« first day (1671 days earlier)      last day (3357 days later) » 

12:00 AM
Pete is the master. And Ted of course!
 
i think you want half-open intervals, because the implication $[0,x] \subset S$ trivially implies $x \in S$.
 
He proves several things using it, including the intermediate value theorem.
Yes. Whoops. Fixed.
 
amusingly, that set of notes has its origins on MSE:
61
A: Induction on Real Numbers

Pete L. ClarkOkay, I can't resist: here is a quick answer. I am construing the question in the following way: "Is there some criterion for a subset of $[0,\infty)$ to be all of $[0,\infty)$ which is (a) analogous to the principle of mathematical induction on $\mathbb{N}$ and (b) useful for something?" The an...

 
Yeah, I know, that's how I found the article.
 
12:01 AM
It basically mimics the "standard" proof of Heine-Borel for the Reals IIRC.
 
@David did you see my ping to you above?
 
@JulianRachman many lecturers do just that
 
What? Turn their stuff into a book?
 
Or at least a set of "notes"-it's quite common, yes.
 
By the way, let's say we know $0\in S$.
 
12:03 AM
Open many texts, and they'll start with: "This text grew out of lectures for a 1-semester class I gave at...."
 
Lol When I become a senior in high school, I will make a book out of all my notes and call it "the journey through higher mathematics"
@David
 
So, using the second property, we know that there is some $y$ such that... etc. So, let's say it's $0.01$. So, in step $1$, we know that $[0,0.01]\subset S$.
Using it again, we can say that (let's call this step $2$) $[0,0.011]\in S$. (Doesn't need to be $0.011$, it's just an example.)
Again, it step three, we know there's a $y$ such that... Let's say it's $0.0111$, so, in step $3$, we know that $[0,0.0111]\subset S$.
The problem is that we want to prove that $S=[0,1]$, but just using rule $2$ an infinite number of times, we just get to prove that $[0,0.011\dots)\subset S$.
 
It's a bounded increasing sequence, it's cauchy, the reals are complete, it has a limit.
 
But that's where rule three comes in. So, using it, we know that $[0,0.011\dots]\subset S$.
@DavidWheeler Talking to me?
 
12:08 AM
The limit is, in this case, 1/90, IIRC.
 
It doesn't matter, we just find a new $y$ then-rinse, and repeat
 
Now, using rule 2 again, we have that, say, $[0,.1]\subset S$. Let's call this: Step $\omega$. (Hope you're familiar with ordinals! If not, just know it's the smallest infinite ordinal.)
 
it's a tad frustrating when the only references on a topic you're somewhat interested in are in book form or pay-walled
 
Again, step $\omega+1$ could be $[0,.101]$, step $\omega+2$ could be $[0,.1011]$, step $\omega+3$ could be $[0,.10111]$, etc...
 
ya, $\omega$ corresponds to the natural numbers as an ordinal type
 
12:11 AM
So, doing it an infinite number of times again (and a rule 3 and the end), you know that $[0,.1011\dots]\subset S$.
 
sure, say step $2\omega$
 
For complicated reasons, it has to be $\omega2$. Don't ask.
 
I don't care what you call $\omega + \omega$
 
Similarly, step $\omega2$ could give us that $[0,.11]\subset S$.
After using it an infinite number of times again, we might know that $[0,.111]\subset S$. This could be step $\omega3$.
And step $\omega4$, after using it an infinite number of times again again, could just tell us that $[0,.1111]\subset S$...
 
I don't care how many steps it takes, we may as well, just "skip ahead"
 
12:14 AM
So, after using rule 2 an infinite number of times an infinite number of times, we might still be nowhere near proving that $S=[0,1]$!!
And even step $\omega^2$ has no reason to be anywhere near proving that.
Also $\omega^\omega$, or even step $\omega^ \omega ^\omega$...
 
Well, yeah, the number of possible "paths" is a very large "number"
 
Ordinal.
 
use curly brackets
 
By the way, the ordinal after $\omega$, $\omega^{\omega^\omega}$, $\omega^{\omega^{\omega^\omega}}$, etc. is called:
$\epsilon_0$.
(Because epsilon should be a very small number.)
These are all countable, by the way. The number of "steps" we've done has been countable this whole time.
 
we can repeat the process with "epsilons", though (it takes a while to cycle back up to $\epsilon_02$).
 
12:16 AM
There's an $\epsilon_1$, by the way.
There is an ordinal after all of the countable ordinals! It's called $\omega_1$, and it's the smallest uncountable ordinal.
 
and, by extension, an $\epsilon_{\omega}$
 
Way ahead of you. I just jumped ahead to after all of the countable ordinals.
(It has cardinality $\aleph_1$, by the way.)
So, Theorem: We will prove that $S=[0,1]$ sometime before getting to $\omega_1$.
 
ya, cardinals aren't as "finely diced" as the ordinals
 
(Small note: The proof in the paper above that $S=[0,1]$ is much shorter and mentions no ordinals. But this is more interesting in my opinion.)
Why? Well, Lemma: You can't divide $[0,1]$ into an uncountable number of intervals.
Can you figure out why?
(Hint: The rationals are countable...)
you still there
?
 
12:20 AM
Just thinking?
 
well, if you could, you'd have uncountably many rationals in the union
 
Because...
 
each interval contains rationals (they're dense in the reals)
 
Yep!
So, back to the original question: Why must we be able to prove that $S=[0,1]$ before getting to step $\omega_1$ (the smallest uncountable ordinal)?
(P.S. Remember that $S\subseteq [0,1]$, as a given.)
 
as a wicked aside, i note dryly, you have to choose to allow uncountable things to be sets (it's not forced by the ZF axioms, per se)
 
12:24 AM
What? $\mathcal P(\mathbb N)$ is uncountable, I think. The power set of the natural numbers.
You can't prove it without the axiom of the power set, but that's in ZF.
(Hello, @GustavoViana!)
 
only if you take the "unrestricted power set" (think downward lowenheim-skolem)
 
(We are talking about adding infinity to infinity an infinite number of times, except you add that to itself an infinite number of times.)
> Think Downward Lowenheim-Skolem
You clearly know much more than me about Set Theory.
 
IF ZF admits a model, it admits a countable model
 
Somehow, that makes sense.
 
but then the power set of the naturals doesn't have all the things in it we think it does
 
12:26 AM
In any case, the "real" proof that $S=[0,1]$ doesn't involve ordinals.
 
you can still form the "normal" power collection, it's just not a SET
 
My head hurts so back to the original question, K?
 
but, I digress...you were saying about $\omega_1$?
 
4 mins ago, by columbus8myhw
So, back to the original question: Why must we be able to prove that $S=[0,1]$ before getting to step $\omega_1$ (the smallest uncountable ordinal)?
(Hint: Remember what each of the steps are doing: adding a small interval to what we know is in $S$. I'm not good with words...)
 
the only thing that comes to mind is that the number of steps must be less than the cardinality of [0,1]
 
12:29 AM
10 mins ago, by columbus8myhw
Why? Well, Lemma: You can't divide $[0,1]$ into an uncountable number of intervals.
Use that.
 
if it was an uncountable number, each step would induce an uncountable partition of the interval so far, so that when we finished, we'd have an uncountable partition of $[0,1]$
I should say "ordinal" but I'm being lazy
 
Correct! So, that means that before getting to $\omega_1$, we will prove that $S=[0,1]$.
In other words: $S$ EQUALS $[0,1]$!!!
 
by the MINIMALITY of $\omega_1$
 
And that means that our induction scheme works.
(Yeah, $\omega_1$ is minimal. But I got to go, sorry.)
 
It occurs to me we might run into problems trying to do this on arbitrary subsets of the reals
well, au revoir
 
12:35 AM
Hello!! Is there someone familar with regular languages??
 
12:48 AM
Hi @robjohn you have been running in my mind all day. =)
 
He must be all sweaty-let him take a breather!
 
@DavidWheeler Do you believe in predestination or determinism of any kind?
 
I'm on the fence about that-I believe in block time AND "many worlds"
 
My struggle with mental problems led me to think about this. I keep trying and trying to get well, but I just cannot. Am I fated to never get well, whatever that means?
 
As I see it, only in those realities where it doesn't happen. Is this one of those? I cannot say.
 
12:58 AM
Ah, I see your many worlds theory now. I am not familiar with it.
 
I think all possibilities happen somewhere, but we only have the single vantage point
 
I believe that the ordinary human mind alone cannot get the answer to the question of determinism.
But possibly people like the Buddha, with an extraordinary mind, know the answer.
 
From a practical perspective, you may as well act as if you have choice, because it's a better frame of mind
 
I am not sure it is a better frame of mind.
 
@ABeautifulMind yes, this exactly. we are so interwoven with it all, it's truly hard to separate "I" from "all"
@ABeautifulMind If believing your path is already chosen is helpful, then believe that. What I worry about, in your case, is just resigning yourself to defeat.
Our dreams are not meant to torture us, but to give us hope.
 
1:03 AM
@DavidWheeler No, I will keep doing my best to get well (of course my best is not perfect). But it is frustrating that things don't happen the way you want even though you do that.
 
We have the functions $f_1 , \dots , f_{30}$ :

$$\frac{n}{\lg n} , \ \ n^{\lg n} ,\ \ (\sqrt{2})^{\lg n}, \ \ n^2, \ \ n!, \ \ (\lg n)! ,\ \ \left( \frac{3}{2} \right)^n ,\ \ n^3 ,\ \ \lg^2 n ,\ \ \lg(n!) ,\ \ 2^{2^n}, \ \ n^{\frac{1}{\lg n}}, \ \ \ln{\ln n}, \ \ e^{\log_{10} n }, \ \ n \cdot 2^n, \ \ n^{\lg{\lg n}}, \ \ \ln n, \ \ 1 ,\ \ 2^{\lg n}, \ \ (\lg n)^{\lg n}, \ \ e^n ,\ \ 4^{\lg n}, \ \ (n+1)!, \ \ \sqrt{\lg n}, \ \ \lg{\lg{\lg n}}, \ \ 2^{\sqrt{2 \lg n}}, \ \ n, \ \ 2^n ,\ \ n \lg n, \ \ 2^{2^{n+1}}$$
 
Well, if Buddha believed it was all 'determined" why would he bother to teach?
 
Even if the world was deterministic, chaos theory (which shows that deterministic systems can have randomness of a sort) says that we'll never prove it.
 
Also, chaos theory says that certain "random-appearing" things, have a hidden order (and I daresay, an exquisite beauty)-like strange attractors.
One's apparently meaningless journey, could be the petal of a lotus.
 
We have the functions $f_1 , \dots , f_{30}$ :

$$\frac{n}{\lg n} , \ \ n^{\lg n} ,\ \ (\sqrt{2})^{\lg n}, \ \ n^2, \ \ n!, \ \ (\lg n)! ,\ \ \left( \frac{3}{2} \right)^n ,\ \ n^3 ,\ \ \lg^2 n ,\ \ \lg(n!) ,\ \ 2^{2^n}, \ \ n^{\frac{1}{\lg n}}, \ \ \ln{\ln n}, \ \ e^{\log_{10} n }, \ \ n \cdot 2^n, \ \ n^{\lg{\lg n}}, \ \ \ln n, \ \ 1 ,\ \ 2^{\lg n}, \ \ (\lg n)^{\lg n}, \ \ e^n ,\ \ 4^{\lg n}, \ \ (n+1)!, \ \ \sqrt{\lg n}, \ \ \lg{\lg{\lg n}}, \ \ 2^{\sqrt{2 \lg n}}, \ \ n, \ \ 2^n ,\ \ n \lg n, \ \ 2^{2^{n+1}}$$
 
1:09 AM
@ABeautifulMind I feel your frustration. You want your hardest struggles to be behind you, so you can focus on your goals. It makes you sad.
 
The purpose of the site is to ask questions and get a lot of input from volunteers, rather than demanding answers from individuals logged into chat, @user159870. :)
 
@user159870 You should not ping random people to answer in chat. =)
 
@ABeautifulMind It's raining... hardly weather for a run.
 
@user159870 Despite what Thomas said, feel free to ask, but realize no one may answer. perhaps they don't know, perhaps they are busy with something else.
 
I wasn't even saying that it is wrong to ask, just not optimal usage of Math.StackExchange.com, which is useful precisely because you open the question to lots of people, and the ones interested in answering can answer.
 
1:12 AM
@user159870 You can just post your question here and those who can answer will answer automatically.
 
@ThomasAndrews Not criticizing you, either, just clarifying. The site gets better results, often not faster, though. Depends.
 
Seems like letting $g(n)=2^{\max \{f_i(n)\}}$ would work to answer the question. @user159870
 
@robjohn It's not raining in my mind.
 
@robjohn My roommate is leaving to go on a run right now.
 
@MikeMiller did you get to go to the show on Friday?
 
1:15 AM
@MikeMiller You don't get single rooms as a grad student?
 
Yes, it was very nice. I think the writing was poor, but the rest of the show made up for it.
The director used to be a photographer and that was apparent.
 
It seems that Mike does not wish to answer my trivial questions.
 
Roommate is a term used to refer to both those who share a room with you, and those who share an apartment with you. I live in a 2 bedroom, 2 bathroom apartment. (Nobody says 'apartmentmate'.)
 
@ABeautifulMind Hey, I asked first.
 
Wow, 2 bathrooms, that is luxury.
 
1:18 AM
Henceforth I shall use "apartmentmate". For everything.
 
Do you call them flats there? You can say flatmate.
 
I've heard apartmentmate occasionally, but roommate is more common.
 
But isn't $g(n)=2^{\max \{f_i(n)\}}$ bounded below by the functions $f_i(n)$ ? @ThomasAndrews
 
Some day, I will invent/discover a mathematical idea and call it a "room." And then I will name a related object a "roommate."
Similarly for "apartment" and "apartmentmate," to make @DavidWheeler happy.
 
@user159870 Whoops, forgot the meaning of $\Omega$.
 
1:22 AM
@columbus8myhw Your use of punctuation suggests you use American English.
 
@columbus8myhw as long as I get a footnote, I'm happy. Apartmentmate.
 
@ABeautifulMind, is it because of saying "stuff," rather than saying "stuff"?
And, yes, I am from NYC.
 
@columbus8myhw It is because you put the period inside the quotes and not outside.
 
So, in other words, yes.
 
@columbus8myhw No, that is different.
 
1:23 AM
New York City? Well, that really chaps my hide....
 
...what?? @DavidWheeler
 
It's from a Pace picante sauce commercial
 
@DanielFischer I have tried to prove that for the initial pseudocode that implements the Horner's method the following relation expresses an invariant for the while-loop at the lines 3-5.
At the beginning of each iteration of the while-loop at the lines 3-5
$$k=\sum_{d=0}^{m-(j+1)} a_{d+j+1} x^{d}$$
That's what I have tried:
Initial check:
Before the first iteration of the loop, we have $j=m$ and $k=0$.
The sum $\sum_{d=0}^{m-(m+1)} a_{d+m+1} x^{d}=\sum_{d=0}^{-1} a_{d+m+1} x^{d}$ is zero, so it is equal to $y$, therefore the invariant holds before the first iteration of the loop.
This is the initial code:
k<-0
j<-m
while (j>=0){
         k<-a_j+x*k
        j<-j-1
  }
 
...I won't ask
 
@DavidWheeler You mentioned you had a gf some time back. Are you still together?
 
1:28 AM
@ABeautifulMind 5 years now
 
Do you have an other idea? @ThomasAndrews
 
Wow Robin Williams had 3 wives LOL
 
Yes, post the question on the site so you can get help from people who actually remember the stuff. @user159870 . But you ignored that advice previously
 
sigh I want. to be. entertained.
 
Let me give you a song.
 
1:47 AM
@DavidWheeler Prove the intermediate value theorem using real induction. Go.
5 mins ago, by David Wheeler
sigh I want. to be. entertained.
(It's in here, by the way. If you give up.)
 
@columbus8myhw I said entertained, not tortured. Besides, it's a duplicate of a question already answered on this site.
 
Why are you on Stack Exchange if you don't like answering math problems?
 
@columbus8myhw I'm messed up in da head.
 
Just like me.
 
See? Jasper gets me.
 
1:52 AM
I am going to sleep. I hope my miracle comes soon.
 
Me too. I've been sleeping in a car for about 7 hours now
 
Bonnuit Jasper
wow, people trying to figure out their homework are really getting slammed here, lately.
 
If you want to be entertained, read about ordinals.
(By the way: There are way too many of them)
There isn't even a set of all ordinals. 'Cause there's too many.
 
What's an ordinal
 
1:57 AM
OK, here goes...
 
? Not everyone doing math has to be entertained every second of the day
 
So, let's say you're counting.
And you start: 0, 1, 2, 3, etc.
 
@JulianRachman Why are they getting slammed? A variety of reasons.
 
(You start with 0 because why not.)
 
But is 0 "really" a number?
 
1:58 AM
So, with the natural numbers, that's all there is. But, with the ordinals, you keep going.
 
Because they to lazy or what?
 
The first "infinite ordinal" is called $\omega$. It's bigger than all natural numbers.
And then you keep on counting: $\omega+1, \omega+2,\omega+3,$ etc.
 
@David ^^
 
Some reasons: poor formatting (too much latex in the title, or too much of their question in the title, or non-use/bad use of latex)
 
And when you're done with that, you have $\omega+\omega=\omega2$. (You don't call it $2\omega$ for complicated reasons.)
 
1:59 AM
Starting ordinals with zero always irritates me. I realize it is the set theory approach, due to the model for ordinals chosen there, but nobody places $0$th in a race.
 
@David so they just don't have a balance of latex? Ok.
 
poor formulations of questions, questions at "too basic" a level...standard newbie errors
 
Well, $1$ is the number of ordinals less than it ($0$). $2$ is the number of ordinals less than it ($0$ and $1$). Similarly, $\omega$ is the number of ordinals less than it ($0,1,2,3,\dots$). See, it makes sense.
In any case, you keep on counting, because there are even more ordinals.
$\omega2+1,\omega2+2,\dots$
 
It is interesting to realize that the reason people wondered if $0$ is a number was because to count a finite set, you had to implicitly sort the set and then figure the ordinal of the last element of the set. And to do that, you'd have to have at least one element. So we were confusing ordinals with cardinals from the beginning.
 
After all of those, you get to $\omega3$.
 
2:01 AM
@David ok
 
@ThomasAndrews Say again?
 
So, many people making their first post here find their questions immediately downvoted.
 
In any case, you eventually get $\omega4$, $\omega5$, etc., and when you're done with all those, you get to $\omega^2$.
There's also $\omega^3,\omega^4,\omega^\omega,\omega^{\omega^\omega}$, etc.
 
So are the ordinals countable or not
 
At this point, you might want to use "up arrow" notation
 
2:03 AM
After $\omega,\omega^\omega,\omega^{\omega^\omega},\omega^{\omega^{\omega^\omega}}$, etc., you get what's called $\epsilon_0$.
 
If you have a pile of beans that you want to count, you take the first bean off. Then the second, then the third. And then the 9060th bean is the last bean, so you have 9060 beans. So the 9060th is an ordinal, but it determines the cardinal (of a finite set.) But that approach doesn't work with a pile of no beans - there is no "ordinal of a last bean." @columbus8myhw
 
Because epsilon should be a very small number.
 
@David SE should restrict the kind of people that are allowed to use MSE
 
By the way, @ʙᴀᴅᴀᴛᴍᴀᴛʜ, these are all countable.
 
@JulianRachman What should the rules for restriction be
 
2:04 AM
@David Any ideas for a proposal?
 
Pretty cool @columbus8myhw
 
@JulianRachman Some people feel that way, yes.
 
By the way, there's also $\epsilon_1$, which is what you get after raising $\epsilon_0$ to itself an infinite number of times.
And they keep on going.
 
@JulianRachman I do not wish to shape, in any way or form, math stackexchange policy
 
At least knowledge up to pre calculus
 
2:04 AM
After all of the countable ordinals, you get $\omega_1$, the smallest uncountable ordinal.
It has cardinality $\aleph_1$.
By the way, there's an uncountable number of countable ordinals. (Just like there's an infinite number of finite ordinals.)
 
@JulianRachman I don't know pre calculus :(
 
@ʙᴀᴅᴀᴛᴍᴀᴛʜ Do you know what "uncountable" means?
 
@columbus8myhw Yes
 
OK.
By the way, since there's an uncountable number of countable numbers, that means you can't possibly make a notation that names them all (since there's always a countable amount of things you can name with any notation).
(Hint: Put them in alphabetical order.)
 
2:08 AM
But you can make a notation that goes up to an arbitrary countable ordinal.
HOWEVER:
 
They are countable, but also uncountable at the same time?
 
The same way that there's an infinite number of finite numbers.
 
@DavidWheeler It's a shame, some people being run out. If you give someone the benefit of the doubt at first, they'll show you soon enough whether they're worth a serious effort on your part
 
They're "infinite but finite at the same time," I guess.
 
the "slick" proof that the uncountables aren't listable is known as cantor's second diagonal argument
 
2:09 AM
HOWEVER: A good notation has two things: You can tell easily if a jumble of symbols is part of the notation, and if someone writes down two ordinals using the notation you can tell easily which is bigger.
There's something called the Church-Kleene ordinal...
...that's countable, but:
 
$\aleph_1$ looks like a cool symbol
 
There's no "good" notation for the ordinals less then it.
 
Proposal anyone?
 
@pjs36 well, the policy-makers here have their reasons. I'm not staff, so I just observe and comment.
 
Either you can't decide whether something's in your notation, or you can't decide which of two things is bigger. (P.S. What do I mean when I say you can't decide something "easily?" Basically, it means that no computer program exists that can decide it for you.)
 
2:11 AM
@columbus8myhw What is the meaning of $2^{\aleph_0}$
 
Ps, everyone should follow my Area 51 proposal NOW
7
Mathematics Advisory

Proposed Q&A site for people who are on the path to higher mathematics and want advise on certain techniques, future topics, reference requests, and "road maps" to related fields.

Currently in definition.

 
That's the cardinality of the real numbers. You can't prove if it has the same cardinality as $\aleph_1$, though, using the normal axioms of set theory.
 
@columbus8myhw What axioms do you need
 
If "$2^{\aleph_0}=\aleph_1$" was one of your axioms, that'd work.
 
@ʙᴀᴅᴀᴛᴍᴀᴛʜ the same as $|\mathcal{P}(\Bbb N)|$
 
2:13 AM
But there are others and I'm not a set theorist.
 
And add questions too ^^
 
Reason that the Church-Kleene ordinal exists: Consider the set of ordinals that have "good notations" for the ordinals leading up to them. These are all less then $\omega_1$, since $\omega_1$ has no notations for the ordinals leading up to it at all (since it's uncountable). ...
... Now, the idea of ordinals is that, given any set of ordinals, there's always something larger. The Church-Kleene ordinal is that "something larger," and hence has no good notation for the ordinals leading up to it.
(Sorry for the headaches.)
(Also, try to prove that the Church-Kleene ordinal is countable.)
 
An interesting side note: since $\omega_1$ is an ordinal, you can impose the order topology on it, creating a topological space.
 
Is this the "long line"?
 
Is that the same thing as the Alexandrian long line
 
2:16 AM
Google isn't turning anything up. What is that?
Also, what kind of kid knows what the "Alexandrian long line" is and doesn't know precalculus??
 
Alexandroff, sorry
@columbus8myhw I never got around to learning it so I just gave up on it and started learning other math
 
Ah. Wikipedia says that that's the same as my "long line."
I don't know what @DavidWheeler was referring to, since I'm not sure what order topology is.
@ʙᴀᴅᴀᴛᴍᴀᴛʜ I'm assuming that means you don't know calculus either?
 
I guess it's just a topology on ordered sets
@columbus8myhw I know a little bit
 
What exactly is "a topology" again...?
@ʙᴀᴅᴀᴛᴍᴀᴛʜ But no trig?
 
@columbus8myhw A topology on a set X is a collection of subsets of X where X itself and the empty set are in the topology, the arbitrary union of open sets is open, and the finite intersection of open sets is open
@columbus8myhw I don't really know any trig, no
 
2:21 AM
I need to go. I'll look it up later.
 
Like I couldn't tell you $\cos(\pi / 4)$ or anything like that
Bye @columbus8myhw
 
TOPOLOGY.......
 
$\cos(\pi/4) = \dfrac{\sqrt{2}}{2}$
 
@DavidWheeler Thanks, I can never remember any of those
 
$\pi$ is a half-turn, so 1/4 is an 1/8-th turn. 360/8 = 45 degrees.
So it's an isosceles right triangle: $a^2 + a^2 = 1$
$a^2 = \dfrac{1}{2}$ so $a = \dfrac{1}{\sqrt{2}} = \dfrac{\sqrt{2}}{2}$
 
2:31 AM
@DavidWheeler Well I know how to do all that, but it just takes too long so I don't like to do it at all
 
Lil
can someone help with an abstract algebra question?
 
C'mon, there are $4 + 3\cdot 4$ special values to remember, which basically collapse down to $5$ if you use symmetry. I know you've got it in you, Bad!
 
@pjs36 My memory is so terrible :(
 
I understand, mine's not great.
 
@Lil What is it?
 
2:42 AM
Oh, if someone shouted "$\cos(\frac\pi4)$" at me, it would take me a minute to work it out. I still say that I know trig, even though I don't have them memorized and on the top of my fingers.
 
Lil
@David
Sorry to bother you again.. it says Let F be the set of all functions f mapping the set of real numbers into the set of real numbers that have derivatives of all orders
<F,+> with <F,+> where ϕ(f)=f', the derivative of f
 
Ok, $C^{\infty}$ functions, yes
 
Lil
so I picked f(x)=x^2 and f'(x)=2x?
 
For what purpose?
 
Lil
I'm not quite sure...Do I want one function or two?
I am still trying to show if it is an isomorphism
 
2:47 AM
Well, what do you want to check first, whether its bijective, or whether it's a homomorphism?
 
Lil
lets do 1-1
bijective
 
By the way, someone seems to have gone through all of the questions I've asked on the site and upvoted them. So, them you, anonymous stranger!
 
Ok, we don't want to pick two "specific functions", we just want to ask the "general question"
 
Lil
ok, makes sense.
 
IF, $\phi(f) = \phi(g)$, then IS IT TRUE that $f = g$?
We have two options: try to prove it for any pair with $\phi(f) = \phi(g)$ (proof), or find ONE pair, for which it ISN'T true (disproof).
 
Lil
2:51 AM
lets try a proof by contradiction
 
well, we could, but I think that's more indirect than you need, here
 
Lil
ok then just disproving is fine
 
Can you think of two functions that are different, with the same derivative?
 
Lil
x^2 and x^2+4
 
Good!
So...what does that tell you?
 
Lil
2:55 AM
its not one to one because 2 functions correspond to the same antiderivative
 
same derivative, the anti-derivative would be going "the other way"
remember $\phi$ goes FROM function TO derivative
 
Lil
oops meant derivative.. sorry my bad
 
But yes, that's correct, and one counter-example is all you need for disproof.
 
Lil
thank you again!
 
So it's not 1-1, and if ALL we have to do is decide isomorphic/not isomorphic, we can stop right there
But...let's check the other conditions "just for fun"
Is $\phi$ onto?
 
Lil
2:59 AM
Which function am I trying to find the pre-image for now?
 
I regret not learning algebra when I took the class
 

« first day (1671 days earlier)      last day (3357 days later) »