« first day (2589 days earlier)      last day (2728 days later) » 

18:00
that's a good question :/
wow that was nonsense
I got nothing :(
How did Graubner solve it ? I don't think Maple or Mathematica can crunch that
Heya Dodo !! :)
Hey Ted! :)
@Danu: You could certainly take the hermitian-orthogonal complement of $\ell$ in $D$, but this moves us out of the algebraic category. I've never seen any such notation for that, though.
howdy @Brody ā€” long time!
@TedShifrin It's the only non-degenerate thing around, right? So it seems the only chance at orthogonal working
18:14
Yup.
And honestly, I have no idea what else to even think about
ALSO
OMG OMG OMG OMGOM OMG DID YOU WATCH DELPO
HOLY F*** THAT WAS AWESOME
And you saw anon's derivation of the double cover? I am totally fluent with this analysis on $\Lambda^2 \Bbb R^4$ with the Hodge star, but never think of generalizing that.
Apparently he had a bad virus. But I'm sad that Thiem fell apart.
I saw it but I didn't follow it tbh
Thiem didn't really fall apart though... The fourth and fifth set were fiercely competitive
It's complexifying the usual way to get $\tilde G(2,4) \cong S^2\times S^2$ in terms of eigenspaces of $\star$.
I kept screaming to my friend: JUST GIVE ME ONE FOREHAND DAMNIT (me being DelPo)
18:16
Well, he sorta fell apart with nerves when he served for it. When he got the break points, DelPo hit two killer aces, so not his fault :P
Ah, the 5-3 in the fourth set, yeah maybe.
But he played some seriuosly good points in CLUTCH situations
remember that one game where he was 0-40 and had an epic point on one of them, went on to save all three?
I think Federer's in for a struggle against DelPo.
Which he?
Hi @Ted! Good to see ya
18:17
Thiem
I hope DelPo will just be too tired to compete haha
But Federer seems pretty weak, my money's on Nadal.
I can't believe how weak the second half of the draw is
Kevin Anderson, honestly? Ridiculous
Carreno Busta?!
I like Schwartzmann though :)
I think if Fed makes it through DelPo, he will totally bring it and beat Nadal.
Schwartzman is down in flames. Down 2 sets and down a break in the 3rd.
@TedShifrin Really? Hmm
How's your math learning going, @Brody?
18:19
@TedShifrin Yeah, I know :( Sad to see because Carreno Busta is so... bland. No flavors to him :P
Pretty cool for the US that we have 4 women in the quarters, one in each match.
Hey @TedShifrin !
@TedShifrin Snail's pace, but it's going
Women's tennis... Is just so weak in comparison to the men's game. I can't really bring myself to watch it without getting bored :\
Yeah, you're definitely not going to be pushed hard where you are, @Brody.
@Danu: Careño Busta just beat Schwartzman. Done.
18:20
Where are you, Brody? :P
@TedShifrin I saw :\
Also @Brody funny you mention mastering some languages in your profile. Where are you from?
I'm not such a snoot about the women. There are some talented hard hitters. And they're getting their heads together. And some of them play the net, too!!
He's from near where I used to be, @Danu.
@TedShifrin I love Ostapenko, becase she's approaching the pace of the men's game.
I also find her very attractive, somehow. :)
Maybe because her tennis is so nice ;D
@TedShifrin Never really thought about it but that's certainly true
Luckily you have Ted ;D
Nah, he missed his chance to get stuck with Ted at UGA :P
But I'm still here to torture people a little.
18:23
We're all masochists here :)
@Danu currently at Kennesaw State University with the Owls, "hootie hoo" :)
Well, @Brody, any time you wanna get back to the serious multivariable stuff, let me know :P
@Wildcard Thank you
Hi, DogAteMy. ... Thinking about deck/covering transformations for the Klein bottle, I saw? That's something I spent serious time on with the grad students studying for their topology qualifying exam.
The second person plural should be doubleyou
18:25
The Owls? lol
@Danu usually reluctant to divulge but... Russian-Asian and Latino heritage
@Brody Nice
Diverse haha
So do you speak Russian/any Asian language/Spanish/Portuguese? Or is that your goal? ;)
@Danu Diverse? Mmm not exactly. It's complicated lol
I'm 3/4 Russian, 1/4 Polish, but my heritage didn't get me born speaking either Russian or Polish. :(
I'm currently slowly working towards some kind of understanding of Greek... Not really studying though.
Just immersing myself (or being forcefully submerged by my Greek friends haha)
18:28
When you say Russian-Asian do you mean like Siberian
Not so useful for reading math journal articles, @Danu :)
DogAteMy: Maybe I need to call you Husky.
Yeah, I guess I should go for Russian. But I have to say that the Russian style of math writing does not charm me at all, as far as I can tell. At least not the research articles, I've found some OK books.
Very little math writing charms me.
Much as I adore Blaschke, I didn't find his differential geometry book in German "charming."
18:29
@TedShifrin Yes of course! My current job situation makes it a little hard to find good self-study time, but it will come eventually
Hirzebruch <3
His lectures are so great. I love it
@Brody: I'm here if and when you need me.
But his book style is pretty dry, @Danu.
His articles are glorious
at least the ones I've read
I haven't read the book I must admit
@TedShifrin Is this true: If $f:I \rightarrow \Bbb R$ where $I\subset \Bbb R$ is interval, is a continuous one-to -one function, then $f$ is either strictly increasing or strictly decreasing.
I really should though. Though I'm a bit afraid of it being sorta outdated.
The treatment of HRR and stuff
18:31
@Silent: Yes, it's true.
but it probably contains a ton of beautiful ideas
@Danu: I still love reading some of Chern's seminal works, despite modern style having obliterated his viewpoint.
@TedShifrin for any kind of interval, closed or open or half-open?
I wish we read the original sources and classics a lot more in mathematics.
@Silent yes
18:32
Sure, @Silent.
@TedShifrin They're sadly... unreadable to me?!
I really tried for a few hours to read the one on Hermitian blahblah introducing his classes. But I found it very unrewarding :(
Well, if Bryant and his students weren't the only ones teaching moving frames (besides me, but I don't count) ...
Yeah, that might be the problem.
It's so much more geometric to use moving frames. I truly cannot understand why the modern world is so scared of differential forms.
Maybe in Europe there are folks using them/teaching them, but I don't know them.
@Danu: If you're going to be successful at reading Bryant, you're going to need them for that, too.
And a bit of Cartan-Kähler theory.
18:34
@Danu English and Spanish from the home. Studying Russian and Chinese (no Chinese heritage however). Greek was always a candidate for its legacy, but probably will not srsly study it
@TedShifrin I don't think so.
Actually, I have a small anecdote about it. My supervisor did a weekly "research tutorial" for his students this semsester and asked us for proposals on what to discuss. I asked about moving frames and he just laughed at me and said "well, you took a course on differential geometry right?". I still don't get the point/joke...
It was never brought up again after that. I certainly was too afraid to ask :D
It should be said here once and for all that we are going to glos over several points that some books spend pages proving, usually by complicated induction arguments broken down into many cases. For example, [...]. How do we know [...]? The student will probably say this is obvious. Some authors spend considerable effort proving this. The author tends to agree here with the student. Proofs of this sort he regards as tedious, and they have never made him more comfortable about the situation. However, the author is the first to acknowledge that he is not a great mathematician. In deference to
@Brody: Keep up the languages :) I learned Latin and French in high school, majored in French (essentially) and learned German and a year of Russian in college. :) Now I regret missing out on learning Spanish thoroughly.
Ch.39, A First Course In Abstract Algebra, John B. Fraleigh
@Brody Cool
18:36
I"m not a huge fan of Fraleigh, @Leaky. Plus he once had an error and I got a rude email from a reader of his saying my algebra book was wrong because I had it right and he had it wrong :D
@TedShifrin what was the error?
It comes late in the book. One proves that a constructible real number $\alpha$ must have degree $2^n$ (for some $n\in\Bbb N$) over $\Bbb Q$. He suggested that the converse was true. It is FALSE.
@TedShifrin Hahahaha that's the real reason you dislike the book, I'm sure :D
There are other reasons I'm not fond of it, but there are worse choices. I don't want to think about a coherent discussion here. For strong students, I vote for Artin (at the undergrad level).
@TedShifrin I've skimmed through that chapter (and basically every chapter), wait a minute...
18:39
Hi Demonark
It was something in an old edition, @Leaky. I'm sure it finally got corrected. I don't have any Fraleighs on my bookshelf at this point.
Hey there!
But I did get a rude email from someone. It was amusing.
@TedShifrin my edition is old enough (2003)
That might not be old enough.
7th edition
18:39
Probably not old enough.
Mine went back to the 80s, I think.
@Daminark \o
What'd you recommend for me to learn algebra from Ted?
I think I needda do it at some point
In the early editions he did have a chapter introducing simplicial homology to show a geometric manifestation of quotient groups. That was sort of a nice idea.
nice
Aren't you just going to take a graduate course, @Danu? And then learn commutative algebra? You need that. Modules, localization, etc.
You don't need to care about Sylow theorems (in the US Ph.D. students have to do qualifying exams, but I presume you don't).
18:41
Yeah, probably. But I actually never really learned the basics. Though I did read through 2/3's of a basic algebra book of Vinberg
@TedShifrin No, I'll have a Master's instead, thank you ;D
I like Dummit/Foote, and it has good exercises.
In the US every Ph.D. student takes qualifying exams. MA is irrelevant.
But Sylow is so fun!
Define $S_n=[\frac{1}{n},n]$ for $n\in\Bbb N_{\ge 1}$. Then $S_n$ under normal multiplication is not an abelian group, but rather a groupoid with commutativity?
Group actions are terrific, Demonark, but the Sylow theorems are pretty arcane except for 5% of mathematicians.
@Brody not closed: $n^2 \notin S_n$
18:43
Group actions are 10/10
I don't know what that notation means, @Brody.
@TedShifrin probably interval
@TedShifrin Yeah but I feel like that is because they have no other quality control since you can enter after the BSc.
yep, sorry
18:43
Also it's a MSc. ;)
When I want to explain something about the beauty of mathematics to laymen I usually start with group actions and symmetries of polygons
2
@LeakyNun Sorry confusing terms... groupoid i.e. not a magma = no totality
In part, yeah, @Danu. But I like the viewpoint that you want to insure everyone has some general knowledge in analysis, algebra, topology (or different things for applied folks).
I also like that. But that's what the MSc. here does too. It's just that I don't havea MSc. in math :(
@Brody A groupoid is a certain type of category
Then closure fails immediately, yeah.
18:44
@Brody right, groupoid doesn't need closure (cc @TedShifrin)
In the math degree, you have to do analysis and algebra courses too
OK, I'm outta here. Hi/bye, @Tobias. Oh, how is your class going?
Oh, I never remember what those things are.
@TedShifrin Going alright. I think I might be going a little too fast, so I will be doing a poll tomorrow to find out
Def: Group is a groupoid with 1 object
What do the TAs say about the homework, @Tobias? That's what's crucial. Or do your students not have to do homework?
18:45
@Brody not associative: $n \times 1$ and $1 \times n$ are defined but not $n \times n$
Nice to have a non-math chat, Ted :)
Good luck moving soon, @Danu!!
@TedShifrin We have only had one week of exercise sessions, and they only hand in the first set of homework this week (well, some of them did last Friday), so too early to tell
Glad you gave me an excuse to see München a little bit.
But the exercises and hand-ins are the same as previous years
18:46
2 months
Ah, so, yeah, too early, @Tobias.
2 months ... I forget how late the German schedule is.
Bye, all, for now.
@TedShifrin Yeah, it was nice to meet up :)
Ch. 32: Geometric constructions
32.1 Theorem: If $\alpha$ and $\beta$ are constructible real numbers, then so are $\alpha + \beta$, $\alpha - \beta$, $\alpha \beta$, and $\alpha/\beta$, if $\beta \ne 0$.
32.5 Corollary: The set of all constructible real numbers forms a subfield $F$ of the field of real numbers.
32.6 Theorem: The field $F$ of constructible real numbers consists precisely of all real numbers that we can obtain from $\Bbb Q$ by taking square roots of positive numbers a finite number of times and applying a finite number of field operations.
I haven't defended my thesis yet @Ted :P
See you!
18:47
@TedShifrin probably the precursor of 32.8
@orlp hi
@LeakyNun damnit I was lurking!
:P
@LeakyNun Oh of course. True cannot imply false...
@LeakyNun will you disappear again when your college starts again? :(
@orlp not really
this is why I entered, was curious what you were talking about :)
18:51
@orlp I do frequent this room.
@LeakyNun myself I've been preoccupied with some neural network stuff lately
@orlp I see
haven't visited PPCG in a while
although I also really like a mathematical card trick I recently read
@LeakyNun maybe you can figure it out
@LeakyNun Hi
18:54
@Mr.Xcoder hi
@LeakyNun You and I are partners, and you give a deck of cards to an audience member, who then hands you ANY 5 cards. You choose 4 cards of those 5 and hand them to me. We do not communicate any further than that, but we did discuss and learn the trick beforehand. I look at the 4 cards you gave me and I proceed to call out the 5th card.
How did I do that?
@orlp modulo arithmetic :)
that's too vague :P
how is it possible :o
@LeakyNun Give it some thought :)
18:56
there are 52C5 possible combinations that the audience gives you
see if you can figure it out
keep in mind that the trick needs to work with humans
you have 5C4 choices for the four cards
and I see 4 cards, which is something like 52C4
and 52C4 x 5C4 is hardly 52C5
@LeakyNun there's more to it than just the choice of cards that you give me
if the cards are ABCDE
I can give you ABCD, or BDCA, or DBEC, etc
right
but essentially you are using 4 cards to encode 5 cards
essentially I'm using 4 cards with order to encode 5 cards without order, without allowing duplicate cards
19:00
what would make it work is if ABCD and BACD are considered distinct
oh, right
we're partners after all
you can hand me the cards in whatever order you please
or you can call them out vocally
or write them on a blackboard, or whatever
Another thing I notice: you get to pick which card is the fifth, which should make it easier, I think.
yes
@El'endiaStarman Hi! How is Tetris going?
@Mr.Xcoder Hey, it seems like we're getting closer to having all answers written and ready to post.
19:04
@El'endiaStarman Good for you. Good luck! Looking forward to see the new most upvoted answers in History!
52P4 = 6497400
52C5 = 2598960
so we allow overlapping but only 2.5 cases to 1...
@orlp I swear if it is one of those algorithms with like 100 cases
@LeakyNun it's not
it's actually a fairly simple algorithm
any hints?
choose the 5th card wisely
lol you don't say
19:13
alright a stronger hint?
sure
I tried choosing the largest card but it wouldn't work
there are 4 suits, but 5 cards to choose from
So if there are exactly two cards of a suit, you can pick the duplicate and leave the other at a fixed position
you can specify that the suit of the missing card is the suit of the first card
In fact, you can always have the top card give the suit
19:14
@LeakyNun that could be an idea
now I have to use 4 cards to encode 13 numbers...
@LeakyNun Except if you give the suit like that you can't be sure to be able to use that card for additional information
except you should probably know which of the two you would pick from any pair of values in the suit
which cuts it in half, leaving exactly 6 possibilities (i.e. $3!$)
that's interesting
@Steamy do you know if there is any way to download a Dutch math book? I'm looking for Inleiding in de statstiek by Fetsje Bijma
but you can't just pick the smallest one, or else there would be 12 possibilities for A
19:21
as far as I know, b-ok.org isn't of much help
@LeakyNun Yeah, you need a more detailed system
@ShaVuklia al mijn boeken waren in het engels, misschien kun je het aan medestudenten vragen?
@orlp ze kopen al hun boeken:l
de rekening was 505,- voor dit semester
likeee jesus.
although I guess we shouldn't talk dutch here, my bad :P
19:24
@orlp this isn't TNB
I guess I have to buy it then:l....
@TobiasKildetoft you guys are on the right path though
@ShaVuklia Nop, Nederlandstalige boeken zijn sowieso heel moeilijk online te vinden...
"if the parity is the same, choose the smaller one; else, choose the larger one" then 1 would give you 3,5,7,9,11,13; 2 would give you 1,4,6,8,10,12
@orlp this is better
but problem is, what happens if there are three cards of the same suit
@LeakyNun You ignore one of them completely
19:31
no problem, since it is just your choice
Wat een gruwelijk Nederlandse naam ook, "Fetsje Bijma" :P
@TobiasKildetoft right
German "sein" is Dutch "zijn" but "so" is "so"?
@SteamyRoot Does gruwelijk mean awful (or awfully in this case?)
Yup :P
@LeakyNun sowieso is 'no matter what'
19:32
very close to the Danish word for it, gruelig (except it is a somewhat old-fashioned word in Danish)
@orlp but it is so+wie+so right
@LeakyNun nope
or am I completely making it up
although the usage of sowieso here is somewhat slang
oh!
wiktionary: Borrowed from German sowieso.
that explains the lack of "z"
11 mins ago, by orlp
@ShaVuklia al mijn boeken waren in het engels, misschien kun je het aan medestudenten vragen?
also I recognize vragen might be German fragen (to ask)
19:35
yes
waren is exactly the same with German
anyway, here's how the original scheme worked
kun should be könn = can
if you order the cards like A 2 3 4 5 6 7 8 9 10 J Q K, wrapping around K back to A
and you're given two cards of the same suit
adding 13 to odd numbers
oh
you can view A as 14
19:36
you can always choose one such that the other one is at most 6 steps away
@orlp oh, that's neater
for example if you have J and 2
J - Q - K - A - 2 is 4 steps
it's modular distance, at the lack of a better word in my vocabulary
so from the 5 cards there must be a duplicate suit
you choose two cards from the duplicate suit
figure out which one comes 'first', and choose that one as your first card, and the other one as the hidden card
now all you need to do is encode up to 6 steps
another way to look at it is as a graph where the nodes are the 13 cards and there's an edge between adjacent consecutive
you do that through defining an ordering on the cards, and choosing one of the 6 permutations of the three remaining cards
@orlp Right, hence my referral to the $6$ being $3!$
19:40
@TobiasKildetoft So it seems that I'm good at causing confusion, as now with the KL polynomials. Like the last question about Stroppel's paper and parabolic subgroups, do you recall?
and if you do numbers first, suit second, and you just remember ABC = 1, ACB = 2, BAC = 3, BCA = 4, CAB = 5, CBA = 6
in which case the graph is just the 13-ring and the graph diameter is 6
it's definitely doable by humans
@Bubaya Yeah
Except I don't think there were any parabolic subgroups in that
people, stop the overkills :P
19:40
@LeakyNun pretty neat trick though, right?
pfff. i like graphs
there's just enough freedom there to make it work consistently
@orlp indeed
however
in the case of duplicates
for example a hand with 4 of 1 suit
we have multiple possibilities of encoding a 5th hidden card
so we're definitely not using up the maximum transmission bandwidth that these constraints allow
@LeakyNun so the question is: how much information (in bits) are we actually transferring, and what would be the maximum we could transfer?
@TobiasKildetoft No, not that KL question, but a few weeks ago, I asked that question (2252854) about category š¯“˛^š¯”­, where a paper by Stroppel caused very much confusion. In the end, it was a notational issue again, but that day, it took really long to understand that.
19:43
@Bubaya Ahh, I see
@TobiasKildetoft But thank you for being also confused!
@Bubaya No problem :)
can one build an infinite word out of $a$ and $b$ not in the form $pq^3r$?
In mathematics, the Thueā€“Morse sequence, or Prouhetā€“Thueā€“Morse sequence, is the binary sequence (an infinite sequence of 0s and 1s) obtained by starting with 0 and successively appending the Boolean complement of the sequence obtained thus far. The first few steps of this procedure yield the strings 0 then 01, 0110, 01101001, 0110100110010110, and so on, which are prefixes of the Thueā€“Morse sequence. The full sequence begins: 01101001100101101001011001101001.... (sequence A010060 in the OEIS) == Definition == There are several equivalent ways of defining the Thueā€“Morse sequence. === Dir...
@LeakyNun You mean not containing any cube of a word?
19:52
@TobiasKildetoft right
Yeah, the Thue-Morse sequence is probably the most well-known cube-free sequence.
thanks
In mathematics, the Kolakoski sequence, sometimes also known as the Oldenburger-Kolakoski sequence, is an infinite sequence of symbols {1,2} that is its own run-length encoding and the prototype for an infinite family of related sequences. It was initially named after the recreational mathematician William Kolakoski (1944ā€“97), who discussed it in 1965, but subsequent research has revealed that it first appeared in a paper by Rufus Oldenburger in 1939. == Definition of the classic Kolakoski sequence == The initial terms of the Kolakoski sequence are: 1,2,2,1,1,2,1,2,2,1,2,2,1,1,2,1,1,2,2,1,2,1,1...
thanks
@orlp How much could we transfer in that exact setup? Or a similar setup with no requirement that the card be guessable?
19:55
@Wildcard I'm interested in both
although the former seems a lot harder to analyse than the latter
Does real-life math chat-room exist?
@LeakyNun what do you mean with 'real-life math'?
@orlp no, real-life (math chat-room)
ah

« first day (2589 days earlier)      last day (2728 days later) »