@Verde Tell you what: if there are at least ten questions entirely about manipulating derivatives, then we make a derivatives tag; otherwise, they can stay under the calculus-and-analysis tag for now.
@acl The $m_i$ are arbitrary unitary matrices? The thing you're proposing isn't obvious to me at a glance; I'll have to think a bit...
@J.m. Yes, but from my brute force approach this should be true for arbitrary matrices (but I haven't checked numerically now that I think about it). Thanks for thinking about it (nevertheless do not spend much time, I can show it by manipulating indices, I was simply wondering if it can be deduced from some property I don't remember)
I'm trying to do this: i = RandomInteger[{1, 12}] If[(p = RandomInteger[{1, 12}]) == i, p, RandomInteger[{1, 12}]]
i generates a random integer, and the function above will generate another integer that must be differente, I want to build a function that will try to re-evaluate it again, in the case of it being equal to i.
@OleksandrR. Yeah, but I don't really know if his intention is to calculate the LCM, or if it is a toy problem for understanding something else. That's why I didn't post an answer
Srinivasa Ramanujan FRS () (22 December 1887 – 26 April 1920) was an Indian mathematician and autodidact who, with almost no formal training in pure mathematics, made extraordinary contributions to mathematical analysis, number theory, infinite series, and continued fractions. Living in India with no access to the larger mathematical community, which was centered in Europe at the time, Ramanujan developed his own mathematical research in isolation. As a result, he sometimes rediscovered known theorems in addition to producing new work. Ramanujan was said to be a natural genius by t...
Word to the wise... FindFit/NonlinearModelFit/(probably the other fitting functions too) have a parameter localization bug. Parameters of the form f[g, x] are localized correctly, but those like f[g[x]] aren't. I found this quite obscure.
Most bizarrely of all, the bug does not manifest if you use the (undocumented) Method -> "PrincipalAxis" option.
I need to solve an optimization problem, which is defined in a Mathematica notebook.
Using Mathematica's FindMinimum is not an option, because it is too slow. So, the idea is to use an external solver for quadratically constraint quadratic problems and use MathLink to get the constraints of the...
@Rojo out of interest, how did you initially learn Mathematica? I learnt it by myself from the docs and through my own experiments, but increasingly I get the impression that this is uncommon and most people learn it from others--a taught course, colleagues, asking questions, etc. I never felt like my doubts would be of interest to others, though. After reaching a certain level of proficiency that of course changes, but perhaps I'm stuck in my ways now.
@OleksandrR. I learned it by myself, docs, experimenting. I was never a forum-active person, and I gave the mathgroup a shot even when it felt weird to ask. I ended up asking 3 or 4 only, not very satisfied with it anyway. Google groups, days between questionn and answer
Receiving answers by mail in private, receiving warnings that someone is currently on vacation
Hehe
And then I think I've learned quite a lot since I arrived here, by asking and reading
I think it helps a lot to be around people of similar and higher skill levels than oneself. There's a lot of experience on MathGroup, no doubt, but it tends to be drowned out by the n00bs.
I never asked any questions of MathGroup either, for that matter. My first post was demonstrating how you can use MathLink to communicate between different parallel subkernels.
The whole mechanics of a by-mail forum isn't tuned for programming questions. You can't wait for a week for someone to come out with ... "hey ... use SetProperty[ ]"
@Rojo agreed with that. I just wrote a package to answer this question. Overkill? Yes, but it's something many people might find useful. Will post in a while.
@Verde you're right. I never bother with bulletproofing. If it doesn't fail for correct inputs, I consider it a job done.
@Rojo it would be nice not to have to care about licences. Unfortunately, software licensing seems to be as much a political as a legal/technical exercise, and people will make a huge fuss if you offend their sensibilities.
@Verde perhaps he's under the misapprehension that FindMinimum, like NMinimize, is implemented in top-level code. I'm not downvoting because the question is potentially interesting as it stands even if the statement is incorrect, but I understand your point of view.
@OleksandrR. I don't think blaming about a function behavior without showing the reasons are of much help for others here. We all know that many implementations in Mma are far from perfect, but the whole trick is understanding where the pitfalls are
I agree in the abstract, but for me it's not so much the why but the what (how to pass lists of symbols using MathLink) that's interesting. Of course, one still wonders whether he tried Method -> "LevenbergMarquardt".
@LukeAllen literally speaking, no. (OpenerBox seems to have a fixed appearance.) But you can replace the Opener with a Toggler, which can look however you want.
@LukeAllen There's a cell style called Section, and you can collapse cells with lower hierarchy, such as subsections. When you do, you can make the section cells have an opener on the side, to open or collapse the section
@LukeAllen well, usually the answer is "very much so". But often it's a bit obscure. I don't know particularly well how to do it myself and I think you have to expect that most things like this are going to be undocumented. In the case of the cell group openers, Cell has an option OpenerBoxOptions that you can use to change what an OpenerBox looks like.
does anyone know the story behind why when you copy a tiny raster image from a pdf and paste it into mathematica, it will be about 5 times bigger than it was in the pdf
is mathematica zoomed in by default or what
ok it seems to be because it will paste the 100% image size of the image in mathematica
@acl that's what I thought too, at first. Then people started mentioning that they knew of Mathematica courses, which I had assumed didn't really exist (except for the training that WRI provides). Given that they do exist, and that some universities presumably use Mathematica in their teaching, I wonder how many people learn that way?
@LukeAllen yes, let me check my notes on how I did that
@Verde but in the MG you'd be drowned out by those (I have no problem with that) but also by a small number of people with strong opinions. my metric for how useless a discussion is is the number of appeals to authority; there are (or were, I don't read it any more), lots of those there. bit like usenet.
@acl I changed ISP and my news server stopped working so I can't post any more, but I do still skim the posts on MG. Most of them are like you say, but there are a few that I think, "would have been better if you posted that on Mma.SE". This, for example... no way can we mention v9 features on MG.
@Verde come on, you're being too harsh! my C skills are negligible and yet my C code is generally faster than my mma code, even though I've spent years obsessively learning tricks to speed mma code up. it's a reasonable conjecture he's making (but the question is too localized as it is)
@Verde see you
@OleksandrR. So, are you procrastinating in preference to writing your thesis?