@TimStone hi

Folks, what do you think of these posts:

The OP keeps dumping his NDSolve code wholesale. I am not really in favor of that. Most of his problems are rather basic and the answers might be helpful, but with all the dump overhead they become too localized.

Any opinions?

@YvesKlett But he looks nice in his suit!

Very stylish

But - what about the question style? I feel it is impolite to let others debug your code...

@halirutan but then you said you love finding bugs :-)

The thing is, I think he needs generally a bit help for a start on SE. All his question votes give a total of 0.

But how to go about encouraging him to help us?

@YvesKlett I would vote to close this too, but maybe we can invite him to chat or write a comment that he has to fix his question style to get answers.

Out for lunch...

@YvesKlett Guten Appetit.

@halirutan Yum

@YvesKlett, @halirutan: people should be politely pointed to the documentation or some videos by WRI or a real Mathematica (live) course or whatever. M.SE should not get drowned by this type of total beginner questions. Is the goal to be a world-wide free Mathematica tutor? Most certainly not. I just usually ignore this type of questions, or give a small comment where to look else (like subbu with his strange InputField question: why should we do the work for them??)

@RolfMertig I hope that some of them are maybe beginners in Mathematica, but advanced computer scientists which adapt very fast with a small boost at the start.

Anyway, I remember being wrong about this more than once.

Or we could make a customisable CDF to edit/extend InputAutoReplacements settings

@RolfMertig I want no replacements, I want suggestions depending on what I have typed so far. I really hope John is working on it.

Btw, I have 9.0.1 since about a week on DVD here but I didn't install it.

Actually: I'd rather want suggestions in WWB ...

@RolfMertig Me too (of course additionally) but since we don't have the code you would have to rebuild the whole plugin by yourself. Pain in the ass.

@Xerxes Thanks again and welcome

@Xerxes By the way, remember to ping, either by clicking on the arrow at the rhs of the message you are answreing to, or by typing @username. I saw your last version by chance

@halirutan We could do the same as this brave lady [here] (undo-for-mathematica.alaifari.com) and start a petition for a more useful and modern WWB or for opensourcing it (which will not happen I think). I wished WRI would pay more attention to heavy-duty WWB users ...

@RolfMertig I've seen this.

Wouldn't it be better to speak with the Wolfram people directly?

I don't know. I haven't seen them doing something similar since I use M

@RolfMertig Are you using the Debugger in WWB or something else beside the basic editing?

Hey all

Hi @Rojo.

Hey @halirutan

@RolfMertig I do share your feelings. Downvoting / closing will probably just work without too much need for extra initiative.

I was just thinking maybe voting to close would be too harsh

@halirutan aaaand: there are computer scientist who do not adapt at all (not here, though, but browse mathgroup for RJF)

@YvesKlett Richard Fateman is funny!

@YvesKlett He is the one with the better rules right?

Did you follow the Mathematica vs. Lisp thread recently?

He is funny, but has no sense of humour at all (strictly judging from his posts)

I´d love to see him participating here - oh the fun!

@YvesKlett I read only the best parts.

OK, guys. I have to work a bit.

See you later

See ya!

@YvesKlett He seems to dislike Mathematica. Though he has at least two valid points: the pattern matcher is confusing to beginners and there is no formal language specification. Or, maybe more generally speaking: documentation needs to be improved for professional (industrial) applications. Right now it is not enough (check out "EnterpriseCDF" docs. Or try to produce a small PDF from a Notebook in version 9 (I mean how to suppress the enormous inclusion of fonts etc. etc.)

@halirutan But I kind of do this now! Since they read it (I hope). Directly makes not much sense. I spoke to Roger G. at the conference in 2011, and urged him to include/integrate WWB by default into Mathematica (just like there is a Java delivered by default): he just said no, "it's not Mathematica-like". What can you say? That programming only in the FrontEnd is a dead end for bigger projects? Maybe they all should once do bigger projects? I don't know.

@RolfMertig every serious user will encounter problems, sometimes grievously so

@halirutan I tried the Debugger. It is too slow for me and usually I can debug code faster by using directly the Debug[] function, or Print, or by inserting Global` debug variables.

@RolfMertig I think to some extent he just likes arguing with people... he is a retired professor after all! Though I would say that many of his points are valid or at least useful observations.

and I keep buggering anyone I meet for the stuff I´d like to have

@RolfMertig agreed. I ignore them as well. On the other hand, if everyone does, the site will be flooded with questions with no votes and no answers

@RolfMertig but I guess it would be fair to say his comments are strictly non-constructive

@OleksandrR. so what? but if we educate total beginners they will stay and come back and attract even more total beginners.

@YvesKlett Problem is he does not understand Mathematica (pattern matching). But of course someone like him should ... Yes, not very constructive.

@OleksandrR. but arguing and whining are two different things. But together with Andrzej Kozlowski this is a sure source of continuing casual entertainment

@RolfMertig the site will not attract anyone if it looks like nobody gets any answers, so we have to do something about the questions rather than just ignoring them. I agree with you; I think questions that are too basic or too broad should be considered either off topic or too localized. But many others do not agree with that

@OleksandrR. would you say we are voting too beneficially?

@All : anybody else correcting students work in notebooks? I wrote a bit of code to look at CellChangeTimes values being the same for different notebooks. Do I need to do more statistics to "prove" that they cheated? Also, of course there is no real way to see if student A copied from B or B from A.

@RolfMertig was there not some Q/A by @SethChandler?

@RolfMertig: mathematica.stackexchange.com/q/17563/131

@YvesKlett not voting, no. I think votes are reasonably cast when they are. But a lot of people are, quite reasonably, ignoring a lot of questions now. My point is if the volume of mediocre question continues to increase, at some point we'll have to decide concretely how sophisticated and general a question has to be and act accordingly, in order to avoid being flooded with low quality material that everyone ignores. This effect is evident for many tags on SO, and it seems to be quite harmful.

but probably not too related. I thought there was a question around concerning that

@OleksandrR. Fine. Maybe then let those people who want to answer that type of question do it. On the other hand I find it sad that experienced people like MH (trying to help subbu today) are then pissed off by subbu changing the type of question (and he also does not provide specific code). So I think if people do not put real Mathematica code into the question , maybe they should be told to do so immediately? I think just basically asking "can you write the code for me for free" is off-topic, or?

@OleksandrR. agreed - so more closures?

@YvesKlett that's a different thing.

bye. need to run.

@RolfMertig yes, I agree with you. subbu and a few others have a bad reputation for incomprehensible questions. Several of these users (subbu included) have been suspended more than once without much obvious improvement... closing the bad questions seems to be the only thing we can do

@ArnoudBuzing hi!

@YvesKlett Hi!

My gripes with the help system continue in spite of 9.0.1

Try entering " ' " in the help dialog - I get a Dynamic timeout ("Disable Dynamic Evalution")...

hm

Sorry to dump immediately - but you are kind of an obvious target around here ;-)

I don't see it yet ... The three characters """ or just the ' or the three characters with spaces " ' " ?

Just the '

ToCharacterCode["'"] -> {39} correct?

@Yves you're running windows, right? Can you tell me the result of SystemInformation["Small"]?

as in Derivative

I don't get a hang, but it is not finding Derivative either ...

@Arnoud That's what I get, on mac.

Which one, the hang or not the hang?

same result you get, no hang, no doc.

ok

@ArnoudBuzing do you have a copy of version 8.0.0 to hand?

If so, perhaps you could try this question and see if you can confirm that this is specific to that version?

@ArnoudBuzing the question mark "?" does the same (not always, but annoyingly often).

? as in Definition...

Or PatternTest.

@all any, what is the best way to set global antialiasing preference programmatically (the option inspector´s "HardwareAntialiasingQuality"`)

@OleksandrR. Installing now

@rcollyer not to be sneezed @!

uh, and patterntest got another timeout. Is my system so messed up? I did a clean start with preference reset already to no avail.

@ArnoudBuzing ah, thanks very much. Sorry, I wouldn't have asked if I thought you would have to install it... I thought maybe you have VMs with every version. :)

@ArnoudBuzing good one - first evaluation stalled and I interrupted, the second evaluation spammed my screen all right :-)

in fact it is still spamming

@YvesKlett Yes, this will open 200+ windows

this reminds me a bit of my oneliner this year - the frontend can take a fair number of windows

@YvesKlett Training day for the close button finger I guess ;-)

NotebookClose /@ ...

@ArnoudBuzing No cheating!

@halirutan just you wait for that oneliner (although I was told it did not work on Mac)

@OleksandrR. The first input gives me: 1/2 ([Pi]^2 Zeta[3] - 11 Zeta[5])

(in 8.0.0)

@halirutan this will train your fingers all right

but I dare not post it here - it is rather badly behaved

@ArnoudBuzing remind me never to run that again ... although the resulting window looks like it has a very large drop shadow now.

@rcollyer a new benchmark is born!

@YvesKlett I thought certain unsavory websites already had developed the window spam test.

@rcollyer but now you get to pay a few k$ on doing it the scientific way :-)

:)

o.k., I´ll say it... a new kind of spamming.

back to work now

Yes, you've caused enough trouble for one day. :)

I try to give as good as I get - see ya!

Bye.

Is there a way to check for permutations without sorting ?

@ArnoudBuzing thanks! That's what it should give. This OP seems to have a problem with their setup, then.

@The-Ever-Kid do you mean, check that two lists are permutations of each other?

@YvesKlett Here, try this too: MathLink`CallFrontEnd[FrontEnd`UndocumentedCrashFrontEndPacket[]]

@rm-rf now that's just mean ... enough for me to star it. :)

@rm-rf wicked!

I´ll keep that to confound my dear colleagues :-P

in times of anger

and April 1st is also drawing near

what about a question: How to most spectacularly crash MMA?

@RolfMertig @OleksandrR. While it's fine if people choose to ignore, I think it would be nice if at least the long time users/experienced users also threw in a downvote/comment/close vote when appropriate (one or all three, depending on personal views on certain actions, etc.). Users like him are repeat offenders and drives everyone crazy, but if people simply ignore and don't take any action, then the lone person that comments on the quality of his post comes out looking like a jerk...

@rm-rf yes, I agree and I do this. What I was concerned about though is whether repeatedly penalizing a particular user (even if entirely justified) might come across as overtly hostile, and for more marginal questions if a close is actually achievable in this manner. I'm thinking of those questions where there's nothing badly wrong with them but still nobody feels that they're particularly worthwhile. There seem to be a growing number of these judging by the # of questions with no upvotes.

@rcollyer yes....

or more precisely two numbers. Sort[IntegerDigits[#]] is what i usually do.

Is this link completely gone? mathematica.stackexchange.com/questions/19316/…

There was a useful suggestion within it (to expose/document enum things like System`DateStringDump`cachedShortMonths["en"])

@ArnoudBuzing The OP deleted it...

@ArnoudBuzing I made that comment and Leonid also made a comment that it could be generalized along the lines of "best practices for defining enum-like types in Mathematica". I'll ping Jagra to ask something along those lines

thanks, it seems reasonable to have these available as enum types (and not having to write code for something that basic)

I agree... I came across that when writing this answer and considered using it, but eventually decided to go with just month number instead of the month name

@The-Ever-Kid I don't think you can avoid a Sort, but there are ways to make it faster. For instance, since you are only looking at single digit integers, I'd do something like a RadixSort which is very fast. You can implement it yourself, or you can re-purpose one of the built-ins for it. Tally works nicely for this, but you must run a SortBy[res, First] on the result. HistogramList, though, works perfectly for this, as follows:

HistogramList[#, {0, 9, 1}] &@{0, 1, 1, 7, 6, 4, 5, 5, 4, 1}
(* {{0, 1, 2, 3, 4, 5, 6, 7, 8, 9}, {1, 3, 0, 0, 2, 2, 1, 1, 0}} *)

where the second list is the "Radix". So, to compare two numbers run

HistogramList[IntegerDigits[num1], {0, 9, 1}][[2]] ==
HistogramList[IntegerDigits[num2], {0, 9, 1}][[2]]

as they will be permutations of each other if their radices are the same.

I am still flabbergasted every time I say PlotLegends->{stuff} and the result doesn't suck

@acl I could tell you who worked on it (so you could thank them), but the NDA would come into effect and I'd forget what I was telling you, and the past year in its entirety, and quite possibly how to tie my own shoes ... :)

@rcollyer ah. thanks then :)

Hi all!

What exactly is happening here? Why does DataRange mess up the plot? Try with and without DataRange:

I'd expect DataRange to simply rescale the coordinates to fill another range.

@rcollyer You might be interested in this one above

@rcollyer Now I can go back to adding legends to the plots in my papers, rather than verbally state things in the captions...

@Szabolcs odd, very odd. I'll look into it.

@acl I resorted to using LevelScheme's energy level schemes to do it on a number of occasions.

@rcollyer It is possible that DataRange was never meant to work for the case when you specify coordinates explcitly, but for some reason I was under the impression that it did in the past. There's an easy workaround with Rescale.

@rcollyer I wrote my own routines for plots in my notes. But I could never get myself motivated enough to make them presentable

@Szabolcs I think it was rewritten a while back, he just never got around to pushing it out.

@Szabolcs try this:

data = Table[Sin[i + j], {j, 0, 2 Pi, 0.5}, {i, -1, 1}];

ListContourPlot[data, DataRange -> {{0, 10}, {0, 50}}]

@rcollyer That works. That is the case when you give the data in matrix format, i.e. only the "z values", but not the x and y values.

@Szabolcs are you seeing what I'm seeing? That worked for you?

Oops

No it gave this:

That's what I'm seeing. Need to investigate. Thanks ...

@rcollyer I think it happens because there are only 3 values in the i direction. If I change the Table iterator to {i, -1, 1, .1} then it's all fine.

Maybe it's something about how the system guesses whether this should be interpreted as a grid of values or a list of coordinates.

Nope, still not right. Look at it without the DataRange.

Never mind.

without DataRange it's interpreted as a coordinate list

I should take my own advice. :P

It's kind of like ParametricPlot[{Sin[j - 1], Sin[j ]}, {j, 0, 2 Pi}] without DataRange

but to me it's still not completely clear how DataRange works

I think you are right in that it shouldn't be applicable with the coord list form. What it is doing, instead, I need to look into.

@rcollyer Have you seen this one?

yes. But, I have to be very circumspect about what I say. See above.

Night all.

@rcollyer Well, the only thing I'm concerned about is that WRI learns about the bug, so hopefully there will be a fix. The workaround is already there (Simon's post)

@Szabolcs upvoted it, too.

@OleksandrR. Are you here?