@belisarius I see. There's still another mystery... Yesterday I think I got about 5 upvotes to a quite old answer, and I didn't find a new link to it... (nor I searched) Interesting
Not a general solution, but works for the case you cited;
(*Definitions*)
an=6;
bn=4;
h[0,x_]:=x; h[1,x_]:=1/x;
h[2,x_]:=1-x; h[3,x_]:=1/(1-x); h[4,x_]:=(x-1)/x;
f[x_] := Sum[a@i x^i,{i,an}] / Sum[b@i x^i,{i,bn}]; (*Proposed form / Rational*)
(*Now find the coefficients a[i] and b[i]*...
Let's say I have a function
formula[x_List] := (x[[1]] - x[[2]]) + (x[[3]] - x[[4]]);
and I want to pass it a variable and get back the unevaluated formula with these inputs in place, so for example with input {1,2,3,4} I would get back
Out[]= (1-2)+(3-4)
I do not want it to return strings;...
@BrettChampion Ah, common typos are a different matter altogether. :) I believe R.M brought up the danger of Quit[]/Quiet[] a while back in this room...
I used to always write a Hide[] function, which does the opposite of Show[]. It's so I didn't have to always set DisplayFunction -> Identity for each series of graphics that I was using...
I've always had the urge to tell one of those code dumpers who post a giant block and ask what's wrong with their code to turn On[] and look for errors
@RM The first I purchased was v.3, and that is the first I recall the version number for. But, I used a prior version on the Sun systems as an undergrad.
@Verbeia hola!
I just spent a little bit of time trying to answer this question. Glad it's over.
@Verbeia you don't give yourself enough credit. Actually, it is pretty simple, just an extension of the calculus you know to functions that transform functions into numbers, e.g. definite integrals.
The trick is to think of it as a gradient with a continuous index.
I came across Luc Barthelet's wonderful game of breakout:
How can I extend this game to three dimensions? The bricks should be cuboids and the paddle a transparent rectangle (e.g ultranium3d)?
segmentIntersectQ[{p1_, p2_}, {p3_, p4_}] :=
Module[{t1, t2, sol, u1, u2},
sol = Solve[
...
How is this not a real question? It may be a big code request, or unpopular, or whatever, but it is quite certainly a real question with a real answer if someone wants to take the time to write and give it.
@BrettChampion I think that was my first use of init.m. :-)
@MrWizard It is of the form "here is some code; how could I extend it to a completely different situation". it would require rewriting more or less the whole thing, and there is no indication of "to do X, I tried Y but got stuck because Z"
If such questions are allowed, I have some toy DMRG code that I'd like to speed up. I am pretty sure it is less work than this question. I don't think it is acceptable of me to ask people to do that for me, and don't think that this being a cool-looking game makes it different.
@MrWizard If you take "not a real question" literally yes. If you interpret "question" in the narrower sense of what constitutes a question on mma.se, then I suppose no.
and it asks for some code in mathematica. I can't see how it is off-topic, if I am going to interpret "topic" literally.
his code could make a nice blog post though (@Rojo and @Heike were discussing Tetris)
"I can't see how it is off-topic, if I am going to interpret "topic" literally." -- that's just the thing; you may not like this question, but that's what down-votes are for. Wearing my pro-tem moderator hat I must find a justification for the closing or consider reopening.
There has been some debate over whether code-review and code-request questions are On Topic, and if so to what extent.
Actually I think this answer by @verbeia and the comment by @Oleks express my position well.
now @Oleks says "unless the techniques used are useful to others" so again we must interpret things--techniques useful for 3D interactive games would likely be useful in many other games, but I still object to people posting walls of code, however well-written, and asking for them to be generalized to different situation. I don't think it is a "question" in the sense of this site.
I read the referenced post and comments. It focuses on "Code Review" questions. How would it change the argument if the closed question included no code and simply asked how is it possible to write 3D Breakout in Mathematica?
Look. "I want to write a flight simulator in Mathematica. How do I do that?" that's on topic, a question, constructive, and not a duplicate at the moment. So if this were to be closed it would be "too localized". So I guess I would accept that it should be closed as too localized, and not "not a real question", but it should be closed.
@MrWizard I thought so, yes. once it has two great answers, though, I did not see the point of voting to close (I saw it after one had already been posted anyway). In addition, this is a much simpler question to answer, and it is a single question: "partition plane into unique, non-interlocking pieces", not a series of problems to be solved culminating in a working game.
Would it be fair to summarize your position as: the reason for closing the question of discussion is a combination of overly broad in request and too localized in application?
@SjoerdCdeVries If I am following acl's argument that question is more acceptable (or has a lower "closing weight" as he put it) because less broad in request: it doesn't ask for a full program but merely for improvements.
Are you inclined to reopen the 3D Breakout question? If you are it might be good to talk to a couple more of the closers first, just to better understand.
@belisarius I think I'll delete it before you know what hit you! (Honestly, I really would prefer NOT to be part of putting PE solutions in public view.)
@MrWizard I'd say it's borderline. Question was put effort in, and some really nice things may come out of it, but it won't be very useful for the general public.
@MrWizard Sounds like something for the blog, though; issue a "challenge", ask people to send in soutions, and then have a succeeding blog entry discuss the "best" submissions.
@JM what do you think of the question closure? I'm okay with it on the basis of acl's reasoning, but I think it's important that I heard his reasoning.
I'm just trying to be fair; if people would have closed my puzzle pieces question but did not only because I have a lot of points or am a pro-tem moderator, that's not fair. At this point I think I both understand and agree with acl's logic.
@MrWizard Fine, here's a more solid position: if my vote were not binding, I'd vote to reopen, just to give the question a fair shake. Still, it seems five people feel more strongly about it than I do.
@SjoerdCdeVries I am not sure it's that important if the question was "really" faced. what does "really" mean? maybe I am idly wondering how to make some code I wrote for work faster. is that a real problem? maybe I've already published a paper on it and I am now done, but decide to lay with it over the weekend. is it "really" now? maybe I am curious on how some numerical technique works and try to implement it; is it a real problem then? it's just not well-defined.
When I work in other languages, I often want to use some of Mathematica's constructs. Today, I'm working in BASH and I have a file name I wish to break up into parts. With mma, I would simply set the results to a list of variables I'd like set. With BASH, it's not that simple. :|
Seeing as there are no good repositories for Mathematica stylesheets and packages, I created a github account for the community. The account resides at github/stackmma. I know there's library.wolfram.com, but we should be independent of WRI.
On the topic of this question, I created a public styl...
The things that I'm looking for are:
1) licensing issues — I know stuff here is cc by sa, but what about the attachments that users host on dropbox? Is there a problem with rehosting it on github? By posting here, is their attachment also licensed the same as their post?
2) How can this be made permanent? I'm not very familiar with github, but it seems like I or anyother collaborator could wipe out the uploads just as easily as the OP who linked it to dropbox unless someone else has cloned and created a repository for themselves.
@JM Good point, I should let the OP know, at least out of courtesy.
FIX: Visual C++ compilers are removed when you upgrade Visual Studio 2010 Professional or Visual Studio 2010 Express to Visual Studio 2010 SP1 if Windows SDK v7.1 is installed
@SjoerdCdeVries if this is about jfklein adding lightweight-grid to a couple of Jagra's questions, I thought it was legitimate and approved it. I can imagine it being used again.
@SjoerdCdeVries I think we are doing a reasonable job of tagging, not creating tags unnecessarily and cleaning up and documenting those tags. István in particular is doing an amazing job tidying things up.
I had been working towards the Research Assistant badge but he beat me by miles.
@SjoerdCdeVries glad to hear you found the source of the problem. Actually in the instructions for installing Visual C++ I posted on Mathgroup I advocated installing directly from that package rather than going through the VS Express + SDK + SP1 + hotfix dance. So, I never encountered that problem and didn't know what its symptoms would be.
@Heike I would rather not have new tags unless absolutely necessary. It strikes me that we have too many already and managing them is causing a considerable nuisance.
@OleksandrR I agree we shouldn't create many more, but the one jfklein was about a specific product and I think those are legitimate. It has been a long time since I couldn't easily find obvious tags for my own questions.
@SjoerdCdeVries IMO it is perfectly fine. That is indeed a specific product which requires its own tag. I don't know anything about the Lightweight Grid, in contrast to, say, Parallel` tools in general.
I always set it all up manually in the past, but I haven't needed to do any parallel evaluations for a long time. (I bought the old Parallel Computing Toolkit merely for my own interest, not because I actually needed it. Then Mathematica 7 came along and made it obsolete.)
@RM No, she doesn't use mma that much anymore. Was never more than an occasional user, but she still has a one premier license, so if mma 9 pops up we'll have it at home ;-)
@SjoerdCdeVries I have a student license, but that doesn't give me access to upgrades. There is a campus "license", but labs have to subscribe (pay) for it, which mine hasn't. Most people use MATLAB so that's what we have a license for.
I managed to convince my advisor to get one for the group (at least a single user one so that I can use it on the computing cluster) and looks like we'll get it this Fall (that's when the license agreement begins)
@SjoerdCdeVries That sounds a bit like two former colleagues of mine who were also married to each other. Only with them it was macs. I think they owned about 6 between them and that was excluding the ones they used at work.
@RM honestly though, if you were director of sales at WRI, and you saw this great community of enthusiastic users all of whom had paid full price for the product (many of them multiple times), would you be inclined to give anything away for free? I know I certainly wouldn't.
@RM Maybe have a contest where the goal is to create an application using new features from mma 9. The winner would get a premier licence. The contestants would be provided with home licences to be able to participate.
@Heike Honestly though, if multiple undo is a new feature, I'll never get an application done for the contest because I'll be busy undoing everything BECAUSE I CAN!
They put in all those fancy finance charts, which are fine if you are a chartist, but didn't put in something simple like a bar chart for dated data, or an easy way to have bar and line graphs together.
I'm wondering if I should post a new question... I just want to know how to add only one 0 to the beginning of sublists, not like what was proposed here
@OleksandrR The c compilers seem to work, but now I don't seem to get CUDA compilation back to work again. Installed VC 2008 as well because the nvidia cudacompiler doesn't seen to like 2010, but it doesn't see it. I'm going to bed, frustated
Not a general solution, but works for the case you cited;
(*Definitions*)
an=6;
bn=4;
h[0,x_]:=x; h[1,x_]:=1/x;
h[2,x_]:=1-x; h[3,x_]:=1/(1-x); h[4,x_]:=(x-1)/x;
f[x_] := Sum[a@i x^i,{i,an}] / Sum[b@i x^i,{i,bn}]; (*Proposed form / Rational*)
(*Now find the coefficients a[i] and b[i]*...
Let's say I have a function
formula[x_List] := (x[[1]] - x[[2]]) + (x[[3]] - x[[4]]);
and I want to pass it a variable and get back the unevaluated formula with these inputs in place, so for example with input {1,2,3,4} I would get back
Out[]= (1-2)+(3-4)
I do not want it to return strings;...
I came across Luc Barthelet's wonderful game of breakout:
How can I extend this game to three dimensions? The bricks should be cuboids and the paddle a transparent rectangle (e.g ultranium3d)?
segmentIntersectQ[{p1_, p2_}, {p3_, p4_}] :=
Module[{t1, t2, sol, u1, u2},
sol = Solve[
...
