« first day (1756 days earlier)      last day (3268 days later) » 

3:00 AM
most every book has copious errata
 
user147690
Ahhh they had my one already. They mis-defined the distributive laws on page 120 for rings
 
how big the errata is depends on how often people tell the author about mistakes more than anything
 
user147690
That errata page is incredibly impressively long
 
And don't second editions have far fewer?
 
yes
 
user147690
3:01 AM
@SohamChowdhury Well I would hope they have none of the original list :P(and don't introduce more)
 
I was having trouble with understanding how $\text{Hom}_{\sf Ab}(G,H)$ is an (abelian) group (what if there's only one homomorphism between the two groups?) until I realised that a group can have one element.
:(
In my defence, the proof used two, and I never noticed that they're free to be the same.
 
Dammit, fluxions sounds so much more kewl than calculus. We should bring back the fluxes. We need flux capacitor.
 
@bjb568 Newton, on seeing Gabriel's Horn: "What the fluxion?"
 
:p
 
The starboard is impressively oligosyllabic today.
(word?)
 
3:05 AM
Wrong problems are probably the most annoying. I as recently looking at 2 and 3 dimensional topology by Moise, and he mentions that many propostions are stated as if they were true, and part of the problem is to figure that out. Maybe all textbooks should do that so there is no need for errata in the problem section
 
What's oligosyllabacism?
 
Having few syllables, I guess.
 
Oh.
 
I made it up.
 
I see.
 
3:06 AM
It exists!
Woohoo!
 
∃ teh wordz
 
user147690
$\exists $ oligosyllabic
 
"Of three or fewer syllables, as a word; trisyllabic, disyllabic, or monosyllabic: opposed to polysyllabic."
 
user147690
Hey @Rem
 
Hm… should I bother to install the latex?
 
3:07 AM
Hello@AlexClark
 
**Theorem**: let $\{o_i\}_{i \in \{1, 2, \dots, r\}}$ is an orthonormal basis for the column space of a matrix $X$ and let $O = \begin{bmatrix}o_1 & o_2 & \cdots & o_r\end{bmatrix}$. Then $OO^{\prime} = \sum\limits_{i=1}^{r}o_io^{\prime}_i$ is the perpendicular projection operator onto $C(X)$.

In the proof, they say $OO^{\prime}OO^{\prime} = OIO^{\prime}$, $I$ being the identity matrix. How do I know that $O^{\prime}O = I$?
 
user147690
@Soham I have to do some assignment work now. More Aluffi later.
 
@bjb568 Yes, you must.
@AlexClark Okay, later then.
 
Agh, is there a userscript?
 
Indeed.
 
user147690
3:08 AM
@bjb568 It's a bookmarklet on the top of the starboard on the right. $\LaTeX$ in chat
 
Read the star board
 
bookmarklet ≠ userscript
 
user147690
Oh Paul's got picked up :'(
 
I cant see the 🌟 board coz I am on my phone:(
 
user147690
 
user147690
3:09 AM
 
user147690
I should be famous( a star so to speak )
 
Lol
I did not notice you had posted, I was wondering why was no chat guidlines up when I woke up
 
yay, script bound to command-1
 
@AlexClark, about how he distinguishes products and coproducts in categories like $\sf{Vect}$: It's sort of similar to how you're supposed to distinguish C# and Db, if you're familiar with a little bit of those things.
 
Sadly, @AlexClark, I somehow got my star badge without posting that message... just by tarting around :)
 
user147690
3:14 AM
@pjs36 Oh I have had it on three accounts(including this) without needing that haha
 
Oh yeah, I forgot you were the account master!
 
user147690
@pjs36 Hahahaha
 
$\square \!\!\!\! \checkmark$ Make @AlexClark realize everybody ignores him, by letting him do something, which everybody ignores, then I do the same exact thing, which everybody thinks is the best thing since sliced bread.
6
 
user147690
@PaulPlummer Hahahahaha
 
☑ this is a better checkbox
 
3:16 AM
LaTeX > emoji brah
 
unicode > LaTeX
dammit… less than and greater than are basically the same thing, right?
 
user147690
@SohamChowdhury <---this
 
@bjb568 LaTeX $\checkmark \left(\frac{10}{10}\right)$
 
user147690
@bjb568 What? Oh haha
 
3:18 AM
@SohamChowdhury 11/10 with rice?
 
Redditor!
 
:D
 
faux-romantic music
 
elevator music, just because
 
user147690
I remember that thread
 
user147690
3:19 AM
But are you then following the button?
 
Nah, already red and satisfied.
I thot about going grey, then I stopped thinking about going grey.
 
user147690
@bjb568 What number though? Better not be higher than 1
 
@AlexClark Non-presser FTW
 
You just have a bad font
 
user147690
@SohamChowdhury Me 2, I want a 0 or a 1
 
3:21 AM
Meh
 
user147690
I will get a 0 or a 1 tonight, since we are down to 280ish users at 8pm where I am from
 
I haven't reddited (or been on FB or HN or wherever) in a week
 
@PaulPlummer What kind of sorcery is this?
 
3:22 AM
Discipline level = pro, @AlexClark
Hah!
 
user147690
This is what I see imgur.com/LDFiiWa
 
The people who work on $\LaTeX$ are doing the lords work, you shame them by using unicode when you can use $\LaTeX$. @bjb568
2
 
WWKD: What Would Knuth Do?
 
user147690
@bjb568 Filthy presser
 
@bjb568 Filthy presser
 
user147690
3:25 AM
It literally goes lower than 10 most minutes these days
 
That looks really good on your computer @AlexClark
 
He probably has a Retina display.
Not sure.
Nope, he doesn't.
 
I don't think the display effects screen shots
 
All is Unicode.
 
3:26 AM
@PaulPlummer Resolution does.
 
Sure
 
That's what I meant
 
user147690
I have a really nice resolution
 
Anyhow, I'll get back to the page that I've been stuck on for half an hour.
 
user147690
 
3:27 AM
@AlexClark OMFG lol
 
user147690
Haha
 
user147690
@SohamChowdhury It's just the uni computers haha
 
I need a 5k display.
 
3:31 AM
^ Classic Redditor
 
A 5k display to look at all your precious unicode? @bjb568
2
 
Hey, when you're coding, you need the best equipment! Wait what, silence you, I'm not just using a text editor, it has colors.
 
Vim is love, vim is life
 
sublime > vim
sublime is just sublime
 
3:32 AM
Is there a good intro to real analysis?
 
vim is text surgery
 
I realized recently sublime is an english word. English is weird.
@SohamChowdhury and I do surgery like a kid on surgeon simulator…
 
I am sure there is a question on stack exchange asking exactly that @AaronHall
 
user147690
@PaulPlummer So is 'field norm' common notation to refer to taking any $a+b\sqrt{D}$ to $a^2-Db^2$ via the standard $(a+b\sqrt{D})(a-b\sqrt{D})$?
 
Not sure @AlexClark
 
3:35 AM
well why didn't I think of that... on SO it's considered off topic to ask for resources, but I've been seeing a lot of those types on the other SE sites...
 
Goodnight! I'm going to go sleep with my unicodes…
 
user147690
@bjb568 Night :)
 
user147690
@AaronHall Yes, we are nicer here
 
@AlexClark Should we be though...
 
top 3 google hits for my question are SE, top 2 are Math, 3rd is MathOverflow
 
3:36 AM
I gave up on SO years ago. It's near-impossible to ask there nowadays.
 
user147690
Oh I finished a chapter of D&F hmmph
 
@AlexClark Now? Assignment purposes?
 
user147690
Yep
 
user147690
Sorry a subchapter
 
@AaronHall Here is a good place to start math.stackexchange.com/questions/62212/…
 
user147690
3:38 AM
431
Blocked malicious login attempts
 
user147690
Holy crap
 
user147690
Better strengthen the password
 
user147690
For my website
 
Someone's into you, @Alex.
 
3:39 AM
that's the same book Math Overflow recommends
 
Stephen Abbot?
 
user147690
@PaulPlummer Oh I have that book(in pdf
 
user147690
Seems really nice
 
user147690
Probably run through that and then hit rudin's principles of mathematical analysis
 
@SamuelYusim: enjoy Kevin's answer
 
3:40 AM
It is a nice book, I remember for my first course we had a book I didn't care for, then future classes were using that book and I thought it was pretty good (although I never really looked at it in detail) @AlexClark
 
user147690
My password is now somewhere between 16-32 characters long(won't specify since it makes bruteforcing really easy[in comparison] :P)
 
I will, it looks interesting
 
ok, I ordered it.
thanks!
 
Hmm, I just got a good idea for passwords, I can have the same password but I hash it in different ways every time, so I can just copy and paste, or figure it out by hashing the pass I already know, instead of going through this buisness of having a million different passwords @AlexClark
 
user147690
Damnit I can't wait until I TA for that textbook money
 
user147690
3:43 AM
@PaulPlummer I just have passwords that are based on nice hand motions
 
Did you guys download that 10M username/password set?
 
@SamuelYusim he facebooked me with a link to it saying 'oh god what have I done'
it's a good answer though
 
@AlexClark I have some nsfc guesses for your passwords :D
 
Wow a lot of stars in the last 25 minutes
 
people star too much
10
 
user147690
3:49 AM
@PaulPlummer not safe for copying?
 
chat
 
user147690
Oh
 
wow that answer was hype
 
user147690
Shoot away I suppose, no way you could guess
 
Oh got into wordpress
 
user147690
3:50 AM
LOL
 
user147690
It has 7 symbols for the record
 
ugh
 
ok guys, gotta hit the hay, or my wife will fume
 
user147690
@AaronHall Okay, night
 
thanks again!
 
3:52 AM
Why would she be mad if we don't hit the hay?
:P
 
user147690
@PaulPlummer Are you actually trying to log in?
 
No
It is sort of funny, it is more convenient for me to go to my blog, then the blogroll in my cite, to find yours, than it is to find yours in my bookmarks @AlexClark
 
user147690
@PaulPlummer Hahaha, I just go through my blogroll to all my blogrollees also
 
user147690
Is that a picture of you in your header btw @Mike
 
haha
no
 
4:02 AM
It is a picture of me...
:)
 
user147690
@MikeMiller Was it with the template?
 
user147690
Is it a musician from the 80s?
 
closer
 
user147690
"The Man Who Fell to Earth"
 
user147690
4:04 AM
An actor from a 1976 movie who is acting as david bowie?
 
@AlexClark god, I'm going to be a poor student in two years, too :(
 
no, that's david bowie acting, as actors do, in the film "the man who fell to earth"
fantastic movie
 
Paul, what's your blog?
 
user147690
@MikeMiller Oops, I thought he was solely a musician
 
4:05 AM
mostly yeah
 
Is anyone willing to guide me through the proof that $C(XX^{\prime}) = C(X)$, $C(X)$ being the column space of a matrix $X$?
 
he's been in a couple of movies; that one's the only one where he did a fantastic job
he was coked out of his mind which made him rather alien to the rest of humanity during the period; as this fit the character, he luckily didn't need to do much acting
 
user147690
@MikeMiller Oh, very interesting
 
4:37 AM
@Clarinetist isn't that not true if $X' = 0$?
by which I mean a matrix of all 0's
 
@SamuelYusim My book says this is true for any matrix. Hmm
0
Q: Using SVDs to prove $C(XX^{\prime}) = C(X)$

ClarinetistLet $C$ denote the column space. I would like to prove $C(XX^{\prime}) = C(X)$ for $X \in M_{n \times p}$. This answer suggests using singular value decomposition. So sure, $XX^{\prime}$ is symmetric and thus $$XX^{\prime} = PD\left(\lambda_i\right)P^{-1}$$ for some orthogonal $P$. Since $P$ is ...

 
oh, you use $X'$ to mean the transpose?
 
Yes, sorry
 
that's fine, I just hadn't seen the notation before
 
Yeah, it seems prevalent in econometrics and statistics
 
4:44 AM
Hey @Clarinetist
 
Hey @SohamChowdhury
 
Does anyone know why free groups are called free?
 
because there are no rules imposed on products I suppose
 
free to do whatever they want
 
I like that there's an isomorphic copy of the free group on 3 letters inside the free group on 2 letters, that's sort of cute
 
4:48 AM
What does $X^{\prime}$ usually mean, if not the transpose?
 
I've been stuck on the last page of the section on free groups for a while now.
@Clarinetist Just another variable.
 
Oh I see
 
the notation I've seen is $X^T$ or $X^t$, not that it's a big deal
 
I think what baffles me most is how much I'm struggling to finish some of these proofs
It's been a welcome change compared to actuarial work, but dang, quite a struggle for me
 
welcome to math
 
4:53 AM
More like welcome back :)
 
this would be a good time for a thumbs up if the chat could to that
so pretend I gave you one
 
$\checkmark$
is close enough
 
Hmm, I wonder if there is a way to prove $C(XX^{\prime}) = C(X)$ without using SVDs
and without using ranks
I suppose I could try containment
 
my first thought, not that I've worked it out, is to expand out the product into a big mess and from there show that every column of $X$ is a linear combo of columns of $XX'$ and vice versa
 
$C(XX^{\prime}) \subset C(X)$ because... you're basically taking a linear combination of the columns of $X$ by multiplying it by something after you look at the column space of the resulting matrix
 
4:57 AM
sounds too fuzzy
 
@Chris'ssis if you can, try $\int_0^1 x^{-\Gamma(x)} \text{d}x$ when you come on
 
@SamuelYusim I considered that at first... might want to try it again :P
If $Z = [a_{ij}]$, $$ZZ^{\prime} = \begin{bmatrix}
\sum\limits_{k=1}^{p}a_{1k}^{2} & \sum\limits_{k=1}^{p}a_{1k}a_{2k}& \cdots & \sum\limits_{k=1}^{p}a_{1k}a_{nk} \\
\sum\limits_{k=1}^{p}a_{2k}a_{1k}&\sum\limits_{k=1}^{p}a_{2k}^{2} & \ddots & \vdots \\
\vdots & \ddots & \ddots & \sum\limits_{k=1}^{p}a_{(n-1)k}a_{nk} \\
\sum\limits_{k=1}^{p}a_{nk}a_{1k} & \cdots & \sum\limits_{k=1}^{p}a_{nk}a_{(n-1)k}& \sum\limits_{k=1}^{p}a_{nk}^{2}
\end{bmatrix}$$
So now I gotta look at the column space of this thing
 
Woo, what latex-fu.
 
I get the feeling this won't be nice
 
@SamuelYusim Yeah, hence why I gave up on that approach :P
@SamuelYusim This book I have basically says that $C(XX^{\prime}) \subset C(X)$ is obvious. Any reason why it would seem that way?
Hey @KarimMansour
 
5:05 AM
obvious? I'm having to stop and think about it, at least.
is it something to do with the decomposition you mentioned earlier?
 
@Chris'ssis sorry, I meant $$\int_1^\infty x^{-\Gamma(x)} \text{d}x$$.
Apparently converges to around 1.04078
 
No idea @SamuelYusim
 
hi @Clarinetist
just had busy day with gf lol
tommrow doing physics all day and some algebra
 
Nice. I've been doing linear algebra all day
 
5:10 AM
@KarimMansour Are you familiar with column spaces and SVDs?
 
yeah
what is SVDs I am familiar with column spaces
is it short hand for something
 
Singular Value Decomposition
 
Maybe you could help me with this one :P
 
I did that in my matrix theory class
 
5:11 AM
1
Q: Using SVDs to prove $C(XX^{\prime}) = C(X)$

ClarinetistLet $C$ denote the column space. I would like to prove $C(XX^{\prime}) = C(X)$ for $X \in M_{n \times p}$, $X^{\prime}$ denoting the transpose of $X$. This answer suggests using singular value decomposition. So sure, $XX^{\prime}$ is symmetric and thus $$XX^{\prime} = PD\left(\lambda_i\right)P^{...

 
tommrow I will look at it oke because its my bed time :)
 
K 'night! :)
 
nights !
 
5:38 AM
It seems like (and this was one of my two plausible theories) that it is because the group is generated by an independent system of generators @SohamChowdhury
 
Do you have a copy of Aluffi with you right now? I figure I need a bit of help with page 77. I've been stuck on it for a long time.
If you're free, that is. :P
 
:) You can ask, (I am guessing it has something to do with free product and coproducts of $\mathbb Z$
 
I basically got lost when he says "Now for the general case."
How does he know to define $H^{\oplus A}$ as that subset?
That bit about $H^{\oplus A}$ having an operation induced on it and being a group needs a little bit of explanation.
 
So, like why not have it just every function $\alpha: A \to H$?
 
Yes, that's one.
Wait. Each $j_a$ corresponds to an $a \in A$, right?
I think I'm getting a sort of handle on this now.
 
5:55 AM
The idea is that every element is a combination of the generators, and from that perspective it does not really make sense to have infinite combinations. That is all that is seen from the structures perspective is finite combinations, so there is never a need to have infinite combinations (which is sort of what direct products are)
Although was that what was confusing you?
 
Hello@Paul
 
@Rememberme Hello
 

« first day (1756 days earlier)      last day (3268 days later) »