« first day (2581 days earlier)      last day (2738 days later) » 

17:01
Not uncountable dimension things ...
Not much, Faust. What have you learned so far?
starting to finnaly feel confident about homomorphisms and linear algerbra =)
Also discovered that Algerbraic topology is interesting today
You have a lot of things on your plate. By the way, if you ever make it to the end of chapter 8 of my book, you'll find all sorts of cool topology in there.
@TedShifrin is your book available at UK/HK?
Thats exciting to know, i will eventually but i have to finish prepping for the classes i have in sept first and none of them involve calculus cept an analysis class
(and should I be reading it?)
17:07
@LeakyNun theres an international edition that i got in canada thats mainly sold in the uk
@Faust7 nice
Leaky, you can get an idea of the content from the videos that are on line. The book has a bit more stuff (and exercises).
@TedShifrin I finally sat down and worked out the curvature, geodesics and isometries of $\Bbb H^2$ :P
@TedShifrin do you think I should read it?
@Balarka: It's even easier to compute with moving frames :D
17:07
I mean, I haven't even started university
Why don't you wait to learn stuff in university, Leaky?
well I have been reading an abstract algebra book
mainly to prepare myself and also for interest
GOing to start soon anyway.
@BalarkaSen jealous how come u get to do fun stuff and im stuck doing number theory?
@TedShifrin grumble grumble
17:09
@Faust7 All math is fun, including arithmetic.
@WillHunting well I do find it interesting
@Balarka: $\omega_1 = dx/y$, $\omega_2=dy/y$, so $d\omega_1 = \omega_{12}\wedge\omega_2$ and $d\omega_2 = \omega_1\wedge\omega_{12}$ tells you $\omega_{12}$ immediately.
Oh and yeah saw your outburst yesterday :P
Guys. Let $G$ be a finite group en let $a\in G$. Then define $\lambda_a\colon G\to G$ by $\lambda_a(x)=ax$ for $x\in G$. Show that $\lambda_a$ is the product of $[G:\langle a\rangle]$ disjoint cycles of length $\operatorname{ord}(a)$. I have no idea what they mean. How can a function be a product of cycles? How are those two things even related?
Sometimes one has to be a bit blunt and perhaps mean, Balarka.
2
17:09
I do not disagree with whatever you said.
@ShaVuklia $\lambda_a$ is a permutation
haha
@Sha: It's a permutation.
I did think the replies to what you said were annoying and somewhat arrogant, however.
@TedShifrin ugh, I wanted to say that lol
17:10
oh
I see, thanks @ted,leaky
LOL, you did, Leaky.
@TedShifrin But there are blunt and mean people who have no good reason to be so too, lol.
Actually, the ODE for geodesics is a bit tricky (as I did it in my notes), @Balarka, but that too is nice to figure out with moving frames. You chose the obvious horizontal/vertical frame and another frame tangent to the geodesic, and solve for the angle between them as you move along the geodesic.
@Faust7 Eh, if you know me any better you'd know that doing calculations is not really what I call fun :P Hyperbolic geometry is fun, on the other hand.
Jasper: I try to be super-encouraging in here, but sometimes I just cannot be.
17:11
Number theory is good stuff. How come you don't like it?
I used to like number theory in high school.
@BalarkaSen i dont mind it theres some interesting results and its not hard which is nice but id much rather be doing hyperbolic geometry
Then I went to college, I no longer liked it.
But now after college, I like it again.
@Balarka: I'm going to claim that this calculation is more insightful and less grundgy than the usual. (Unless you resort to what's known from complex analysis and know that the Möbius transformations are the isometries. Then you can do everything softly ... But ... how did you prove that Möbius transformations take generalized circles to generalized circles?)
Im mostly just said that theres only one class im taking in september that im really excited about taking
Anonymous
17:13
I used to like number theory a lot in school. But I forgot most of it now. :P
sad*
Maybe you'll be pleasantly surprised, Faust.
@TedShifrin I have heard of the term circline before. I wonder how common it is.
I've never heard it, Jasper :)
@Ted Oh I guess you are right that $K$ is easy to find with moving frames. Maybe I should work out the moving frames calculations of all this.
But then I wouldn't accomplish my world wide hate game with moving frames :(
17:15
@Faust7 I recommend this book: Elementary number theory by James Strayer. Even a high school kid can read it.
@Faust Strange. I actually quite like number theory.
Slowly you're going to crumble, Balarka.
I never much liked number theory, but in teaching algebra and writing my book I began to appreciate elementary number theory for its pedagogical strengths.
The book by David M Burton is horrendously expensive and not even properly updated after so many editions, a total ripoff.
@TedShifrin hope so but i doubt it ive already gone through like half the textbook for all but one my classes in September
If you see my profile page, I have complained that I am "Interested in number theory." but I have been forced into "Studying topology." @Faust7
17:17
?
seems odd
@TedShifrin lol, let's see about that
Number theory! \('-')/
I think a good mathematician should know logic, combinatorics, number theory, algebra, analysis, geometry, and topology.
don't u choose what u want to do?
@WillHunting diff.geom?
17:18
@LeakyNun Included, lol. Anyway, that was just a random list, lol.
Balarka will argue that Mike and I forced him to do all sorts of stuff.
And lol your profile doesn't say forced, but it does say "Interested in Number Theory. Studying topology."
And so I turned out to be falling endlessly through the geometry & topology rabbithole.
So why was Ted ranting yesterday?
@Daminark You have to fill in the blanks, man.
17:19
May 24 at 1:16, by Leaky Nun
and I have a question that I solved by exhaustion, but I would like to have a better proof: "every commutative binary operation on a set of two elements must be associative"
is anyone interested?
I learned lel and sick from the cool @Daminark
Woohoo, I'm cool now!
:D
lol
17:20
no he sucks
he's a normie
and he posts normie memes
Whos normie?
You're an algebraic topologists who still does point-set so I think you're more normie than I am
Point set topology gets boring after a while.
17:21
resigns from the adolescent games
I might need to summon @AlessandroCodenotti then
@Daminark algebraic topology is for normies
I would never claim to be an algebraic topologist
But I think point set topology was my favourite subject.
I am a higher category theorist
Okay good
17:22
It never was mine, Jasper, although it made me grow up a great deal when I took it.
Lol @Ted
I loved teaching it, though, to instill an instinct to check out examples/counterexamples.
(regarding adolescents games, not point-set)
@TedShifrin And after you grow up, you just grow old. =D
Well, at the time I was only 19, Jasper :P
17:24
[Chemistry] The 4 stage script is finally working. First 3 out of 40 batches calculations were submitted. Expect to finish them within 3-5 days
Well, since I believe in rebirth, after you grow old, you just grow up again. =D
There's actually a lot of drama not just in chat but also on the main site, sometimes, especially on meta.
Math people are quite fun, lol.
?
Drama?
Re topology, I'll grant that was my favorite part of analysis before measure theory, so there's that. Though I've heard (from sources you can guess) that stuff gets quite interesting when you work on model and infinity categories
I have been summoned
how may I help?
hi chat
It depends on who summoned you.
17:29
do you grant 3wish's?
How about 1?
@Faust7 Did you know that when English teachers keep looking at spelling mistakes for a long time, sooner or later, they will also spell wrongly, lol.
Hm, I don't think so
by that same logic there will come a day if i read long enough that i will spell correctly?
17:31
@Leaky regarding the associative/commutative stuff, I think the answer is that you can be just commutative in general
Both seem equally unlikly
@Faust7 Depends on what you read, lol.
So I don't know if you're gonna have a slick proof in the 2 element case
But in general, I never use spellcheck anywhere. It's not reliable. Grammar check is even worse.
Seems im doomed if i keep reading this chat then?
17:32
May 24 at 1:16, by Leaky Nun
and I have a question that I solved by exhaustion, but I would like to have a better proof: "every commutative binary operation on a set of two elements must be associative"
shrugs in doom
@Daminark alright
I actually do like the example of the commutative-only operation
@LeakyNun But there's not much to exhaust there, right?
@WillHunting right
17:34
But still you want a better proof, LOL. Fine, LOL.
Take two elements in a totally ordered set of words and concatenate them based on their comparison.
@WillHunting that's generally a fair request, even if brute force is short, it's quite unsatisfying
@Daminark Yeah, but sometimes after using brute force, it feels very satisfying to me, lol.
@Daminark it doesn't even map back to itself
@AlessandroCodenotti i have a tshirt that is nonorientable
I have a question I asked years ago, which you should ping me if you solve by non brute force.
9
Q: Existence of $\vee$ or $\wedge$ for non-monotonic functions

user4594This question is inspired by a discussion in chat with wj32. We allow for equality in the definition of increasing and decreasing and call a function monotonic if it is increasing or decreasing. If $f:\mathbb R\to \mathbb R$ is not monotonic, are there three points $x<y<z$ such that $f(y)<f(x),f(...

I have an answer there, and I accepted Christian Blatter's answer because it was the only other answer there, but it is essentially the same as my answer.
17:38
The answer is 4
@WillHunting what happened to the account?
@LeakyNun I'm taking a set of words, concatenating two words gives another.
@LeakyNun Deleted, just like many others, lol.
@Daminark oh
Not going to explain the motivation of this remark: dealing with University bureaucracy suuucks
17:39
@Semiclassical trust me working for a big company is 1000x worse
@LeakyNun This question seems trivial at first but is actually extremely hard. The simple answers all have some flawed logic there. I think they were deleted. In fact, it was wj32 who pointed out the errors in my initial attempted solution in chat or something.
You say that, but with a company the main task is convincing them to give you money
Payroll/salary etc
you dont want me to melt your brain with incompetence of managment
17:42
With a university the task is occasionally To convince them that you shouldn't owe them money
@LeakyNun Actually, that is the only real mathematical question I have ever asked on this site, the others were like soft questions.
For example say your manager decided one day that in order to enter your company vehicle you must climb in from the passengers side not the drivers side door and that this had henceforth become company policy.
It's especially annoying that the excessive management is partially responsible for the fact everyone owes so much nowadays
this would not even make my top 100 of ubsurd statements given by a manager in a very large and succeful company i workedfor for 10 years
It may make the top 1000 but not by alot... and it is a rather ubsurd statement...
In my case I'm dealing with an $8k tuition refund stemming from credits I took at the direction of my prof and other admin people had given
17:48
Can you imagine trying to explain to someone that, that statement was a stupid idea and a waste of time and them simply not caring? telling you its company policy and you MUST do it?
(Had to file an appeal for the refund, but the committee apparently doesn't meet until two weeks into the semester. So in order to register for classes I'll have to pay 8k out of pocket and take it on faith that they'll refund me)
Guys. Let $G$ be a finite group en let $a\in G$. Then define $\lambda_a\colon G\to G$ by $\lambda_a(x)=ax$ for $x\in G$. So I’ve shown that $\lambda_a$ is the product of $[G:\langle a\rangle]$ disjoint cyckles of length $\operatorname{ord}(a)$. Now my book is asking the following: Show that $\lambda_a$ is an odd permutation iff the order of $G$ is even, and the order of $a$ is divisible by the highest power of 2 that divides the order of $G$.

So I’m guessing what we’re looking for is for the order of $a$ to be even in any case. And then we want $[G:\langle a\rangle]$ to be odd. I’m not rea
(I am not happy about this.)
@ShaVuklia youtube.com/…
watch that video lecture also highest power of 2 not needed
that's a 47 min video...
17:53
your forgeting the trival case
im not good at explaining things
@Sha: Once again, have you tried lots of examples to see what's going on?
@Ted uhm nopes. but I'll try some examples then
I will repeatedly give you that advice :P
Demonark: When is Eric's lecture on forms?
Hi @TedShifrin
Hi @Danu
All done?
18:02
Naw
Some shady signs and factors in Besse's book holding me up today
pretty sure he done goofed somewhere
they
What? You mean books aren't flawless? I'm shocked! [waits for Hippa]
For instance, at some point they have the following:
$n a+b+c=a+nb+c=a+b+nc=x$ and hence $a=b=c=2/(n+2)x$ ($n\geq 2$)
The factor 2 is wrong
Yup, even I can tell that.
and it comes in crucially in a bunch of later computations
to cancel a 1/2
Oh, that's not good.
18:05
I wonder if it's REALLY needed in the final computation (I'm doing the prep work for that)
It was gonna happen today but he's sick
But I really hope it isn't.
If it is, I'm in some trouble.
I wonder if there's a list of errata somewhere on the web. I know Bryant has taught courses out of the book and he's super careful.
there are errata in the back of the book but just like... 20 :P
Right now is the calc of variations lecture but the guy who is lecturing is... Not that good, I'm just gonna read on my own
18:05
We can't all be as good as you, Demonark.
I mean I don't think I'm particularly good but when he lectured before he got tripped up on stuff that made it suggest he was really malprepared
(is that a word?)
Lots of full professors do that every week.
It depends how familiar the prof is with the material too my dif geo prof was amazing in that regard
@TedShifrin You figure Bryant might have errata of his own? Google reveals no page for errata.
Even when I was learning stuff I was teaching (like probability), I'm pretty sure my students didn't complain that I was unprepared and messing up.
18:09
never used notes and would make the class come up with a valid approach to a problem or theorem and then use their idea to solve it
That's fantastic teaching, Faust.
itwas amazing
@Ted OK, so I’ve found an example. Let $G=\mathbb Z/10\mathbb Z$. Let $a=\overline 5$. The order of $a$ is 2, and the highest power of 2 that divides the order of $G$ is two. And indeed, we have an odd permutation: $\lambda_a=(1~6)(2~7)(3~8)(4~9)(5~10)$. But I have to admit that I'm still lost concerning the proof:P
Sometimes working a few examples will give you insight, Sha, so try more complicated examples.
He really love the topic though... i think that makes a big diffrence
18:10
@Faust7 regrettably it sounds to me like that'd cause major slowdown (unless you've got students that already know the material hehe)
I actually have thought about this question before, Sha, but I don't remember it.
@Faust: I basically loved the material in every course I taught. Walking in bored and contemptuous of the course never helps anyone.
@danu it was a soft course with no defined end point the idea behind it was to give us an appreciation of dif geo and he succeeded profoundly in that regard
Oh that's super neat
I certainly hope he had Gauss-Bonnet for surfaces as a goal.
My own appreciation of classical diffgeo (surfaces n stuff) has recently gone up a LOT through self-discovery too
18:12
@ted we made it there with a few weeks to spare
You guys should appreciate some of my exercises for the undergrad course — I totally came up with them without help :P
Good, Faust, me too. Then I did hyperbolic geometry for 2 weeks. But most people teaching the course don't even get to Gauss Bonnet. Makes me as upset as teaching multivariable calculus without getting to Green/Stokes/Divergence Theorems.
Hi chat
yeah we did hyperbolic geometry for about 2 weeks then we drifted off into wierd things he found intresting for the last 2 or 3 classes
@ShaVuklia don't try nice examples. find strange examples. groups that have few properties.
Hyperbolic stuff was fun
18:13
Like what @Faust7?
@ShaVuklia try a group of order 7 or 5 or 11
lol
the group has to be of even order @Faust7
@danu honestly i didnt understand the last 3 lectures almost at all... lol
are any of the algebraists here today?
i do have the notes though!
18:15
@Faust7 But what was the topic?
@LeakyNun right I'll try a different group then. not show what you mean by 'few properties'
@Daminark that stuff gets even cooler, maybe check out bridson and haefliger or Danny calegaris notes
" random things our prof found intresting"
Eric, aren't you supposed to be preparing your lecture? :D
he went all over the palce i dont belive it was in one particular area i can look it up but i didnt understand it very well
18:15
@ShaVuklia for example, $\Bbb Z_{10}$ has the nice property of being abelian
@Ted I was meant to give it today but I'm quite sick :(
oh like that @Leaky
Oh, dear, Eric. I'm sorry. :(
@ShaVuklia why?
I lost my voice and have a doozy of a sore throat
Unfortunate cause I was looking forward to lecturing
18:16
@LeakyNun that's part of the theorem right. we need a group of even order for the permutation to be odd
Oh, yeah trying to lecture in that state would be quite painful
25 mins ago, by Sha Vuklia
Guys. Let $G$ be a finite group en let $a\in G$. Then define $\lambda_a\colon G\to G$ by $\lambda_a(x)=ax$ for $x\in G$. So I’ve shown that $\lambda_a$ is the product of $[G:\langle a\rangle]$ disjoint cyckles of length $\operatorname{ord}(a)$. Now my book is asking the following: Show that $\lambda_a$ is an odd permutation iff the order of $G$ is even, and the order of $a$ is divisible by the highest power of 2 that divides the order of $G$.

So I’m guessing what we’re looking for is for the order of $a$ to be even in any case. And then we want $[G:\langle a\rangle]$ to be odd. I’m not rea
Hope you feel better
@ShaVuklia you also need to show that $\lambda_a$ is even if $G$ is odd: that's what "iff" means.
Thanks
18:17
@WillHunting I gave up
I have had a virus that did that to me a few times in my life, Eric ... including when I was teaching. I used lots of lozenges and made the class be very quiet.
Eat some chicken soup and sleep m8
@LeakyNun Yeah, maybe there isn't one after all. =D
Or chicken inf, that works too
That's the plan @Faust
18:18
@LeakyNun right
extends Demonark no-pun period for another month
i tried eating some chicken oxo in hot water once as a substitute but it went terribly
@TedShifrin And the way to make the cllass very quiet, is to make them have the same sore throat you are having.
Come on, that was good!
It was a bit of a stretch
18:20
Im having a Glass of guiness
Although I guess it's p good if you pronounce sup like soup, I don't personally
I do.
But maybe I'll change.
Wait how do you pronounce it?
He says 'sup, like wassup?
I say sup like wassup
18:20
ok @Leaky, well $\implies$ is easy then, given the previous exercise. As soon as $G$ is odd, we know that the cycles must be of odd length, so the permutation is even.
@EricSilva I pronounce sup like soup
Sniped
Amazing
10/10
18:20
@ShaVuklia what is the previous exercise?
@EricSilva I take sup to mean supper, not wassup.
Lol @WillHunting
i take sup to be the roof or in merican ruf
I don't say supper
I've shown that $\lambda_a$ is a product of $[G:\langle a\rangle]$ disjoint cycles of length $\operatorname{ord}(a)$ @Leaky
18:21
Eric is getting sniped by Ted
Epic moment
Hardly, @Balarka.
@ShaVuklia oh. alright.
I will say I prefer either pronunciation to lub
Sally does that and it's just ugh
facepalm
I think lub sounds funny
18:22
lol
Lub glb
Like you can't say glb without it being awkward, and mixing is just sacrilegious
Pronounced lub glub
<-- agrees with Eric sans snipe.
any pronunciation of glb would just be... glib
18:24
puts Balarka on Demonark's pun-detention program
Question: Give an example of a ring with unity $1 \ne 0$ that has a subring with non-zero unity $1' \ne 1$.
Answer: Consider $\Bbb Z_2 \times \Bbb Z_2$, whose unity is $(1,1)$. Then, $\Bbb Z_2 \times \{0\}$ is a subring with unity $(1,0)$.
Further thoughts: this is possible because there can be $fa=a$ with $f \ne 1$, which is because $a$ might not have an inverse.
@Ted More like... pun detonation
big bass-boost explosion sounds
ba dum tss
I wish I had that background sound to puns irl
The min max theorem for characterizing eigenvalues would be hilarious if I heard it called the glb lub theorem
@Daminark The best teeth in the game's explosion sounds are the best
18:26
@Leaky oh I think I'm almost there. Assuming the thing on the right side of the iff, we know that $G$ is even and $\operatorname{ord}(a)$ is even. So this thing with the highest power of two is going to give us an odd index. I just have to figure that out and then I'm done:P
Oh this reminds me
@ShaVuklia how do you know that $\operatorname{ord}(a)$ is even?
@Akiva when you see this, in my last lecture I actually had the opportunity to use Proof: [DATA EXPUNGED]
My life is complete
because the order of $G$ is even, and we have that the highest power of 2 that divides the order of $G$ also divides the order of $a$
Bye, all.
18:27
@ShaVuklia alright.
bye @Ted
36 mins ago, by Sha Vuklia
Guys. Let $G$ be a finite group en let $a\in G$. Then define $\lambda_a\colon G\to G$ by $\lambda_a(x)=ax$ for $x\in G$. So I’ve shown that $\lambda_a$ is the product of $[G:\langle a\rangle]$ disjoint cyckles of length $\operatorname{ord}(a)$. Now my book is asking the following: Show that $\lambda_a$ is an odd permutation iff the order of $G$ is even, and the order of $a$ is divisible by the highest power of 2 that divides the order of $G$.

So I’m guessing what we’re looking for is for the order of $a$ to be even in any case. And then we want $[G:\langle a\rangle]$ to be odd. I’m not rea
See you @Ted!
Bye @Ted
@Daminark is summoned
18:28
@Daminark lool
I uh what was the context
@Leaky OMGOODNESS. I SEE IT.
:D:D:D
@ShaVuklia nice
thanks for NOT telling me the answer :P
@Sha "the answer"
18:30
ohhhh haha. yea whatever
:P
@ShaVuklia how did you show that $\lambda_a$ is the product of $[G:\langle a\rangle]$ disjoint cycles of length $\operatorname{ord}(a)$?
lol I wrote down a messy proof
let me share it
@Akiva there was this theorem I was gonna give but the TAs told me to skip it, and then one of the students asked me to write it and I was like okay, and then wrote the proof was expunged
@ShaVuklia more than happy to help you proofread it
Say $\vert G\vert=n$. And say $\operatorname{ord}(a)=k$. I wrote out the elements in the following way:
\begin{align}
ax_1,a^2x_1,&\dots,a^kx_1\\
&\vdots\\
ax_n,a^2x_n,&\dots,a^kx_n.
\end{align}

We know that $a,a^2,\dots,a^k$ are all different elements. So we know that for each $i$, $ax_i,\dots,a^kx_i$ are different elements. Now I basically have to show that those sets (the rows) form a partition. So if $a^mx_i=a^nx_j$, I need to show that the two sets are equal. Obviously $a^{m+1}x_i=a^{n+1}x_j$, and so on. It might be useful to use modular arithmetic here (even though, technically, it's
18:31
@Daminark You were teaching a class
@Daminark what was the theorem?
Yeah
@Daminark This is beyond funny
100/100
It was taking the deMoivre Laplace integral theorem but extending one of the bounds to infinity
with different order I mean starting from a different element. the order an sich is the same @Leaky
and I'm not very rigorous when introducing $i,j,m,n$, but that's just because I was too lazy
18:36
@ShaVuklia might I suggest a similar-minded but more rigorous approach?
yes of course, please
what do you associate with "partition"?
like, it's a set of sets, where each element of your original set is on exactly one of those sets
(those two concepts are mentioned together in Ch.0 of the book I'm reading)
so $A=B$ or $A\cap B=\emptyset$
18:37
E.R.
and the partition consists of all the element of the original set
@AkivaWeinberger I was trying to let her figure out by herself
though it's technically the sets in the partition that contain the elements
what's E.R. tho?
@ShaVuklia you must know what comes with partition
just like homomorphism comes with quotient groups
oh
equivalence relation?
18:39
@ShaVuklia bingo
Technically I didn't give away the answer
ohhh lol
now I know what ER meansXD
@AkivaWeinberger alright
I was trying to be cryptic
lol you succeeded at that @Akiva
18:40
kripˈtäɡrəfē
@ShaVuklia so... what to do now?
probably show that those cycles form equivalence classes
and find the equivalence relation :P
@ShaVuklia what is the relation?
@Faust7 ˌkriptəˈnaləsəs
I'd say $x\sim y$ if $a^kx=a^ly$ for some $k,l$
18:42
@ShaVuklia you can do better
you mean I have to phrase it more rigorously? or try something entirely different?
@ShaVuklia I mean there is a neater way of putting it
I wouldn't know, I'm afraid @Leaky
@ShaVuklia it's a trivial modification
18:48
oh
$x=a^ky$
I guess I'll first show it's an ER then @Leaky
continue :P
I was having dinner @Leaky, but I don't think there is a better proof than bruteforcing it like you did
@AlessandroCodenotti oh, ok thanks
18:51
@Leaky ok I've done that. (not going to write it out)
@ShaVuklia ok
ok, so now we have a partition by the virtue of the ER :P
I guess that's it then?
that was actually a good remark. when I want to show that we have a partition, I might try to think in terms of ER
Every equivalence relation induces a partition, and every partition induces an equivalence relation.
lol yep
@ShaVuklia well you still have to show that the partition is actually a cycle
and that there are $[G:\langle a\rangle]$ that many partitions
18:56
@LeakyNun omg I'm dying. alright, let me do that
wait, just to be a bit precise on the terminology here
the partition itself is not the cycle right
it's the elements (which are sets) of the partition that are the cycles
right?
lol yea ok just to be sure
I got messed up in my terminologies, sorry
we call them the cells of the partition
This shitpost is too good

« first day (2581 days earlier)      last day (2738 days later) »