@Mac Dude 

The basic idea of copying the image really works, but... there are a few important extra details: To save it as a file so that you can rotate it and , after rotation, you can select that png file for use in the components property window, you have to paste it first in some other program such as ms Word from which  you can save the image as a file. Then you have to open it in a program like Windows Photo Viewer so that the rotation is saved when you save the picture. (If you save the rotated image in Word (or try to) what is actually saved is the original unrotated image. Only certain programs will save the rotated image.)

@Carl Love 

Yes, I usually want to use the previously saved version of output so that the code automatically uses the previous output.

However, there are times when I find copying output to input produces much more readable and immediately understandable input in which I cannot make typos and don't have to go through using the palette to get the same result. There are even occasions, such as copying the probe location from a plot, when maple does not allow you to directly save the result without the copy and paste or without going through an indirect setup using an embedded component plot.  

In addition, the copy and paste of output almost always works correctly. In fact, this is the first instance I have seen where it does not work, and here the problem is that the printed output actually has a different structure (using tildes) than the saved output and, on input, the parser does not reverse the process (within all datastructures) and put the input with tildes back into the usual saved structure. Thus, it turns out that some Maple functions (or compositions thereof) when operating on some structures cannot connot handle these tildes.

One just has to keep this in mind and not "throw out the baby with the bath water", so to speak.

This Maplet started life as a classic 1D worksheet. I then attempted to convert it to a 2D worksheet. It seemed that all techniques to do that produced some inconsistencies in the code. Even after everything looked right, it wasn't. The problems were hidden in the "behind the curtain" parsing information. I had to pull many execution groups apart and piece them back together. To make a very long story short, after fixing all these hidden problems, the exported maplet would run independently.

Apparently, the inconsistencies didn't matter when running off a standard worksheet. However, the expoting operation seems to actually turn the code into something like command-line code. The inconsistencies seemed to produce a command-line code that wouldn't run.

@acer 

It's been a while since I wrote Maplets. However, I always used components just like the ones you call Embedded Components for which one can write code which is carried out when the user does various actions, in other words object programming. Perhaps Möbius Apps lack this object programming behavior (and flexibility) since they don't mention this aspect anywhere in their documentation. This limitation to having a separate Maple Player to run the App seems a problem.

Since the Möbius Apps you referred be to all seemed to require Java, that hardly seems different than Maplets. But then again, the Möbius Apps just use Maple, and Maple itself runs on a Java virtual machine.

I myself certainly do not want some "helper" program writing my Maplet or other App. I need to write it in Maple and know and control what it does. Having such a helper program is the kind of thing I always have to spend some time shutting down in other programs because it tries to "out think" me, which means it usually does exactly what I don't want.

Well, I would like to see what these Möbius Apps do. The ones on the websites you directed me to do not run. That is, they often put up a screen of information, but the EC's, like the sliders, don't work. There is no information on how to download the mw files or what is required to run the apps. However, from what I see, there is no improvement over Maplets, but instead some movement in the other direction.

@acer 

Well, none of the deimostrations for either Math Apps or Möbius Apps work on the web. There is also no information on downloading them so that I can run from my own Maple.

I also don't understand the difference you lay out. As I remember it, I set up Maplets on a worksheet also, although it then gets saved to a file with a maplet extension so that students do not have to encounter a full fledged worksheet and cannot mess with the code in it directly.  This way I can always go back to my worksheet to update the code and resave it. I would suppose that Möbius Apps must be set up the same way. Otherwise students have to first deal with the worksheet, see all the code, and start the app correctly, or, after I save it the first time and exit, I cannot get back in at the code to change it. If both formats just have their own extensions to save to for students, then there is not much of a difference since both use embedded components and, I gather, object programming. So where is the hidden difference?And why not just move to Möbius Apps? Also if there is a difference, why is it not explained in the Möbius Apps information.

Same problem, but the documented option is linestyle=solid. Actually, it seems to work with any linestyle option, but the problem seems to be that the documented default value (or any default value) doesn't seem to have been made active.

This problem even extends to the help page on spacecurve. None of the examples show an actual curve. Didn't any proofreader notice this? Of course, if the examples are copied to a worksheet and some linestyle option added, the curves can be seen.

Thanks for the attempt to answer my question, but, of course, I had already gone to the Account page to look for a removal choice. However, I am asking the question because I have found no choice on the account pages to remove an account.

Does anyone know how to do the removal? Perhaps someone from Maplesoft?

@pagan  Yes, that Java fix did it. And after looking at your link I saw what others saw: When I passed the cursor over where the curve should be, I saw the points that would have been used for the plotting, even though no actual plotted curve was visible. Now, with the new mathdoc.jar my curve comes out exactly the way I expected from Carl's original post on the "Eraser" function with transform.

Thank you, Carl and Preben.

@pagan  Yes, that Java fix did it. And after looking at your link I saw what others saw: When I passed the cursor over where the curve should be, I saw the points that would have been used for the plotting, even though no actual plotted curve was visible. Now, with the new mathdoc.jar my curve comes out exactly the way I expected from Carl's original post on the "Eraser" function with transform.

Thank you, Carl and Preben.

Carl,

Thank you for the key to how transform works and how to get to its local procedures.

Actually, my attempt to use transform started with your maple prime response (on 1/17/13) about using transform in a BubblePLot problem as an eraser.

I also wanted to do some erasing in a plot.

You substituted "undefined" (without quotes) for the erased values. However, when I tried that, it actually substituted HFloat(undefined) into the PLOT array for the erased values and the new PLOT instructions would then plot nothing as if there were no data, although I could verify that all the other values in the array were still there and correct and the axes drawn were correct as if it was looking at all the data. Does your code work because there is something special about the BubblePlot?

Carl,

Thank you for the key to how transform works and how to get to its local procedures.

Actually, my attempt to use transform started with your maple prime response (on 1/17/13) about using transform in a BubblePLot problem as an eraser.

I also wanted to do some erasing in a plot.

You substituted "undefined" (without quotes) for the erased values. However, when I tried that, it actually substituted HFloat(undefined) into the PLOT array for the erased values and the new PLOT instructions would then plot nothing as if there were no data, although I could verify that all the other values in the array were still there and correct and the axes drawn were correct as if it was looking at all the data. Does your code work because there is something special about the BubblePlot?

Some CLAPACK routines are used by Maple (according to the help pages). However, the whole package is not bundled with Maple, in the sense that the code (dlls) comes with the Maple program distribution. In fact, none of NAG, LAPACK, CLAPACK, or scaLAPACK are really bundled with Maple in this sense, although some of the routines are apparently used in the Maple code.

All of these packages require as least linking of the appropriate library functions into a dll to be used by Maple's external calling procedure or purchasing a package (say NAG or the Intel math library MKL) with such a dll.  One can also purchase MatLab and use it externally through Maple. None of these are impossible, just many more steps and/or $ past simply running a Maple procedure.

Some CLAPACK routines are used by Maple (according to the help pages). However, the whole package is not bundled with Maple, in the sense that the code (dlls) comes with the Maple program distribution. In fact, none of NAG, LAPACK, CLAPACK, or scaLAPACK are really bundled with Maple in this sense, although some of the routines are apparently used in the Maple code.

All of these packages require as least linking of the appropriate library functions into a dll to be used by Maple's external calling procedure or purchasing a package (say NAG or the Intel math library MKL) with such a dll.  One can also purchase MatLab and use it externally through Maple. None of these are impossible, just many more steps and/or $ past simply running a Maple procedure.

@Alejandro Jakubi 

There certainly seems to be a sort of GUI level kernel operating. Clearly it is parsing input as it is being typed and prevents various types of editing unless one finds a workaround. On most of these issues I have found a workaround.

So I am pleased that someone has found a workaround for the printlevel bug in 14. I still have 13 also, and it works fine there. The bug is not due to using document mode. One of my students emailed me with this problem about an hour ago.

@Alejandro Jakubi 

There certainly seems to be a sort of GUI level kernel operating. Clearly it is parsing input as it is being typed and prevents various types of editing unless one finds a workaround. On most of these issues I have found a workaround.

So I am pleased that someone has found a workaround for the printlevel bug in 14. I still have 13 also, and it works fine there. The bug is not due to using document mode. One of my students emailed me with this problem about an hour ago.