Wolfram Mathematica

J. M.

00:00

Yes, they're meant for exact symbolic manipulations. But for pictures, you can use N[] on them before feeding to Graphics[] (I find things are slightly faster if the coordinates are not exact numbers, at least on my system).

Pickett

@J.M. If it's just another introduction to Wolfram Language then it's not so interesting, but WRI already published one of those this year; will they really publish another already? Rather I am hoping that he will explain his vision for the language, who will use it in the future and for what, etc.

PlaysDice

@MichaelHale @J.M appreciated. I baulked at even trying to plot them but now you've kindly parsed them, I can see clearer how they work.

J. M.

@Pickett Hopefully, he took the writing at least as seriously as NKS. :)

belisarius is forth

@Pickett "If it's just another introduction to Wolfram Language then it's not so interesting" ... but what if it's a New Kind of Introduction :p

J. M.

@PlaysDice One other thing: Root[1 - 20 #1^2 + 80 #1^4 &, 3] // ToRadicals. Sometimes, one's lucky. Most of the time, due to Abel and Galois, ... no.

@belisariusisforth We're not going to stop that joke until WRI ceases to exist... :)

PlaysDice

00:07

@J.M. now you're going all "group theory" on my dodecahedron ;)

Pickett

@J.M. @belisariusisforth I'm sticking to the idea that he has something important to say about the language, we'll see :)

belisarius is forth

@J.M. Well, he did his better to keep the joke coming

J. M.

@PlaysDice I'm just following the tradition of Felix Klein. ;)

PlaysDice

@J.M Well don't bottle it up on my account

J. M.

@Pickett "It's so important, I named it after me!"

belisarius is forth

00:12

@J.M. See? you can't keep that jokes for yourself either. It's too tempting.

@J.M. I can't imagine being at a company meeting at WRI and bursting into red implosions as I try to restrain my laugh in front of Him

halirutan

01:10

@Pickett If you are still around, maybe you like to give me a readers digest on google analytics and what are the important steps to set it up. I thought I have it for the plugin page, but that was really the google webmaster tools. Very confusing for someone not working with web-tech every day.

At least I could verify which pages link to my pages with it.

Pickett

@halirutan I'm still here. Basically all you have to do is find the right form (start here) and copy a piece of JavaScript code into your website HTML code.

halirutan

@Pickett OK, so far it was the same for the webmaster tools.

I had to include some ID identifying my webpage

Pickett

@halirutan Well, for the webmaster tool you just copy some ID to your webpage so that Google can verify that you own the website. Google Analytics is different because it loads JavaScript into your page that does advanced tracking and reports back to Google's servers in real time.

halirutan

@Pickett Ah, I see. Is there a possibility this already exists as plugin for a Joomla based page?

Pickett

@halirutan Yes, there should definitely be plugin that will ask you for your Google Analytics ID and then injects the required JavaScript code automatically. I've written such code for WordPress themes several times.

halirutan

01:18

@Pickett What happens, if I have several URL's that lead to the same page?

E.g. we have mathematicaplugin.halirutan.de and wlplugin.halirutan.de

Is this a problem? It's just sub-domain pointing to the very same page

Pickett

@halirutan They will be tracked separately, technically they are not the same page (unless you mean that both redirect somewhere, then wherever they land is what matters.)

halirutan

@Pickett OK, let me play around. Thanks

Pickett

np

halirutan

@Pickett Is there an easy way to see whether it works?

Pickett

@halirutan If the real time data works, then it works.

(there's a real time menu item in the Google Analytics interface)

halirutan

01:31

@Pickett Found it. Let me see

@Pickett Awesome :-)

Pickett

:)

Szabolcs

08:42

Any comments here?

0

Q: SyntaxInformation: take allowed options from multiple symbols – possible?

Suppose I have a function like Options[f] = {foo -> 1}; f[opt : OptionsPattern[{f,g}]] := h@g[Sequence @@ FilterRules[{opt}, Options[g]]] I purposefully do not want to append Options[g] to Options[f] because I want the defaults to be inherited from g instead of setting them separately for f. ...

Axiverse

09:17

I just started using mathematica and I have a really basic question:

I'm trying to drop certain elements from subarrays like: Map[Drop[#, 1], {{1, 2}, {3, 4}}]

and I'd like {{2},{4}}, but it's not doing that.

Szabolcs

@Axiverse You missed the & at the end of the pure function.

An easier way is arr[[All, 2]] where arr is your matrix.

Axiverse

Ah, thanks for the &. I can't seem to use the arr[[All,2]] method, because I actually want two different ranges - something like arr[[All;{2,4;;}]]

Basically the second and fourth to end element of subarrays. Thus I'm trying to map Drop[#,1,3]& to the array.

J. M.

@Axiverse, if the nesting is needed: Drop[{{1, 2}, {3, 4}}, None, 1]

Replace the third argument with the sequence specification for the columns you want to discard.

Axiverse

09:33

How would I describe non-continuous sequences - like drop {2,2} and {4,4} without nesting like this? Drop[Drop[wb, {1, 5}, 1], None, {2, 2}]

Kuba

I'd expect Echo to respect PrintToConsole setting since documentation says "prints expr...".

One work around is to screen $Notebooks which is checked

Block[{$Notebooks = False}, Echo[1]]

but if I have to create custom function anyway I feel like Echo is useless

@Axiverse Delete[Array[a, {5, 5}], {{1, 1}, {5, 5}}] // MatrixForm is this what you want?

J. M.

09:56

I'm still hazy on the difference between Echo[] and Print[]…

Axiverse

@Kuba I'm not sure what the Array[] is for in your suggestion above. But Delete is helpful

Map[Delete[#, {{1}, {3}}] &, wb[[6 ;;]]]

is what i ended up with

Kuba

@Axiverse it's an array to test since you haven't attached wb definition.

Axiverse

ah okay

Kuba

@J.M.

<< GeneralUtilities`
PrintDefinitions@Echo

J. M.

@Kuba, I don't have 10.3 yet. Can you spoil me? :)

Kuba

10:03

@J.M. basically:

Echo[expr_]:=(echoPrint[expr];expr);
echoPrint[expr_]/;$Notebooks:=CellPrint[echoCell[expr,$echoFormatType]];
echoPrint[expr_]:=Print[$echoPrefix,expr];

J. M.

So, it prints and returns, while Print[] just… prints.

Kuba

@J.M. it seems there was some time left after they fixed all pending bugs, so why not?

Jacob Akkerboom

10:29

@halirutan a sanity check would be appreciated in my following answer. Is it true that even when the data is already distributed, parallellization does speed things up?

5

A: ParallelMemberQ

Short answer It is not a good idea to make a parallel version of MemberQ. This is mainly because membership testing is generally faster than copying data and copying the data is necessary in Mathematica to allow other kernels to work with the data properly. About the rest of this answer This ...

Sosi

@JimBaldwin Since you told me about the Welch t-test being performed by `TTest[]` with the appropriate options, maybe you'd like to contribute with an answer to this?
http://stats.stackexchange.com/questions/176785/comparison-of-medians-in-samples-with-unequal-variance-size-and-shape/

Jacob Akkerboom

10:47

@halirutan Nevermind, I see that the data is not already distributed (of course). Still your feedback would be very welcome as I make some claims about parallelization which I am not completely sure of.

Jacobadtr

11:33

Hello

Can anyone tell me why this throws up an error?

NDSolve[{x'[t] == (x[t] - 2) ** 2, x[0] == 0}, x[t], t]

"Equation or list of equations expected instead of True in the first \
argument \!${\*SuperscriptBox[\"x\", \"\[Prime]\",
MultilineFunction->None][t] == \((\(-2$ + x[t])\) ** 2, True}\). "

belisarius is forth

@Jacobadtr What is **?? :D

Jacobadtr

@belisariusisforth That is me with my python hat on... I guess ^ is what I want?

Jim Baldwin

@Sosi. I have added a comment to that question. A more specific answer would require knowing what software you can use, the nature of how you obtained the data (subsamples within random locations, any blocking or randomization restrictions, etc.), and what data values were actually obtained. So the question is just too big for me to give an answer. I can only poke at it with comments.

belisarius is forth

@Jacobadtr Yep

Jacobadtr

It still isn't happy.

"Equation or list of equations expected instead of (-2+x[t])^2 in the \
first argument {(-2+x[t])^2,x[0]==0}."

belisarius is forth

11:45

@Jacobadtr ClearAll["Global`*"]; NDSolve[{x'[t] == (x[t] - 2)^2, x[0] == 0}, x[t], t]

Jacobadtr

@belisariusisforth This outputs "NDSolve[{True, x[0] == 0}, x[t], t]"

Solving DE's by hand is easier at this point D:

belisarius is forth

12:04

@Jacobadtr Restart Mathematica. You've quite a mess there :)

Jacobadtr

12:19

@belisariusisforth Indeed...

Michael E2

13:36

@Jacobadtr Most likely at some point you executed something like x'[t] = (x[t] - 2)^2, with = instead of ==. This sets a value for Derivative[1][x][t] that is equal to the RHS. When you changed to ==, the DE evaluated to True. You need to restart the kernel, ClearAll[Derivative], or unset the value with x'[t] =..

4

Q: Error entering equation in DSolve

I entered a command incorrectly as follows: DSolve[{y'[x]=y[x]},y[x],x] I am now experiencing: DSolve[{y'[x] == y[x]}, y[x], x] During evaluation of In[26]:= DSolve::deqn: Equation or list of equations expected instead of True in the first argument {True}. >> (* DSolve[{True},y...

differential-equations clear

kirma

14:36

Oh God.

EdgeDelete[Graph@{UndirectedEdge[1, 2]}, {}] // InputForm puts kernel into uninterruptible busy loop on v10.3...

Well, with any graph.

And InputForm is just one function that makes this happen.

Bug reported.

Kuba

@kirma it seems that releases with odd decimal digit are just hidden beta.

kirma

Hmm... :o

kirma

15:53

Oddball method of approximating maximized minimum of pairwise distances between N points (arbitrary distance measure): Construct a fully connected graph of large amount of samples, find random fully connected subgraph of N vertices, remove all edges smaller or equal to minimum of distances on subgraph. Repeat until no such subgraph can be found; last one found is the optimal solution on those samples.

I'm pretty confident this could be implemented quite efficiently using bit-vector representations and such. Proof of concept work on Mma is easier, though.

hftf

17:40

Related question here: How do I scale SmoothHistogram to Histogram using "Count" instead of the other way around (using "PDF")? I can't find an analog for the third argument hspec "Count" in SmoothHistogram.

J. M.

18:52

@hftf "Intensity"?

hftf

19:19

@J.M. I don't believe that's right: imgur.com/xJ3yM6c

xslittlegrass

19:36

The book is online wolfram.com/language/elementary-introduction

2

Michael Hale

@xslittlegrass Thanks. Sent to my dad. He's had a copy of Mathematica for almost a year now, and he's been telling me he's going to learn it for two years. He says he definitely wants to learn it before he retires, which is next summer.

Last week I started getting more interested in the NC =? P problem. Does showing that the results of Simplify on increasing depths of rule 110 logic circuits count as evidence that NC < P?

f = (#1 && ! #2 && ! #3) || (! #1 && #2 && #3) || (! #1 && #2 && ! \
#3) || (! #1 && ! #2 && #3) &

Table[Function[n,
   LeafCount@
    Simplify@
     First@NestWhile[f @@@ Partition[#, 3, 1] &,
       Table[v[i], {i, 2 n + 1}], Length@# > 1 &]][n], {n, 6}]

kirma

20:08

"Today the Wolfram Language has emerged as something else: a new kind of general computer language, which redefines what’s practical to do with computers."

argh...

Pickett

20:32

"there’ve been two millennia of development in the teaching of mathematics, that have progressively optimized the sequence of presenting arithmetic, algebra and so on. The problem of teaching the Wolfram Language is something completely new, where everything has to be figured out from scratch. Existing programming education isn’t much help, because so much of it is about just the kinds of lower-lower structure that have been automated away in the Wolfram Language."

He needed to invent a new kind of programming education in order to write this book.

4

Wolfram Blog

20:53

Wolfram Blog: Wolfram Community Is Turning 10,000

posted on October 20, 2015 by Vitaliy Kaurov

Community is about to turn 10,000! Members, not years. We launched Wolfram Community in July of 2013. After two short years, it has grown to just a hair’s breadth away from 10,000 participants. Join now and help us reach this milestone! We’ll award prizes to new members with the most creative profiles who join until [...]

J. M.

23:46

@Pickett The "visionary" doth protest too much...

Transcript for