Wolfram Mathematica

@Heike looks like we have the same problem then

Verbeia

12:18 AM

@Heike I've added a bit to your answer to my question - thanks again!

12:29 AM

@Szabolcs that didn't provide as much of a diversion as I'd hoped. :-)

@Szabolcs nice to see that they are not taking an "ideological" position on this. they sound reasonable

12:49 AM

that confetti code is amazing

2:35 AM

@MrWizard yeah, the history shows it (even though you might not have done it explicitly).

and holy confetti code, batman!

2:55 AM

@yoda Now it shows as Deleted. What's that about?

Hey @MrWizard

@MrWizard Yeah, that puzzles me too. Usually such stubs are only deleted on the source site, not the destination. I pinged Rebecca about it in TL

I'm not sure whether you really misunderstood the question. It's pretty amazing, because I noted the difference of your function and mine (and Szabolcs) only in a sequence of 100 numbers.

Your function does mostly the right thing.

From this {20, 56, 57, 62, 20} I expect

{20, 56, 62, 20}

the 57 is left out, because abs(56-57)>5 is not true.

At this point, you keep 56 as your reference number and compare 56 with 62 which makes 62 into the output list

@MrWizard I think thats why you cannot use Differences here, because you don't always compare two neigboring elements

3:11 AM

@halirutan I see. In that case I made a big mess of it. Also, it sounds like a duplicate question because that sounds really familiar.

@MrWizard Maybe you can fix and still use your approach. Anyway, it's late here, so I should go to bed. Good night everyone.

@halirutan one moment if you can:

yep

Q: Finding time-series direction reversal of certain magnitude

does this appear to be a duplicate?

9

I'd like to find the points of a time-series that are a certain distance away (in value, not in time) from the previous maximum, which I consider a reversal. For example, for {1, 2, 3, 5, 10, 8, 6, 3} with a threshold of 4, the reversal point would be 6, since it's 4 units away from the previous...

list-manipulation

I'm really tired too. I cannot think any longer.

No, it's different. In the newer question we compare all differences, not only from the maximum.

3:20 AM

Time for bed then. Brain failure. :-)

See you

10:04 AM

@MrWizard Delete this? It already has two votes, only needs one more. Might as well come from you (the diamond won't make a difference this time)

10:26 AM

also this one and this one (any one who can vote)

Also I think this should be reopened. All of the close votes were for the original version of the question which seemed like it was asking for a tutorial... the edit makes it narrower, and is also what the OP actually wanted, but didn't phrase it well

10:43 AM

@Szabolcs Are you working on creating an antialiased version of the confetti?

Q: Point Renderings Slightly Off in Mathematica

@yoda Not worth it because a GIF only has 256 colours and Mathematica will start using dithering

21

The Mathematica code f := #1 + Sin[2 #1] &; inflectionPoints = Table[{x, f[x]}, {x, -Pi, Pi, Pi/2}]; stationaryPoints = Union[ Table[{x, f[x]}, {x, -2 Pi/3, Pi/3, Pi}], Table[{x, f[x]}, {x, -Pi/3, 2 Pi/3, Pi}]]; Show[Plot[f[x], {x, -Pi - 1, Pi + 1}], ListPlot[{stationaryPoints,...

graphics plotting

I am looking at this:

It was migrated from SO so another question can be marked as a duplicate of it

I always thought this was because the plot markers are characters from a special font

(but haven't finished reading the answers yet)

@Szabolcs has already been marked and merged

Yes

I'm focusing on the content

but yes, it is because the markers are actually font characters and not graphics (polygons)

Hm, actually the answer there says that this behaviour was explained by me on MathGrop, but I don't remember at all :D

10:47 AM

haha, lol... eventually, you'll post enough here that you'll start forgetting what you posted on SE

or you'll go insane by posting too often, whichever comes first... ;)

Mr.Wizard is already half way there to an Epic badge and you're almost half-way

I slowed down a bit and it feels better

Did you know ESC fn ESC?

yeah, I use it occasionally

I wouldn't say it feels natural to me, since I'm not a mathematician, but I do like using it sometimes to candy up a notebook

11:17 AM

@Szabolcs what are you trying to do?

Slow day today, eh?

maybe that's a good think. Gives me a chance to improve my xcode skills.

how are you liking cocoa and objective C so far?

It still feels like black magic to me

with those .nib files and stuff, but I'm slowly getting the hang of it.

Are you trying to develop for OSX or iOS?

or is that a silly question?

11:24 AM

OSX

nice, git support is built in to xcode

(I just fired up xcode and am fiddling with it)

I'm still trying to find my way around xcode. The fact that the book I'm learning from is written for xcode 3 doesn't help.

111

Q: Hidden Features of Xcode 4

Now that Xcode 4 is officially released it's time for a follow up to my previous question: Hidden Features of Xcode So what are they? What are those hidden little Xcode 4 hints & tips that you can't live without? Please limit your tips to those that are new and specific to Xcode 4!

cocoa ios osx xcode4 hidden-features

those shortcuts help

11:44 AM

i like how easy it is to play with all sorts of profiling and debugging tools in xcode (last time I wrote C I used gdb without a frontend); shame it has to be C though

11:54 AM

I got another downvote here: mathematica.stackexchange.com/a/613/12

I wonder if that's because a better solution was posted or because something is wrong with mine that I hadn't noticed.

12:21 PM

This is a very interesting article that is relevant to us (mma.se)

2

somehow I suspect that this kind of thing has appeared numerous times as an FAQ list over the last few decades; guess it didn't occur to anybody to publish it until now!

who knows how many such missed opportunities there are in condensed matter physics ;)

@yoda now that you mention it...

I'm pretty sure you could take lines at random from articles uploaded to arxiv in a week and make an entirely new article out of it :P

pdos.csail.mit.edu/scigen

shows how little some conferences review their submitted papers

12:36 PM

@yoda Those 10 rules seem pretty obvious to me.

@Heike I'm still constantly surprised by how unobvious it is to a good number of people...

but hey, free paper! (don't know how much that counts for in their field)

@yoda This has been done by a few French physicists publishing for a social sciences paper.

@yoda At least now we can point such posters to this paper and ask them to come back when they have figured out which rule they broke.

12:49 PM

Funny that it appeared as a paper.

Maybe it's something to link to from the FAQ---some might take a different attitude "because it's a peer reviewed paper"

1:11 PM

@Szabolcs Apparently the wolfram demonstrations are peer reviewed as well.

3:03 PM

The front end froze because I accidentally printed something big and I didn't save :-(

3:15 PM

@Szabolcs need a shoulder to cry on?

5:16 PM

Should I ask this poster to post a new question? This question is about using NonLinearModelFit, but it turns out for what he is doing NonLinearModelFit is just not a very good solution. (See the comments)

Since there are already several answers there, how about leaving that question alone and asking a new one about fitting a sum of sines to that data?

becko

6:24 PM

@Szabolcs I sent you the email with the data.

10:40 PM

I really wish the core language of Mathematica were faster. This is beautifully and extensibly implemented, but it is quite slow.

One could write much faster code, but then it would not be so nicely written.

11:21 PM

@Heike are you here?

@MrWizard Yes I am

Hello.

Hi

For a second time I posted an answer that duplicates a method you showed in an earlier answer. If it does not go against your own sensibilities, I ask you to change the format of your answers.

@MrWizard Is that the multiple bar chart question?

11:24 PM

I cannot read every line of every answer. I need a way to see the major points and methods of an answer. If you would include a divider line between your methods, or use Title (##) tags, or something similar, this would help me.

Yes, that's the one.

What do you think?

I know I tend to write solutions in a chronological order

Maybe I can try putting the conclusion at the top and the derivation at the bottom in future.

That is one option. Divider lines would be just fine too, as would titles. Do you have a problem with either of these latter options? They would require a more minor change in your style, IMO.

I don't mind using titles and divider bars.

Anyway, thanks for considering this to accommodate my reading comprehension problems. :^)

I just didn't think the answer for that particular question needed titles.

By the way, I don't think that the three panel question is really a duplicate of the linked question.

11:33 PM

Okay, I'll reopen it.

@Heike Would you try the divider in that case?