« first day (200 days earlier)      last day (1156 days later) » 
00:00 - 22:0022:00 - 00:00

10:00 PM
oh this guy was your colleague right?
@acl Eons ago, yes
@acl Did you see my orbiting balls?
A: Animation of Differential Equations from NDSolve with ParametricPlot3D and Evaluate

belisariusFor Example s = Table[k[j] -> Interpolation[Table[{i, Sin[i j]}, {i, 6}]], {j, 1, 6}]; Manipulate[ ParametricPlot3D[{{k[4][t], k[5][t], k[6][t]}, {k[1][t], k[2][t], k[3][t]}} /. s, {t, 1, a}, PlotStyle -> {Red, Green}, Evaluated -> True, PlotRange -> {{-1.5, 1.5}, {-1.5, 1...

@belisarius ha yes
even voted for them (although you should insert the phrase "my orbiting balls" somewhere in the answer)
@acl Sometimes I think how to that without Mma!
@acl Done!
@belisarius hehe
10:06 PM
@acl Now, just wait for the flags comming
> flags comming
@belisarius so, he seems to have worked on nonequilibrium dynamics extensively
The benefit of not being a native speaker is that I always may say "sorry, was that the meaning in English? Ohhhh!"
You should slow your balls down, belisarius... kinda like those chinese temper balls that you gently roll in your hand
10:08 PM
@R.M nasty
@R.M I guess it depends on your processor speed and your imagination
no, I was serious... lol.
@belisarius I have a very fast processor!
they are moving slowly here
@belisarius must be the winter
@R.M Let me check if "DisplayDurations" works
@R.M check it now
10:12 PM
so, I've been stuck in an airport for about 12 hrs now... not fun.
2 more to go before my flight
@R.M I hate airports. All of them
@R.M at least you have somewhere to charge your laptop. imagine if you had to sit and strike up conversations with people!
subtle, no?
belisarius is very tactful and discreet
@R.M Sorry, I had to star that
@R.M indeed:
10:18 PM
New WRI motto: Put your balls in orbit with Mathematica
now do it for n>3 and solve the problem too while you're at it
@R.M I wrote it at the margin of my book. Let me search
10:33 PM
This is nice one to try in Mma
Q: Finding matches between high quality and low quality, pixelated images - is it possible ? How?

Simon KielyI have a problem. My company has given he an awfully boring task. We have two databases of dialog boxes. One of these databases contains images of horrific quality, the other very high quality. Unfortunately, the dialogs of horrific quality contain important mappings to other info. So I have be...

@belisarius yes!
@VitaliyKaurov Hi! See my link just above!
@acl mmmm could we copy it? is it "legal"?
@belisarius reading the post
@belisarius we could pose an analogous question and link to that
@belisarius cc license, so yes, as long as you attribute
but make it more interesting
10:36 PM
@R.M but it's not nice to steal someone's question :)
@R.M Yes, but I mean ... the OP here will grab some rep points
We could set it CW
@belisarius unless it's me, I seem to repel votes :)
@acl "My company has several blurry images of Lena. I think this is a sin, so I need to replace it with high res image of Lena"
@R.M The full one, of course
10:37 PM
no other
Question on Signal Processing: "How can I use inpaint to fill in the missing parts of Lena? MATLAB code pls"
@R.M Buy Playboy
On a more serious note, anyone here familiar with elliptic integrals? I have an integral where mma pops out an EllipticK as the result, but I'm trying to figure out where to start so that I can get to it with pencil and paper (the answer almost always involves Abramovitz and Stegun)
@R.M gradshteyn and ryzhik is probably more useful
@R.M A&S is the way to go if the word "Elliptic" is there
@acl That means WAR
so basically you have an integral and want to find which transformation brings it into a form that can be looked up?
10:42 PM
@belisarius I wouldn't be surprised if my integral was already there as formula #xyz
@belisarius BS. how do you look up integrals in A&S if you don't already know the answer?
@acl I can answer in two different ways. 1) I always know the answer, or 2) I browse the book until blind
@belisarius see here, RM doesn't know the answer and he doesn't want to go blind. ergo...
@acl poor soul
@acl I know the final answer. I need to work out the intermediate steps so that I can "derive" it in an appendix. So I need to know a reasonable starting point
10:46 PM
@R.M Use Landau's method
@R.M in seriousness, I'd say that A&S is really not a book to look up integrals. it's for properties of functions. G&S is explicitly a catalog of integrals, arranged so that you can look things up
but in your case, you basically know you need to bring your integral into the form that is the definition of K, so you just have to manipulate it until you get that
@acl I'll give G&S a look. Hadn't heard of it before
@acl exactly
so is your integrand long (no jokes please)?
@acl In its simplest form, it's this: Integrate[Abs[Exp[-x] BesselJ[0, x]]^2, {x, 0, \[Infinity]}]
@acl Like this one ?
@R.M well G&S have sections "integrals of bessel functions" etc so you can expand this and try to look it up
but the Abs will complicate things
@acl yeah, the abs is what makes it harder... Watson also has some results. I'll have to look it up when I get back
or maybe not. I am now itching to get pencil and paper but I should quit for the day (it's a sunday!)
here a lot of properties listed explicitly: functions.wolfram.com/EllipticIntegrals/EllipticK
10:56 PM
@R.M actually you need Integrate[Exp[-2 x] BesselJ[0, x]^2, {x, 0, \[Infinity]}] so probably this can be looked up immediately
@acl probably. I'll give it a try when I'm closer to the monster books.
Speaking of which, they should make an A&S app for the ipad or the mac. I'd definitely pay to be able to search
or, from this form, you can write $I_n=\int dx \exp(-2x)J_n(x)^2$ and obtain a recurrence relation between the $I_n$ (by integrating by parts) and try to look up the solution to that
or other things. there's an infinite number of ways to waste time with these things
@acl most certainly
@R.M how did you end up with this integral?
The bessel comes up as the correlation between two waves in 2D and exp is an attenuation term. I need to show it is in L2, hence the integral
The exp is actually exp(-abs(x)), but I left that part out
11:13 PM
@R.M I see. Since asymptotically $J_0(x)$ decays like $1/\sqrt{x}$ and it is everywhere bounded on the real line it seems obvious, but I guess standards of rigour differ in different areas!
@acl Just J0 by itself isn't in L2... The exp makes it L2
@R.M I know, but you have the exp
Is the fanatic gold badge something to be proud or ashamed of?
Yes, the obvious argument why I've been ignoring it until now when I'm writing it down, but since it isn't obvious to the audience I'm writing to, I'll have to expand on it a little.
11:17 PM
(it decays like $\cos{x}/\sqrt{x}$, so the integral of its square diverges logarithmically I guess)
@belisarius \o/!
@belisarius end of discussion then!
so how did you spot this? do the Orbiting Balls sport a telescope?
page 116
11:19 PM
@Rojo hang your head in shame and join the club
Thanks @belisarius You've ensured that I didn't put my time to good use and can now spend it on something useless :)
@R.M that's the idea :D
@Rojo The fanatic is probably the easiest to earn, but I can imagine it being pretty annoying if you miss out at 99 (even though badges mean nothing)
@R.M I don't know for how much I missed when everyone got it
but I was more proud than annoyed
Not any more
11:24 PM
I need two votes to cap. Someone else want to look my balls?
Its good to have a gold badge even if it is the lamest
Sorry, I can only do that once
@Rojo still mesmerized?
@R.M Right
Hey, now that there's activity in here
about the undo
Anyone wants to give opinion, help out, test, etc?
@Rojo I can give you an opinion. No need to test.
11:27 PM
@Rojo Ok, link me to it... I have a few minutes to test before I have to leave
but give us a link ...
@belisarius You don't need a link to not test
@Rojo Just in case my opinion is too broad. For example Politics is not a science. My opinion
So if you load that and then do on a notebook SetUpUndo[InputNotebook[], {"Text", "Input"}] it should use an invisible notebook to store the stuff
but it works smoother in the kernel, in which case it's ` SetUpUndo[InputNotebook[], {"Text", "Input"}, "Method"->"Kernel"]
The option "StoreFrequency"-> takes an integer with every how many keystrokes it stores the state, defaulting to 15
and you undo with alt+click
if it works
Should I buy some insurance first?
11:31 PM
No re-do's
@Rojo alt-click what? nothing happens
alt click on the cell
Input or text cell
that you have started to write
after running the setup
nothing happens (mac here)
@R.M You used the option "Method"->"Kernel" or without it?
11:33 PM
yes, I used it
nothing here
Sad when this happens
Grr, perhaps I didn't copy the code right? I just did this on a fresh notebook SetUpUndo[InputNotebook[], {"Input", "Text"}, "Method" -> "Kernel"]
and it works
@Rojo By "Loading" you mean just running it, right?
Just evaluating that
ok lets try again
11:35 PM
Remember that it defaults to 15 keystrokes. If you just type "hello" nothing will happen
yes, I typed quite a bit
Hey, it works now
Thank god
@Rojo Maybe it isn't recognizing Alt.. Mac might need option key assigned explicitly
@R.M Oh, I see
Well, it's just a matter of searching the code for AltKey
and changing that to ShiftKey to test
11:38 PM
Ok, I'll do that later. Gotta go now. Bye
@Rojo wonderful
@belisarius :)
Test it without the kernel option
and add some shorter frequency, "StoreFrequency"->5 or something
We have the slowest pcs in here
If it works for us
it's good
lets see
how to set freq?
and then, the options
11:41 PM
please write down the full comand, without using the kernel
SetUpUndo[InputNotebook[], "Text", "StoreFrequency"->5]
Try that in a new notebook
nothing happens
@belisarius With that command, it will only work on Text cells
The second argument is the style or list of styles with undo
it works now
Does it feel sluggish or something?
11:47 PM
there is something strange
when I ADD text, the undo works 5 chars at a time
but when I delete text, it works straight thru the end
OHH, but it just doesn't count when you erase!
Yeah, that isn't right, right?
ahh ok
I have to think about it
I am not programming, just typing randon text
Perhaps it should count every keystroke
SetOptions[UndoDataNotebook[], Visible->True] for a peek at the internals
another strange behavior
I type
f[x_] := Sin[x];
Plot[f[x], {x, 0, 1}]
then I delete some chars
f[x_] := Sin[x];
Plot[f[x], {x
When I use undo>
f[x_] := Sin[x];
Plot[f[x], {x, 0, 1}
one char less than the original
It just stores the state every now and then
and can only recover there
00:00 - 22:0022:00 - 00:00

« first day (200 days earlier)      last day (1156 days later) »