« first day (2504 days earlier)      last day (2813 days later) » 

04:00
Point in that picture being that the angle $\phi$ is measured from the vertical, so the adjacent side for that is vertical as well.
So i might have to remember that in the first and third quadrant
However, if you draw your angle the same way as in the unit circle picture, then x=cos(theta) and y=sin(theta) is correct.
then it is equal to 1/2, sqrt(3)/2
Careful. In the third quadrant, both terms are negative.
and in the 2nd and 4th it's sqrt(3)/2,1/2
right, I am ignoring that at the moment
for simplicity
04:01
Fair.
to explain myself more quickly
@Semiclassical Are you good with surfaces stuff?
Basically: If you a point reflect across one of the axes, the sign but not the value can change.
Same set of Eigenvalues of two matrices doesnot mean that the two matrices are same, as for example the similar matrices have same set of eigen values but the similar matrices may not be equal,is this correct?
I wouldn't say so. @TheGreatDuck
04:02
in the first quadrant is is +,+ in the second it is -,+ in the third it is -,- and in the fourth it is +,-
which is simple to remember
It's handy, yeah.
well, I mean
So a unit circle is good to draw even if you're not trying to make things totally precise.
04:03
I just look at it like I would any graph
@LeakyNun i wanna say that $a^+ \in P(a) $
x is always positive on the right
and y is always positive on the top
@Faust7 can you justify that?
"of course it's true, it's so simple"
04:03
anyways
I think I'll study a bit more tomorrow
thanks, semi.
sounds like a plan.
glad to be of help
@LeakyNun well im fairly certain that $\{a\} \in P(a) $ im just not sure about a
@Faust7 can you justify that?
"fairly certain" isn't a proof
@LeakyNun well the P(a) is the set of all subsets of a and a is a subset of a. so im going with a being an element of the power set of a
that isn't what you said
04:07
yeah im lost
think i need to approach it with a diffrent axiom was just a thought
you said $\{a\} \in P(a)$
and you just proved that $a \in P(a)$.
yeah but im less sure about $\{a\}$ being in P(a) after proving that a was
alright.
@LeakyNun i think that $\{a\}$ also has to be in the P(a) from the definition
@Faust7 why?
@Faust7 sidenote: are you shocked by this level of formality?
04:14
well is i let $b= \{1,2\}$ then $\{\{1,2}\} $ is in the P(b)
uh well i have taken classes that have brushed on analysis in calculus on R and C but really the only formality i have is from Algerbra classes so im sturggle alittle formulating the arguments
i know what i want to say but everytime i write it out it looks technically wrong
@Faust7 I see.
@Faust7 justify it.
its gotta be wrong
$\{a\} $ isnt in the power set of a
the set a is
but not the set of the set a
correct.
@Faust7 any idea?
Yeah im trying to put it togetehr with The axiom of pairing
one sec
let a^+ = a \cup \{a\} where a is some set is we apply the axiom of pairing is we set a=x=y
then z=a^+
yeah this will work ^^
@LeakyNun for 1 use the axiom of pairing. which states \forall x \forall y \exists z(x\in z and y\in z) let a=x=y then z = \{a\} \cup a = a^+
therefore $a^+$ exists
04:33
explain "let a=x=y then z = \{a\} \cup a = a^+"
First assume that a is a set
then we can define the set x to equal the set a
and define the set y to equal the set a
wait thats not enough :(
what i stated would mean that$ z= \{a\}$
but we know that a is a set and z is a set
so by the axiom of extensionality
shit no
by the axiom of pairing
we know that there exists a set a^+ that contains the set z (z= \{a\}) and the set a
im sorry
im tired thats still wrong
ok...
im fairly certain i found the set \{a\}
and i know i have the set a
and i want the union not a set that has those 2 sets as elements
so i want the axiom of union
then i want to claim that is the set a^+
that make sense?
04:41
wonderful.
sorry that took me so long >.>
it's alright. it's your first time.
for 2) can i define the set a =\{ 1,\{b\} \} then use the axiom of regularity?
you can't say what a is.
where b is non empty and doesnt have 1
ah
04:44
a is any set
lets assume for all non-empty sets a that no such b exists
that isn't the negation.
dam :(
are a and b non-empty?
a is non-empty as I have said.
I have said nothing about b.
im kind of confused is a is a non empty set
04:50
a is a non-empty set.
then the axiom of regulaity says that a contains a member y such that a and y are disjiont sets
cant i just take y=b ?
exactly
but there's a simpler solution
hmm
can you just let b= the nullset?
04:52
exactly
lol
question 3 i need to use the Axiom schema of specification
its the set ofall sets paradox
can i just show that the cardianlity of x and \{x\} isnt the same?
what is the cardinality of x?
hopefully not 1
>.>
mathematics is not built on hope.
lol
very true ima get some advil for my headache n give it a whirl ( thanks for being patient)
05:02
circle of imaginary radius anyone?
link to hyperbolic rotations?
@bolbteppa $\mu = i$, I hope.
omg im an idiot
@Faust7 this needs to stop.
sorry
@LeakyNun Thanks for the information, sorry for asking so many questions ^^
@Faust7 I appreciate people asking questions.
I don't like those who learn without asking. They learn not.
@Faust7 any idea for question 3?
05:11
@LeakyNun The solution has to come form the axiom of regularity
but i dont know how to sya it.
Communicating is a strong point for me i dont really know how to ask a question im stuck and have no idea how to move forward
not*
@Faust7 try to apply the axiom of regularity.
i dunno it doesnt make sense it feels like the statement is saying that $\{x\} \in x$
i just dont know why i think that
it's right but you need to justify it.
@Faust7 any idea?
05:28
i wanna but x and \{x\} in one set
then use the theorem of regularity to show that both are non
you're overcomplicating everything
like i want a set where x and \{x\} are both an element of it
then i want to use the theorem of regularity to show that there both elements of the set
that's what you just said
contradicting it
where's the contradiction?
05:30
im sorry i cant write it out logically
i think this material is allittle too beyond me
i think its time i gave up ^^
Thanks for all the help though =)
no problem
Well, as it turns out, I have 4 typos in my definitions (not just 2), fixing them all now

though I seriously think chopping off the front of the supremum series so it starts at something larger than omega may not really work
For starters, by defintion of epsilon numbers, they will be epsilon numbers, but it is possible the index that name them will become a bit weird (if the rules can be fixed to make them exist)
@Secret do we agree that ${}^{\omega+1}\omega = \varepsilon_0$?
05:45
It will (Probably?) no longer once I fixed the limiting case in the definition in the ordinal hyoperation

My current attempt is to force the limiting case for any ordinal ${}^{\lambda}\omega$ where $\lambda > \omega$ to have only supremum series $\sup ({}^{\beta}\omega|\omega<\beta<\lambda)$
(in essence I chop off the first omega terms of the series to avoid the roadblock
@Secret $\omega+1$ is not a limiting ordinal
and there is no number between $\omega$ and $\omega+1$
I mean, orignally the series will look like:
$\omega,{}^2\omega,{}^3\omega,...,{}^n\omega,{}^{n+1}\omega,...$
and we all know that once we take $n <\omega$ we will go to $\epsilon_0$
My current approach is to make the series to do as follows:
I think you mean $n \ge \omega$.
${}^{\omega+1}\omega,{}^{\omega+2}\omega,{}^{\omega+3}\omega,...,{\omega+n}^n\omega,{}^{\omega+(n+1)}\omega,...$

that is, we don't allow any ordinals of the form ${}^{\lambda}\omega$ to have a series to start at terms $<{}^{\omega+1}\omega$.

Finally, to plug in the hole and complete the introduction of the discontinuity, I define ${}^{\omega+1}\omega$ as a fixed point of the exponenetial map $x\mapsto\omega^{x}$ and using the modified hyperopations, I can show it is not equal to $\epsilon_0$
in other words, ${}^{\omega+1}\omega = \varepsilon_1$?
05:55
It will not be equal to $\epsilon_1$ either and I can show it
Can you show me how ${}^{\omega+1}\omega$ is defined? @Secret
${}^{\omega+1}\omega$ satisfies $\omega^{\alpha}=\alpha$ and is defined to be the only ordinal that can only be reached by the constant series ${}^{\omega+1}\omega,{}^{\omega+1}\omega,{}^{\omega+1}\omega,{}^{\omega+1}\omega...$

Therefore if you construct any other series, it will lead to $\epsilon_0$, as expected
I don't see it as a definition.
It isn't specific enough. And the definition cannot contain itself.
SBM
SBM
hello
Definitions must be based on pre-existing concepts. @Secret
SBM
SBM
06:03
oh
I advise you to go through the formal definition of negative integers. @Secret
negative integers is defined to be a number a and integer $b\in \Bbb{N}$ such that a+b=0
@Secret that's exactly where the mistake is.
You can't just define a number to satisfy certain equations.
You need to use what is already defined.
but isn't that's how negative integers are defined, as additive inverses of the positive integers?
No, but that's the motivation behind the concept.
Care must be taken so as not to confuse definitions with proof of existence.
@Secret If you define it as such, you're going to need an axiom for its existence.
And you would also need to tweak the definition of addition to make it compatible with the negative numbers.
06:08
> Formal construction of negative integers[edit]
See also: Integer § Construction
In a similar manner to rational numbers, we can extend the natural numbers N to the integers Z by defining integers as an ordered pair of natural numbers (a, b). We can extend addition and multiplication to these pairs with the following rules:

(a, b) + (c, d) = (a + c, b + d)
(a, b) × (c, d) = (a × c + b × d, a × d + b × c)
We define an equivalence relation ~ upon these pairs with the following rule:

(a, b) ~ (c, d) if and only if a + d = b + c.
So you mean a construction is not the same as a definition?
I consider the construction to be the definition.
4 mins ago, by Secret
negative integers is defined to be a number a and integer $b\in \Bbb{N}$ such that a+b=0
And this is informally a definition, formally a motivation.
Or you can call it the informal definition.
Just a small check, is $\epsilon_1$ defined to be the 1st fixed point of the equation $\omega^{\alpha}=\alpha$ or is there something else I need?
the second.
that is, if there is no $\epsilon_0$ then $\epsilon_1$ will become undefined?
but we can show that $\varepsilon_0$ exists.
We can also show that $\varepsilon_1$ exists.
we have an algorithm to construct $\varepsilon_n$.
the proof of its validity is not what I would go through...
06:14
ok
You can view $\varepsilon_1 = {}^\omega \varepsilon_0$ as a definition, or a construction.
You would need to show that there is no fixed point of $\omega^a = a$ between $\varepsilon_0$ and $\varepsilon_1$.
Point is, you can't just define things.
@Secret do we agree?
ok, so I need to establish the existence of something in order to define it
rather, you need to establish the existence of something in addition to defining it.
06:39
@Secret are you typing?
kinda
$\textbf{Ver 7}$
define ordinal hyperoperation as follows:

Let $\alpha,\beta,\gamma,\delta,\lambda,m \in \textrm{On}$ where $\lambda$ is a limit ordinal and the rest are arbitrary. Then:

Associativity
\begin{align}
m=\{0,1,2\}: \alpha [m] (\beta [m] \gamma) & = (\alpha [m] \beta) [m] \gamma\\
\end{align}

Illustration:
\begin{align}
\alpha [0] \beta = \alpha^+\\
\alpha [1] \beta = \alpha+\beta\\
\alpha [2] \beta = \alpha\beta\\
\alpha [3] \beta = \alpha^{\beta}\\
\alpha [4] \beta = {}^{\beta}\alpha\\
$m<3,\alpha < \beta: \alpha [m] \beta = \beta$ is false.
ah yes, forgot $\omega(\omega+1)=\omega^2+\omega$ oops
A simpler counterexample would involve $\alpha = 1$ and $\beta = 2$.
> for all $\alpha < \beta$, $\alpha + \beta = \beta$ if and only if $\beta = \omega^\delta$ for some $\delta$.
I thought $\alpha = 1$ and $\beta = \omega + 1$ would also satisfy the equation...
so my link is wrong?
well $\omega+1 = \omega^1+\omega^0$
and plus is associative
thus $\omega$ will eat the one, leaving $\omega+1$
06:52
yes, so I'm saying that the link is wrong.
07:05
[Cont.]
\begin{align}
\epsilon_0 & =\omega[4]\omega = \omega [3]^{\omega}\omega = a_{\omega,3}(\omega) = a_{0,4}(\omega)\\
\omega[4](\omega+1) & = \omega [3]^{\omega+1} \omega = \omega [3]^{\omega} (\omega [3] \omega) = a_{\omega,3}(\omega^{\omega}) = a_{0,4}(\omega^{\omega})\\
\omega [4] (\omega [2] 2) & = \omega [3]^{\omega [2] 2} \omega =a_{\omega 2,3} (\omega)\\
\omega [4] (\omega [3] \omega) & =\omega [3] ^{\omega [3] \omega} \omega = a_{\omega^\omega,3} (\omega)\\
\epsilon_0^{\omega} & =(\omega [4] \omega) [3] \omega = (\omega [3]^{\omega}) [3] \omega = a_{\omega,3}(\omega) [3] \omeg
Oh, the link isn't wrong.
The "for all" is attached to the condition.
Meaning, $[\forall \alpha < \beta: \alpha + \beta = \beta] \iff [\beta = \omega^\delta]$
@Secret how is $\omega [3]^{\omega} (\omega [3] \omega)$ defined?
In the definition of the hypoperations above:

$\alpha [m]^{\lambda} \beta = a_{\lambda,m}(\beta)$

Let $\beta = \omega [3] \omega$, $\alpha = \omega$ and $\lambda = \omega$ and we are done
I don't mean in terms of layering.
In terms of its actual value.
@Secret concept test: how is $\sup$ defined for ordinals?
via its fundamental sequence if I recall $s_1, s_2, s_3, .... s_n, ....$
and letting $n \in \Bbb{N}$
there's a very simple one in terms of set theory... but I'll let you continue.
07:13
Well that's all I know..., I might be not familar with the set theory one
well how would you define $\sup(\{1,\omega,\omega^2,...\})$?
$1 \cup \omega \cup \omega^2 \cup$ ...?
ah, axiom of infnity
yes
@Secret axiom of infinity + replacement + union
Every ordinal can be represented in terms of layering numbers, thus the rules of the layering numbers will allow us to compare their sizes , hence establish a well ordering, thus you can say its actual value is determined in terms of layering.

which means, these new ordinals are inaccessible via any supremum operation, and can only be accessed by doing hyperoperations in a specific way

does that sounds valid?
(Since I did laid down the rules on how to do hyperoperations, I think we can define them in terms of constructing them using the hyperoperatations)
@Secret so are you telling me that $a[3]^\omega b$ cannot be defined via supremum?
07:21
@LeakyNun Yes
they can only be compared in terms of layering numbers with the usual ordinals when expressed in terms of layering numbers
and you can't constrct those ordinals?
except for $a=b=\omega$?
well e.g. $2 [3]^{\omega} 2 = a_{\omega,3}(2)=a_{0,4}(2)$
meanwhile:
$2 [3]^n 2 = a_{n,3}(2)$
so $2 [3]^{\omega} 2$ is larger than any exponential towers of $2$
meanwhile:
$\omega = \omega [1] 0 = a_{0,1}(0)$, $\omega = \omega [2] 1 = a_{0,2}(1)$, $\omega = \omega [3] 1 = a_{0,3}(1)$
I'm not interested in the $a$ forms. I'm interested in the cantor normal forms.
so is this statement true: $2 [3]^\omega 2 = \sup\{2[3]^n2|n<\omega\}$
nope, because the supremum will lead to $\omega$, which in a_forms they are not equal

(I have not read much into cantor normal form yet, will do it now...)
so how do you construct $2[3]^\omega 2$?
If you can't construct it, then it isn't an ordinal.
07:32
That, will require investigating whether they can be represented in terms of cantor normal forms, which is what I will be doing now to see...

But if they cannot ,then they cannot be ordinals as every ordinal has a cantor normal form
Is your current problem "how to get faster than tetration" iirc?
Nope, my current attempt is "how to get a tetration that grows at constant height"
i see
The existing ordinal tetration in the form of epsilon numbers already grows faster than compared to the finite case, since the height is a function of the input ordinal
a word of advice: you can't just define the problem out of existence.
you can't solve the question "how to get a tetration that grows at constant height" by defining a tetration that grows at constant height.
07:37
right, so I need to show (or prove cannot) that I can construct them
exactly
Zee
Zee
Is anybody here?
@Zee just ask; don't ask to ask
Zee
Zee
Alright, I wanna show f + x is bijective from unit interval to [0 ,2] where f is the cantor function
Can I simply express x as binary
And then say since f is bijective to [0,1]
And has the same value on x (on cantor sets)
what is the cantor function 😂
Zee
Zee
07:46
The result follows?
It's rhe binary representation of cantor sets and constant between them
I don't see how the result follows
Zee
Zee
Well since f is bijective from cantor sets to unit interval
If you double it you get the result
You double it since the binary rep of x equals f at x
Idk if it's right though...
I don't see how that accounts for x not in C
Well, intuitively speaking, I can see that f+x is bijective to [0,2], you are basically multiplying every point not in the cantor set by 2, thus making the devil staircase to slope upwards
we don't operate on intuition
Zee
Zee
07:55
Silly me, am gonna have to think about it more
@Secret I think you mean "point IN the cantor set" ?
@Zee hint: both functions are continuous and increasing
(not really sure if that helps)
is it surjective?
Zee
Zee
I don't like hints...
But thanks
I don't really think f+x is surjective
but I may as well be wrong
o nvm, I might have mixed up something
Zee
Zee
Well I need to prove it's bijective so it better be subjective
Surjective *
08:00
@Secret are you thinkin?
Zee
Zee
Wait...
nah no longer on the cantor set stuff
Zee
Zee
It can't be that easy
I didn't ping you
@Secret I mean ordinal
@LeakyNun Currently trying to find a good source to read up about cantor normal forms
Zee
Zee
08:02
Can I say x and f are continuous, and x is strictly increasing and f is increasing, since the sum is continuous and strictly increasing and starts at zero and ends at 2 we are done??
@Secret let's ignore cantor normal forms. I'm just requiring your numbers to be constructible
@Zee I don't know if your professor would accept that
Zee
Zee
I don't have a professor, do you guys accept that?
@zee can you prove that it is surjective?
Zee
Zee
Am not sure, do I need that for the above argument?
bijective functions must be surjective
Zee
Zee
08:06
I know
injective is easy to prove: it follows from the increasing
@LeakyNun I don't know of any other way to construct an ordinal in an unambigurious manner other than cantor normal forms, as you have said, I cannot have the thing I want to define to appear in the defintion, thus constructing ${}^{\omega+1}\omega$ as "the set that contains every ordinal smaller than it and including itself" is not acceptable
for surjective you need continuity but you need to elaborate
@Secret yes
Zee
Zee
Increasing as in strictly so?
@Zee yes, which is what you need to prove
08:17
Cantor normal form isn't all that useful. The normal form of $\omega_1$, for example, is just $\omega^{\omega_1}$. — Zhen Lin Sep 17 '12 at 4:45
o great...
@Secret please ping me when you have any result; I'll check 2 hours later.
@Secret which is why I said ignore the Form
(looking up other methods to construct ordinals now)
Zee
Zee
@LeakyNun for surjective, is it sufficient to say that it must be so since we can take lim on both sides of any point?
It seems trivial...
08:55
can't answer to math.stackexchange.com/questions/2318997/… (because of reputation) My answer is " yes, if you count inside the rectangle as along the border "
then you can place exactly 26 points
(each circle diameter = 5cm)
@JDoe I don't get where you place your points. In the middle of each circle ? Does the color change anything ?
I might have misread. Time to sleep @Hippalectryon
Alright :-)
09:29
@Zee I think so
or, image of continuous function is inteval
or, intermediate value theorem
@Secret you only have the axiom of infinity and replacement, if you ignore powerset...
${}^{\omega+m}\omega=\sup ({}^{n+m}\omega|n \in \Bbb{N},m \in \Bbb{N}\textrm{ fixed})=\sup(\omega,{}^{1+m}\omega,{}^{2+m}\omega,{}^{3+m}\omega,...)=\epsilon_0$

${}^{m+\omega}\omega=\sup ({}^{m+n}\omega|n \in \Bbb{N},m \in \Bbb{N}\textrm{ fixed})=\sup(\omega,{}^{m+1}\omega,{}^{2+m}\omega,{}^{3+m}\omega,...)=\epsilon_0$

${}^{\omega 2}\omega=\sup ({}^{n 2}\omega|n \in \Bbb{N})=\sup(1,{}^2\omega,{}^4\omega,{}^6\omega,...)=\epsilon_0$

${}^{\omega^2}\omega=\sup ({}^{n^2}\omega|n \in \Bbb{N})=\sup(1,\omega,{}^4\omega,{}^9\omega,...)=\epsilon_0$
Is $x\mapsto \omega [4] x$ a function?
I know that $x \mapsto n [4] x$ is a function for $n,x < \omega$ because it is bijective
but when $n > \omega$...
09:44
@Secret the first step in the first line is wrong
because $\omega + m \neq \sup\{n+m\}$
ah yes, always forgot that
(NB there is a slight typo in the 2nd line also, there are + that is in the wrong way around
the second line can be simplified by noting that $m+\omega=\omega$
but either way (modulo the mistakes) the conclusion should be unaffected
the third and fourth line are wrong by the same reason as the first line
likewise all your lines are wrong
Wait, if ${}^{\omega+1}\omega \neq \sup \{{}^{m+1}\omega\} = \epsilon_0$ then there is no way to show the object ${}^{\omega+1}\omega=\epsilon_0$ and by extension no way to show ${}^{\epsilon_0}\omega=\epsilon_0$??
09:50
@Secret how do you define ${}^{b^+}a$?
I am thinking about $f: x\mapsto \omega [4] x$ but it only works if $f$ is a function
why would it not be a function?
it wouldn't be a function if it is undefined for some ordinal $x$.
or if the range is multivalued
(i.e. no unique $y$ using the definition within the axiom of replacement)
here it's the former
$x=\omega+1$ is well defined, the problem is there seemed to be no way to show whether $f(x)$ is well defined or not since no supremum can reach it other than itself
(I mean, there seemed to be not even a way to show it is not well defined)
09:55
so how would you define it?
A set is an ordinal if it is transitive wrt $\in$ and is well ordered. I think the issue is transitive because what could be "smaller" than ${}^{\omega+1}\omega$, if there is one, it's probably not very obvious
you haven't defined it, so you can't talk about its elements
well, maybe ${}^\omega\omega+1$ would be smaller than it, depending on your definition

« first day (2504 days earlier)      last day (2813 days later) »