I have a huge list and when I try to use listplot on the list, I get a blank graph without any error message..however excel can easily plot it

any suggestions on how to troubleshoot it

Found the error : Excel screwed up the file because I processed it in excel before importing it to mma

@psimeson Well, your data are not number, but strings.

Thanks to everybody who voted to reopen! I'll write a summary of that paper I was talking about tonight. :)

Would anyone have a good (simple?) example of something that would cause the kernel to hang ...ideally that an Abort Evaluation can end without incident.

@MikeHoneychurch a For loop?

@halirutan I guess an infinite loop would be okay. I am testing something in which i want TimeConstrained to be able to stop it if it runs wild. Wanted to try something controlled first

@MikeHoneychurch OK, use this:

A = {{1, 2, 3}, {4, 1, 0}, {0, 5, 4}};
{p, d} = JordanDecomposition[A];
FullSimplify[p.d^n.Inverse[p] == MatrixPower[A, n], n > 0]

It takes about 10-30 seconds.

If you want it even longer, you could remove the n>0 so that it doesn't find the solution but chokes to death.

@halirutan the choking to death version is appealing :) many thanks

@halirutan You just had that ready... ;)

@J.M. No! My brilliant mind made it up as perfect test-example for @Mike :)

@J.M. I was amazed at the speed that the code materialized

@MikeHoneychurch And THAT was the whole point of the exercise :D

@MikeHoneychurch In that case, stay away from the front page for a few hours or so. ;P

have to go. thanks again @halirutan bye @J.M.

@MikeHoneychurch Bye.

I wonder if Method -> "Semialgebraic" in DiscretizeRegion is some sort of a "preview" feature. I'd love to use it, but it is able to sometimes fail spectacularly even on cases like polygons in 3D space constructed as intersection of HalfSpaces and a plane.

The sad part is that semialgebraics would be the perfect method to perform precise discretization like this.

Should I write a bug report? I don't feel this Method is even intended to be usable at this point, although it's exposed through warning messages.

@kirma "Semialgebraic" is not mentioned in DiscretizeRegion's documentation so it means it still under development. An this is what you will get as an answer from WRI. Write a report if it's released and still broken.

@Kuba Thought roughly so. Also, some fleeting mentions of ongoing development related to semialgebraic sets in WRI videos would give a hint this specific Method is probably one of those features which are currently at most preview quality.

It seems that NotebookOpen doesn't respect SetDirectory, should I expect that?

ahh, yes, because it respects NotebookPath only while Import/Export and friends - DirectoryStack[], moreover NotebookPath "has not been fully integrated into the long-term Wolfram Language, and is subject to change.". #funWithMMA

@Kuba I guess that's because Directory[] is the current directory of the kernel process, while NotebookOpen is executed by the front end.

@Szabolcs I suppose so. But current setup doesn't look clean with all those different paths that different things take care of. Not to mention that FE related are usually with "not full implemented.." tag.

@Szabolcs The problem is, I was surprised by the fact NotebookOpen failed.

Maybe it is a problem with me only, but I often am surprised :) And I don't like that because I should have gained experience already :P

@Kuba What I find really annoying is that autocompletion is not sensitive to Directory[] at all ...

@kirma How did you find that option and what else is available? I am currently struggling with BoundaryDiscretizeRegion.

@Szabolcs I think I have a Method that wasn't available to 3D regions and it listed the alternatives on an error message.

I feel it's a recent addition, I have had reasonable amount of interest on semialgrebraic sets (and limitations of handling them on Mma, although there's some quite nifty functionality!), and I should remember seeing it before if I had actually seen it...

There's one mention on Mma.SE, too.

BTW, it is really not very fast, at least in my case. Yet, it's clearly riddled with some bugs.

I do like the possibility of it maturing, though.

Holes, which seem to occur at perpendicular axes in this case, are clearly bugs.

Argh! No, they are not! I actually have set too tight bounding box for this discretization!

Clearly it helps to "think aloud" these things.

It works too well. :)

@kirma, have you heard of Goldberg polyhedra?

@J.M. Yes, but I haven't really spent much more time on that.

Than just hearing a bit... :)

@kirma your comments on using the spherical Lloyd algorithm reminded me of Goldberg polyhedra: lots of hexagons, with 12 pentagons thrown in just so it closes up.

Yep yep.

I really wish there was a way for DelaunayMesh[]/VoronoiMesh[] to support custom distance functions...

It actually works when I don't shoot my foot by using restrictive bounding box. Good that I didn't write a stupid bug report. :)

Hi All, is it possible to connect, with a line, the origins of two plots in a GraphicsRow?

I plan on asking this as a question, but I thought that the question of connecting features of two objects might already be explored, though I can't seem to find the right words

@EricBrown Nothing easy (apart from converting the object to an image and then overlaying a line), it looks; in particular GraphicsRow[] does not have an Epilog option.

@J.M. Thanks, this is what I have deduced after a little bit of trial and error. Though I am seeing things like: mathematica.stackexchange.com/questions/2011/…

@J.M. Above pseudo-buckyball has 60 faces, out of which 12 are pentagons and rest hexagons. I'm not entirely certain how regular it actually is... it's a bit hard to think about on the screen.

@EricBrown Well, if you don't insist on GraphicsRow[], you can Inset[] the two plots, and then append a Line[] primitive.

@kirma It's not very regular, but I think that's what makes it a bit interesting to begin with. You also used Lloyd to produce this, I presume?

@J.M. Well, sort of Monte Carlo Lloyd method. Instead of computing Voronoi cells and their centroids during the iteration, random points on a sphere and means of them nearest to previous vectors were used to approximate the process.

@J.M. One of the major inspirations behind this tinkering is interest to produce a 3D-printable piece that demonstrates four-coloring and closes on itself, without being too regular, but still almost so.

@kirma Ah. I haven't seen anybody four-color an (approximate) soccer ball before. ;) Nice!

I already ordered a 3 cm version through Shapeways... being mostly afraid of color reproduction. It seems to be one of the things that is awfully hard to get right anywhere.

@kirma I had thought that as long as you don't ask for really fancy colors, it would be alright.

I'm hoping that they're clearly separable and not too fuzzy on the edges (I did some calculations under which I tried to convince myself of this). I'd want colors to be reasonably faithful with colors on display and preferably comparable on levels of lightness and saturation, but that might be a bit too much to ask for, especially that I think there are at least two points along the way where this color workflow has some great unknowns...

Well, it's just a 20 euro print, time to write that code has been much more valuable than that. It's just that half-a-month latency of submitting an order to receiving it is a bit disturbing. ;)

@kirma That's not much compared to springing for your very own 3D printer, no? :P

@J.M. Yep, and technologies available (precision, materials, features such as colors and coating) through a professional shop are way better than what hobbyist printers can offer now...

I have several of my hollow sphere tinkertoys printed in polished silver. It would never happen with home equipment!

That's what I thought. ;)

Its been exactly 4 years now, on Feb 17th 2012, when Theodore Gray blogged on Wolfram Blog about CDF on iPad and a little teaser movie showed that there was a working version of CDF Player/Mathematica on iOS. Clearly WRI gave the Cloud a higher priority... On WTC 2015 a talk was given about "mobile deployments" of WL on iOS and the presenter assured that the iOS CDF Player is real and will possibly be available in late 2015.

I wonder why they still do not deliver it along Mma 10.3.1.0, I still think a native iPad version would make a nice deployment platform for WL.

It has been a bit quiet on the Wolfram Language in Intel Edison front too, I think.

@Xacobeo2002 I noticed an iOS front end mentioned in this presentation: library.wolfram.com/infocenter/Conferences/9254

@Szabolcs "Not to mention" can go both ways :)

I haven't watched the video, but now I got curious.

He says that the stuff shown in this presentation can be expected "if not by the end of the year then very early next year"

Some parts of the presentation appear to be cut (??), so many things from WTC are still NDA'd then?

64-bit Cocoa port for Mac, he says it's more work than was anticipated. I guess we might have to wait longer for this? :(

I really-really want some customizability for auto-completions.

Screenshot above: File means that file-name completion is active for the first argument; the second can be any string from a give list (like $ExportFormats).

I'm just not getting the point of TextGrid ...

"It's for output but for output that you might want to edit as well", but it doesn't support things like TextGrid[Dynamic[table]] ...

@Xacobeo2002 I don't think they were ever intending to produce/release any of them when they were announced. There's a term for such marketing in software — Vaporware. These mythical "breakthrough products" that will "revolutionize X" and is a "new kind of" stuff is all bogus marketing designed to keep you interested and wanting their products. Some of these might be even empty words to gauge how popular a "product" is without actually building it.

> In the computer industry, vaporware (also spelt vapourware) is a product, typically computer hardware or software, that is announced to the general public but is never actually manufactured nor officially cancelled.

I don't feel full vaporware is a fair description of it. Pompous overpromising due to oversight by only one decision maker might be it.

@R.M. So iOS Mathematica is the Duke Nukem of the CAS scene?

As far as I have understood, Apple is consistently hostile towards full "alternative" runtimes on their mobile platforms. That means, one has to restrict usefulness of the app to the extreme to even get it in the app store. Full Mma? Highly unlikely.

@kirma Hard to tell the difference between pompous overpromising and a vaporware :P

@R.M. It depends on limits you set for vaporware...

Also, for some of them, I'm sure there's at least a few developers at WRI who did actually put in a lot of effort into building the product or at least a prototype of it, only to have it be killed by the man at the top. And they can't speak about it because of NDA and all that...

Good for them that their inventions are not classified export-restricted weaponry. It makes life one more level more interesting...

-> bed.