Wolfram Mathematica

6:29 AM

Supose I do this:
Flatten[Take[Solve[a x^2 + b x + c == 0, x], 1]]

It will take the first element of the list, but it's still going to be one list with one element. How to take it out of the list?

I've done this in the past by transforming it to a string and replacing the string with StringReplace, from "{" to "".

Is there a function for that?

David Zaslavsky

@GustavoBandeira Part, or you can use its special syntax [[ ... ]]

@GustavoBandeira Better do x /. Solve[a x^2 + b x + c == 0, x]

and work with that list

Thanks for the answers, guys.

$10

David Zaslavsky

Part also makes Flatten and Take unnecessary: Solve[eq, x][[1,1]]

6:36 AM

@belisarius PayPal Account?

@GustavoBandeira A check will do

@belisarius What about...

@belisarius A free strip-tease on stickcam? =D

@GustavoBandeira If you feel like so, go ahead. I promise not to distribute your naked images

I do with a mask.

This mask. =D

If it's told that the guy in the video is from Mathematica SE, our chat will have an increase in the number of users.

7:25 AM

@GustavoBandeira many of my friends with a math PhD are now working for banking or trading companies… there seems to be good money and a number of jobs in that area

(which might be a sad commentary on our world)

@F'x Cool. I trade stocks in the past, I have a little background on it.

@GustavoBandeira I have no idea what that means :)

@F'x Why?

@F'x Bad formulation.

@GustavoBandeira it sometimes seems that there are more jobs purely revolving around money, than jobs that involve helping others or creating things

@F'x Yes.

7:30 AM

@GustavoBandeira and that is somewhat sad

Don't you think this work revolving around money could lead to something good?

@GustavoBandeira yeah, that's an argument… “by performing millisecond trading, I help others gain money and I make people happy”

@F'x But thinking so, you're excluding the rest of the chain.

@F'x For example: The money extracted from these milisecond tradings could be invested in another corporation that will hire more people, etc.

This is only an example - but it also exclude other parts of the chain, which is huge.

@GustavoBandeira but in the end, no money can be gained overall by financial means

I mean, when you do trading, you don't create money…

anyway, that's me ranting, sorry about that

Yep, when you're trading, you're trying to mine money from the market.

7:37 AM

I have no problem with individual people making carrer choices that bring more butter on their bread, and I have made my fair share of those

I just think the system is too heavily biased toward money

4

But that's what I'm thinking about, money itself is meaningless. - He's trying to mine more money from the market, but this money may go somewhere in the future.

anyway, I've gotta go, but your statement about math degree caught my eye and I wanted to tell you the situation as I see it around me

so, in Europe at least, there's no shortage of employment for people with math skills

hope that can cheer you up

@F'x Yep, it was nice to chat with you.

see you!

Cya.

9:38 AM

@BrettChampion I'm on a break from Mathematica-generated avatars for the time being. Thank you for the interest in my mini-gallery!

xzczd

10:37 AM

Ahahaha, I answered a question for the first time in mma.SE just now!

Grace Note

2:18 PM

Just in case it went a little under-known, your moderator elections have started already.

@GraceNote It's known, considering the spate of "Caucus" badges I've seen...

Grace Note

Ah, that's one way to look at it

Mother ... should I run for president ...

@belisarius You had a nice platform, if memory serves...

@J.M. My best platform are a pair of Nikes

2:22 PM

@belisarius That works nicely for the rousing speeches. Make yourself look imposing...

@belisarius are you worried they'll try to break your balls?

@R.M Of course. Paranoid gliding is my way of life

They rotate too, if I remember correctly...

@R.M Wanna see?

Again?

2:27 PM

@R.M Sorry, I am not really aware of the toad's psychology. Am I offending you?

You'll have to try harder than rotating balls for that

btw, I thought you were quoting Pink Floyd, in case my reply came as out of the blue

@R.M BTW, my old rectangles reordering answer is not useful for the map question

@R.M I WAS quoting PF

Why not?

@R.M Because the way it works. the merit function is based on the angular displacement only.

squares could end up in China

Oh I see... I didn't look through your solution in detail

2:37 PM

@R.M Anyway that code was done for a non-Mma programmer. I could get lynched if I post it here :)

:)

@R.M But I have a nice idea :)

`Complement @@@ (Reverse /@ Partition[ Reverse@NestWhileList[Delete[#, List /@ ConvexHull[#]] &,
centers, # != {} &], 2, 1])`

Proceeding from the center to the borders, adding a convex hull each time

I'll be back in abt 30 mins... Leaving for school now

2:58 PM

This might be the first time somebody's been propositioned...

4

@J.M. I am still trying to understand what he meant ...

OMG

rcollyer

@belisarius cleaned up my comments, too. I agree with the sentiment, the OP needs to show more effort, but he did add some formatting, so I revised it to my standards.

@rcollyer Reading his stuff exacerbates the throbbing in my head...

@rcollyer I understand it is a subjective matter, but I feel inclined to do that only with new users

@J.M. agree

I am not going to try to understand his q

3:42 PM

@belisarius that's a neat idea... it should work. Reminds me of your stochastic cell growth answer

or was it realistic cell growth? Either way, you probably know which one I'm referring to

@R.M yep. similar indeed

I am trying to improve Daniel's answer now

his map has too much slack space

but my machine is too slow and his minimizing takes too much time. boring

acl

@R.M they're "orbiting balls" actually

@acl My platform includes a new NASA program for putting in orbit some weird things: balls, lawyers and paper reviewers

@belisarius ...what kind of lawyers?

4:08 PM

@J.M. Those on the B Ark

@J.M. only those that don't like the number 5

4:47 PM

Can someone explain why the OP goes from {_, 1, 4} to {_, 2, 1} instead of {_, 1, 5} in this question

ohh, I get it now.

@R.M {1,5} is approx 50

yes, I see that now

8:38 PM

anybody around?

ListPointPlot3D[Subsets[Range@10, {3}], PlotStyle -> {Opacity[1], Red}]

I SAID RED

however {Red, Opacity[1]}, works

Directive

ListPointPlot3D[Subsets[Range@10, {3}], PlotStyle -> Directive[{Opacity[1], Red}]]

@F'x damn, forgot it. Thanks

this is one of those things I hate in MMA… if you style a lot of plots, you end up spending your time writing Directive… PlotStyle really should accept lists

@J.M. What's with the new avatar? It's my friend hobbes! Let's play Calvin ball.

@SjoerdC.deVries I think he wants to eat you

8:48 PM

@F'x Can't be

Hobbes? The Leviathan's owner?

The philosopher

The social contract inventor, yes

You're on my heels

Luckily Jens hasn't been very active lately.

Interesting guy. Too monarchic for my taste

However, looking at current democracies ...

8:58 PM

I got 105 rep for FaceForm[]. Amazing.

whoa

I meant at all the (removed)

I thought so much

@R.M No I saw you. You answered "whoa" to Sjoerd

shame on you

I note we still have no nominees

Are you considering running?

8:59 PM

I am asking my mother

I'm afraid I have to ask Subbu

Subbu for mod!

Mother ... should I run for president

Wait. That should be "subbu For........ Mod!"

@belisarius The Wall

9:01 PM

don't you cry

All your nightmares are gonna come true

Classic

optimistic, yes

mama, should I build ... the Wall?

There is one video at youtube with images of "the" bomb and that song. Amazing

9:03 PM

My ringtone is the beginning of "Time", though

I prefer "Money" :)

I've transcended that

worst for you

I find the lines "Ten years have got behind you. No one told you when to run. You missed the starting gun" so true

They offered me transcendence for a few millions. But I decided to keep the important things

@SjoerdC.deVries Hehe ... they are.

9:06 PM

hora ruit

Sounds like a tango lyrics

Tempora si fuerint nublia solus erit

etc

Errare humanum est

en.wikiquote.org/wiki/Latin_proverbs

humanum est errare

better

9:11 PM

@belisarius I miss "homo homini lupus"

@SjoerdC.deVries You shouldn't. Rabies is dagerous

How close does Latin feel for you as Spanish speaker?

@SjoerdC.deVries The declinations are the main problem

no decl. in Spanish

Ah

Just got a nice answer badge for FaceForm. What's happening here?

I can understand simple sentences, having read how the thing works in my youth. But when there are many individual terms, I get lost

@SjoerdC.deVries where?

Q: How to plot empty rectangle?

9:18 PM

3

I know there is a Rectangle[] function, but it is always filled with some color. What I need is a rectangle, which is empty inside (only it's boundaries visible). I just need to overlay it on plotted function to indicate some of it's area.

plotting

@SjoerdC.deVries You mean the votes?

@belisarius uh, yes

I discovered voting is random and irrational after my answer about randomness on SO

2

@belisarius Is the voting random() or random()*random()?

I am still getting upvotes for that one

9:22 PM

I now also got the enlightened badge 8-O

maximum effect,zero effort

@SjoerdC.deVries The multicolored image helps :)

I know. I like graphics.

Perhaps that's why I love Mathematica.

@SjoerdC.deVries Yeah. remember the Wally Q

impossible to answer. I flagged it NARQ

still the most voted q

Talking of that, any sign of heike lately?

She checks in every day... she even answered a question couple of days ago

She probably deleted now since belisarius answered first

9:27 PM

I contacted Szabolcs to see if he is still alive...

Did you get a reply? I sent him an email a few weeks ago and didn't hear from him

Announced that he will be back in a couple of days.

good then :)

Many of the regulars are taking a bit of leave

9:34 PM

I can't wait for the nominations to start. The comments will be most fun

@belisarius They are already open for a day. Go ahead, nominate yourself.

@SjoerdC.deVries No, no. I will write the comments.

:)

Now it's the time for all good men to come to the aid of the site

women are ok too

My vote to anyone who bans For :)

@SjoerdC.deVries Now we only have to find one good man around

9:39 PM

except in compiled code, since I just used it... :D

We can have an all-women mod team. That would be a change.

@belisarius -1 for adding print just before I did

@SjoerdC.deVries hehe I saw you ---

looking for master-delete button

9:50 PM

it works

@SjoerdC.deVries another alternative is to flag… as a mod, I think it deletes the message

hum, FAIL… can't flag my own message :(

yep, it's gone

I have a delete link. Anyone wants it too? I know a way...

@SjoerdC.deVries I have one too, no need to play with ownership :)

mathematica.stackexchange.com/election?tab=nomination

@SjoerdC.deVries :)

9:55 PM

F'x for president... sounds good, doesn't it?

@SjoerdC.deVries don't do that, I'm too easily tempted… as you can see from me being already mod on two sites…

plus, the site already has great moderation

@F'x it does. But the meaning of a List is different to that of a Directive. I wouldn't advocate combining them: that would get confusing and arbitrary, I think.

thanks, but I'm already mod on... o no wait

Have to find a different excuse

Here is a rather philosophical point I was wondering about today: what is the value of a named pattern that has not matched anything? But more than that, what should it be?

If a tree fell with nobody present, did it really fall?

10:07 PM

Consider, for example, a /. {_, x_} | _ :> HoldComplete[x]. This gives HoldComplete[]. So the value of x is simply nothing, not even Sequence[] (HoldAllComplete also confers SequenceHold). But why should it give this rather than anything else?

I cannot recall that this sort of thing is documented anywhere. So, the question is, is this just an implementation artifact, or is this properly specified behaviour of the pattern matcher?

@SjoerdC.deVries Wow, your deep thought deserves a warm one handed applause

@OleksandrR. Looks like x gets the value Sequence[], but somehow it escapes HoldComplete

If a tree falls in a forest

@belisarius Well, it's an oldie

"If a tree falls in a forest and no one is around to hear it, does it make a sound?" is a philosophical thought experiment that raises questions regarding observation and knowledge of reality. History Philosopher George Berkeley, in his work, A Treatise Concerning the Principles of Human Knowledge (1710), proposes, "But, say you, surely there is nothing easier than for me to imagine trees, for instance, in a park [...] and nobody by to perceive them. [...] The objects of sense exist only when they are perceived; the trees therefore are in the garden [...] no longer than while there i...

I'm off to bed.G'night

@R.M I think that is merely how it gets represented if you let it to be returned by itself (the front end has to display something, and the answer is clearly not Null). If it really has a value, it should not be able to alter itself inside HoldComplete.

@SjoerdC.deVries Mine too

Kōan

A ; ; ; ) is a story, dialogue, question, or statement, which is used in Zen-practice to provoke the "great doubt", and test a student's progress in Zen practice. Etymology Kōan is a Japanese rendering of the Chinese term (公案), transliterated kung-an (Wade-Giles) or gōng'àn (Pinyin). Chung Feng Ming Pen (中峰明本 1263–1323) wrote that kung-an is an abbreviation for kung-fu an-tu (公府之案牘, Pinyin gōngfǔ zhī àndú, pronounced in Japanese as kōfu no antoku), which referred to a "public record" or the "case records of a public law court" in Tang-dynasty China. Kōan/kung-an thus serves as a m...

perhaps older

:)

nights!

10:18 PM

@OleksandrR. Hmm.. you're right. I don't know the answer to that... Hey, this could be your first question on the site! :)

@R.M if no-one has any other ideas in chat over the next couple of days, I think you may be right. I don't want to post a question that has a trivial answer that I'm just not seeing.

@OleksandrR. {a} /. {_, x___} :> HoldComplete[x]

10:39 PM

@belisarius I think that is a simpler example of the same thing (the situation I had in mind needed Alternatives, but it isn't necessary in general).

@OleksandrR. Compatible with

{a, Sequence @@ {}} /. {_, x___} :> HoldComplete[x]

@belisarius List is not SequenceHold. So they are identically the same.

Now, you can compare

{a, Sequence @@ {}} /. {_, x___} :> x

with

{a,} /. {_, x___} :> x

Yes. One has Null, the other has simply nothing. {a, Sequence @@ {}} is the same as {a, Sequence[]} is the same as {a}.

But {a,} is the same as {a, Null}.

That is why both are different when you use HoldComplete

{a, Sequence @@ {}} /. {, x__} :> HoldComplete[x]

10:46 PM

@belisarius I don't follow?

We have a nomination!

mathematica.stackexchange.com/election/1#post-10470

@R.M one answer???

@OleksandrR. What I am trying to conclude is that HoldComplete[] is the only possible answer compatible with the rest of the results

@belisarius but it is just the same situation presented in various different ways. The puzzle for me is why HoldComplete[] rather than HoldComplete[Sequence[]] rather than HoldComplete[some random garbage off the stack]. Since the pattern didn't match anything, how can it legitimately have any particular value, even if that value is... well, nothing?

@belisarius no participation either... only 7 votes cast (not even 1/question), no edits and no meta/flags.

@OleksandrR. That is what I am thinking. In {a, Sequence @@ {}} /. {_, x___} :> HoldComplete[x] The pattern acts like matching Sequence @@ {}

10:57 PM

@belisarius Not sure I follow... what's the difference between {a, Sequence @@ {}} and {a}? The evaluation has already happened and Sequence[] is out of the game, no?

@belisarius ReplaceAll is not HoldFirst. So your Sequence @@ {} has already disappeared by the time you get to matching.

@OleksandrR. I am not talking about how it works, I am only talking about compatibility of results

@belisarius but the results must be compatible, since the situation is absolutely identical in both cases! Perhaps I'm missing your point?

@OleksandrR. Not your fault, My English is not up to the task

@belisarius I think your English is fine, but perhaps we are coming at this from rather different perspectives. The documentation on BlankNullSequence is not very enlightening on whether it matches a Sequence or not.

{a, b, c} /. {_, x___} :> HoldComplete[x] gives HoldComplete[b, c], so it seems that even if it did match Sequence[b, c], the matcher stripped out the Sequence wrapper before returning.

11:11 PM

@OleksandrR. My last try :). Treatment of border cases in languages are usually defined for compatibility. Here you have an example of a failed one in Mma

{a, b, c}[[1]]

{a, b, c}[[0]]

{a, b, c}[[1;;1]]

{a, b, c}[[0;;0]]

the last one should return {List} :)

@belisarius agreed. Someone forgot a corner case when implementing Span.

{a, b, c}[[;; 0]] returns {}

@RolfMertig because Span[1, 0] is an empty span, so the result is an empty list.

That corner case is, I think, handled correctly.

@OleksandrR. Yes, it just looks funny after [[0;;0]] giving an error ...

@RolfMertig you can get an even funnier error if you specify a step of 0!

{a, b, c}[[;; ;; 0]]

In fact, any nonpositive step gives an error. I think it would be better for negative steps, for example, to return a reversed result.

@belisarius I think I see your point now. It's still not clear to me whether this is definitional or just a coincidence, though.

11:24 PM

@OleksandrR. Ok. At least I was able to sound less incoherent. That is a point :)

@belisarius I don't think it should return List. Using Span indirectly uses Take, which specifically extracts elements of a list, not the head.

@R.M "take" is an operation defined on lists, not by the Take function specifically. (See, e.g., ref/message/Part/take.) So I think it is an omission in the definition of what it means to take a span.

@R.M I am talking from the POV of language consistency.

if a[[n]] is defined for all valid n

a[[n;;n]] should also be

I am not saying that this is a design flaw. Just inconsistency.

@OleksandrR. I'm saying that it doesn't mean anything to take the zeroth element of a list, which is why it gives an error. Rolf's example doesn't lead to an error because it quietly ignores taking something from a larger index to a smaller index.

;;0 is 1 to 0 and 0;; is 0 to -1 which are both ignored.

@R.M the zeroth element of anything is its head. Doesn't (or shouldn't) matter what the head is specifically.

11:32 PM

@OleksandrR. I know, but that's only extractable via Part, not take.

You can't "take" the head from an expression (at least, not to my knowledge).

@R.M okay. But why should using Span in Part imply Take? It isn't documented that you can use Span in Take, after all: only examples with Part are given. IMO, consistency with the function it is used in would be desirable.

Admittedly, it is stated that "m[[i ;; j ;; k]] is equivalent to Take[m, {i, j, k}]." I just think this is not a very helpful equivalence.

@R.M. Extract[ff[a], 0] gives ff

Especially given the next example: "m[[i ;; j]] = v can be used to reset a span of elements in a list or other expression." Can't do that with Take!

@OleksandrR. I agree there are differences, but it seems like Span (which is relatively new) is implemented over take with some layers of additional functionality like the one you mentioned

Probably at a low level, in that it is not calling Take explicitly...

@RolfMertig Yes, that works too... and I think Extract is also just a layer over Part with added functionality that lets you hold or operate on subexpressions before evaluation

If in a paper I read " ... using fast Fourier transform zero-padded to avoid aliasing." how do I do this with Fourier? (I mean: in ListConvolve I can have a third argument 0, but then I don't quite see how to "fix" that convolution theorem relation in M (like, from the Help: {a, b} = RandomReal[1, {2, 31}];
ListConvolve[a, b, {1, 1}] ==
InverseFourier[Fourier[a] Fourier[b]] Sqrt[31] ) )

11:47 PM

Yes, there are some differences, but there probably shouldn't be. :) Range[5] ~Part~ Range[0, 5] works fine, for example. I think it's an example of a loosely defined behaviour, which in some ways is worse than being undocumented: at least we know that undocumented functions are liable to miss important edge cases. But a purely operational definition of what is proper behaviour is confusing when you're trying not to rely on undocumented functionality.

@RolfMertig Use PadRight?

@RolfMertig I use a padded fft in this answer (I've been meaning to get back to it to add some stuff)... what I did there is essentially: Fourier[PadRight[foo, nfft], FourierParameters -> {1, -1}]