@Szabolcs I'm sorry, I really don't know. It sounds like an interesting problem, though.

@halirutan unfortunately, it doesn't. I think .VersionID files were a new addition in version 6.

@halirutan the .CreationID file is still there, though. On my system it contains 530575, whatever that means.

@OleksandrR. This is very unfortunate.

@OleksandrR. I assume a .Revision file is missing too?

Newbie Q, anybody: say a is not defined; then defining a function f[a_] = a^2 with Set = or SetDelayed := is the same, true?

What I'm curious about too: := localizes formal parameters' names, and = does not?

For me it's the same. See DownValues[f] : they are the same.

@BoLe The difference between = and := is that = evaluates the right side while := does not.

Therefore, it is wrong when you say := localizes the formal parameter.

Because by explicitly evaluating the rhs, the localization is lost:

x = 13;
g[x_] := Evaluate[x^2];
f[x_] = x^2;

And now compare

DownValues /@ {g, f}

@halirutan What does the different syntax coloring mean in f[x_] = x and f[x_] := x on rhs (is x removed or not)?

@BoLe I would say in = it clearly tells you that the Global value of x is used.

I wouldn't pay much attention to the coloring. It is just convenient because usually you use := for function definition and there x is the formal parameter. Since the explicit Evaluate is rarely used, you could think of x as never having a value and just being a placeholder.

@halirutan But coloring very helpful. So aside from Evaluate and whatnot x is localized to the setdelayed function's body, or I can think so at least?

@halirutan And I delocalize it with Evaluate?

@BoLe Think of f[x_]:=x as replacement rule because this is what it is.

Is anyone finding this Free CDF useful when you cannot import files from the local drive?? I really ask myself why Wolfram makes such restrictions, when a 7year old can circumvent them..

@halirutan You can circumvent the law by pick pocketing a four YO at a bus stop. That doesn't mean that the restrictions are not in place and are protecting something

@belisarius OK, you lost me in the first sentence.. this is either too much slang or I'm just too dumb.

@halirutan Probably my English isn't clear enough. I mean that the easiness in breaking a law isn't related to the punishment for doing so

And I find the restrictions really annoying. I mean, in real life I don't want to make little small demonstrations.. I want to make something useful. This client/colleague of mine wanted to quantify some (approx 600) images and I wrote a small tool which works nicely. The only thing is, she has no Mathematica but on the other hand she just want to measure some images.

She doesn't need to write files but she needs to load the images..

@belisarius I'm not breaking anything.. I'm just bending the interpretation of "loading from websites".

@halirutan I'm NOT trying to say that Wolfram's policy is reasonable!

@halirutan They are just betting that restricting I/O is going to help them in selling Mma licenses. I think they are plain wrong

@belisarius that's what I basically mean.. We have a large contingent of licenses in our university and I myself have 5 licenses for my group. We have premier service and bla bla bla...

BUT I cannot make a useful CDF without my colleagues having Player Pro or me buying the Enterprise thingy..

:-(

I mean, how should I promote Mathematica when I cannot give useful things away?

@halirutan That's why I think their strategy is flawed. Satisfied users can't recruit others because they are unable to show them the whole thing

@halirutan But you should have a webMathematica Amateur license already. Then just use that to make a small JSP which loads the images, then you generate the CDF's on the fly and deploy them by CDFDeploy. It's not that hard. I'll send you a private email.

@belisarius No, senior. The strategy is fine. Actually one can get quite far with FreeCDF. And then one just needs Enterprise Mathematica. That, I agree, is probably a bit expensive, but it is just cool.

@RolfMertig Hi Rolf! I think WR should encourage sharing Mma apps with non users instead of that being a discussion theme here ...

@halirutan are you here?

okay, I'll try again later. bye

what am i missing: trying to reconstruct sum of sin / cos from Fourier data

then fft = Fourier[test]

how do i get a function out of this?