« first day (3088 days earlier)      last day (1932 days later) » 
02:00 - 23:0023:00 - 00:00

11:07 PM
Hmmm, neat.
So whenever you transition from one face to another, you alternate between high and low.
 
Mobius bagel
 
@Rithaniel No, Hopf bagel
Möbius bagel looks slightly different
(though it's made by basically the same idea)
It would be one piece instead of two
 
Ah right. I love both concepts just the same.
 
But the reason I bring this up is because the crust of each half is a Seifert surface for the Hopf link
The circles are each Villarceau circles
 
11:12 PM
Ah, hmmm, so what characterizes a Seifert surface?
 
Look at those youngsters with fancy 3D renders, back in my days I had to learn how to draw to think about Seifert surfaces
 
@Rithaniel A surface whose boundary is knotted or linked or both
Oh and also they have to be orientable I think
 
Given a link are its Seifert surfaces equivalent up to some reasonable notion?
 
Another image of a Hopf thingy
 
Okay, so a Mobius strip is a Seifert surface as well?
 
11:16 PM
Well I said orientable
So no
 
Ah, okay, not 100% on what that refers to.
 
The Möbius strip is famous because it's nonorientable
Nonorientable = has one side
Orientable = has two sides
I'm 90% sure the boundary of that^ is a trefoil
 
Looks like it.
 
Hm apparently it's from this link
Like four clicks in
 
There is a general procedure to construct Seifert surfaces but the ones you get out of it dont look as cool :P
 
11:24 PM
When I initially saw that 3-layer Seifert box, my mind jumped to an infinitely layered structure similar to that one. Does anything necessarily prevent such a structure from existing?
 
@Rithaniel I didn't find a three-layered version of the box
As far as I can tell, nothing prevents you from doing it infinitely many times, other than the fact that you'll have to go infinitely far in and infinitely far out
 
Ah, sorry, disconnect between brain and keyboard. Meant the 2-layer box.
 
(I imagine the inner layers get smaller and smaller)
(or maybe they just approach a nonzero limit)
 
I was envisioning smaller and smaller surfaces, yeah.
 
There's this paradox that VSauce talked about a while ago
 
11:32 PM
Supertasks, yeah, I recall that video
Zeno's paradox and all that.
 
where you have a cake with two colors: the first layer is one color, the second layer is the other color, the third layer is the first color again, and they alternate like that
and the layers get half as thin each time so they fit in a finite space
and the question is, if you look at the top, what color do you see
There we go
 
I always like the "superposition" answer. Ie: You see both colors.
 
Now with less black bars
 
This is interesting
 
Though, you could argue that the end result would be an artifact of the nature of the infinite process, in which case you'd need something like the Ramanujan sum to assign values to divergent series.
 
11:42 PM
OK so for future reference, if the URL of a video looks like https://www.youtube.com/watch?v=ffUnNaQTfZE, then the URL of its thumbnail looks like https://i.ytimg.com/vi/ffUnNaQTfZE/0.jpg
 
(I wonder if other numbers instead of 0 can show up)
 
@AkivaWeinberger mathematically if your cube is to be a unit cube then it is clear that there would nothing on the plane z=1
the "cube" would instead occupy the region [0,1]x[0,1]x[0,1)
 
Doesn't help us decide what color you'd see though
 
there would be no top side of the cube, is what I mean
it is as if you would need some access to a "time beyond infinity"
the situation is isomorphic to the question "if the lamp toggles between on and off repeatedly every second, what would be the state of the lamp after an infinite amount of time?"
or, "if I tell you that $f: (\omega+1) \to 2$ satisfies $f(0)=0$ and $f(n+1)=1-f(n)$, what is $f(\omega)$?"
you can treat the bottom side of the cube as $0$, and then go up, and add $1$ every time you change colour
the top side would be labelled $\omega$
@AkivaWeinberger do you like this resolution? :P
aha I have another resolution
 
11:59 PM
But what would I see
 
02:00 - 23:0023:00 - 00:00

« first day (3088 days earlier)      last day (1932 days later) »