WHY?

    You don't understand what the operation defined is.


    What used it up?

    Remember, we are using the k's to mark off the m,n, that doesn't
    make the k go away.

    They go away from their initial positions in the first column.

    Nope.

    Yes, they do.

    Nope, not if you are trying to follow Cantor.


    And "exchanging" isn't the operation being performed,

    The game is defined so. You are not allowed to change it.

    But you are playing the wrong game.

    No, you are not understanding it.

    You can't win the game if you aren't playing the right game.

    I do the the game exactly according to the rules given by Cantor with
    one exception: First I use a bijection between ℕ and the first column of

    1/1, 1/2, 1/3, 1/4, ...
    2/1, 2/2, 2/3, 2/4, ...
    3/1, 3/2, 3/3, 3/4, ...
    4/1, 4/2, 4/3, 4/4, ...
    5/1, 5/2, 5/3, 5/4, ...
    ..

    to get

    1, 1/2, 1/3, 1/4, ...
    2, 2/2, 2/3, 2/4, ...
    3, 3/2, 3/3, 3/4, ...
    4, 4/2, 4/3, 4/4, ...
    5, 5/2, 5/3, 5/4, ...
    ..

    If however that destroys Cantor's game, then it has never worked.

    So, you don't understand Cantor's game.

    He pointed out that there were many ways to do it wrong, but the key was
    that if you do it right, you show that matching



    Cantor played a different game, which it seems you don't even know
    how to play.

    I repeat exactly his game:
    k = (m + n - 1)(m + n - 2)/2 + m
    The k come from the first column

    1, 1/2, 1/3, 1/4, ...
    2, 2/2, 2/3, 2/4, ...
    3, 3/2, 3/3, 3/4, ...
    4, 4/2, 4/3, 4/4, ...
    5, 5/2, 5/3, 5/4, ...
    ..

    and go to
    1/1, 1/2, 2/1, 1/3, 2/2, 3/1, 1/4, 2/3, 3/2, 4/1, 1/5, 2/4, 3/3, 4/2,
    5/1, 1/6, 2/5, 3/4, 4/3, 5/2, 6/1,  ...

    XOOO... XXOO... XXOO... XXXO... ... XXXX...
    XOOO... OOOO... XOOO... XOOO... ... XXXX...
    XOOO... XOOO... OOOO... OOOO... ... XXXX...
    XOOO... XOOO... XOOO... OOOO... ... XXXX...
    ..........................................


    And we show that every element of the "matrix" has a matching point in
    the list of Natural Numbers,

    No, You show that all known natural numbers match all known unit
    fractions. Therefore they appear to be in bijection. Only a very, very
    stupid mind can believe that there are as many natural numbers divisible
    by 10^10^1000 as fractions or even as algebraic numbers.

    Yep, he proves it.

    Just shows how unintuative infinity is.


    But since you do not even understand the basics of this discussion, it
    is useless to continue. EOD

    Regards, WM



    You are just proving the point.

    You are too stupid to understand the actual logic, and have FAILED.

    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)