My concerns is about using Maple in research. For me, it would be much more productive that the vertical range were chosen by default similarly as it is done in the plot linked above, than the current behavior. 

If this capability were introduced in plot, it would be quite simple to disable it by some setting, if needed for educational purposes.

My concerns is about using Maple in research. For me, it would be much more productive that the vertical range were chosen by default similarly as it is done in the plot linked above, than the current behavior. 

If this capability were introduced in plot, it would be quite simple to disable it by some setting, if needed for educational purposes.

There has been several posts/subthreads speculating about different options of opensourcing Maple, either complete or partially, e.g.:1, 2, 3, 4.

I get also an error message:

    |\^/|     Maple 13 (IBM INTEL LINUX)
._|\|   |/|_. Copyright (c) Maplesoft, a division of Waterloo Maple Inc. 2009
 \  MAPLE  /  All rights reserved. Maple is a trademark of
 <____ ____>  Waterloo Maple Inc.
      |       Type ? for help.
> proc(x); local a; end proc;
Error, unexpected local declaration in procedure body

A quick search shows plenty of sites where different previous versions of Maple can be downloaded for free. So, I do not see how sales that are not being done could be anyhow canibalized.

And browsing through fora, what I collect is that most users of these unlicenced copies look for quite traditional stuff (like ploting functions). So, I find it quite unlikely that they would pay a costly licence for any "killer feature".

Do you think that this is a 64 bit issue? (later I will boot into Linux 32 and check).

I get an error here:

    |\^/|     Maple 13 (IBM INTEL NT)
._|\|   |/|_. Copyright (c) Maplesoft, a division of Waterloo Maple Inc. 2009
 \  MAPLE  /  All rights reserved. Maple is a trademark of
 <____ ____>  Waterloo Maple Inc.
      |       Type ? for help.
> proc(x); local a; end proc;
Error, unexpected local declaration in procedure body
> quit
memory used=0.9MB, alloc=1.1MB, time=0.06

What I can tell are facts, not speculations. For instance, I can tell that while I was ambassador of Maplesoft in Argentina, quite successful, large scale math courses were developed in a faculty of my university, based on the usage by the students of unlicenced copies of the then already old Maple V Release 2. These good results could be checked in situ during a visit by International marketing staff in 1996.

The usage of unlicenced copies was the only practical way for them to use Maple. On the one hand there was no local commercial representative of Maplesoft. On the other hand, the prices in the Northamerican market may be fit for the buying power of users there, which were several times larger than here (currently, the difference is even larger) . Moreover, local prices of licences of other software sold locally are tipically much higher than those in the Northamerica market.

So, Maple was being used anyway without Maplesoft receiving anything. Had they sold old copies at a resonable price for the local market, as they promised, they could have made some money. I could check that quite a large number of persons were interested in that possibility.

By the way, I wonder what do you mean by "Poor Dogs".

On this issue you may see e.g. the thread Maple Express and this post.

I find it a very good idea. In fact, I have posted about previous versions support and backporting.

However, I know also that it is a hopeless idea.

Precisely, this is the current behavior. And I think that it could be improved. It means, trying to identify a more useful vertical range than just fitting all the values that have been calculated.

My guess is that for piecewise smooth functions like tan, it should not be difficult to implement or computationally too costly. While, if implemented, the advantage for the user would be significant.

Precisely, this is the current behavior. And I think that it could be improved. It means, trying to identify a more useful vertical range than just fitting all the values that have been calculated.

My guess is that for piecewise smooth functions like tan, it should not be difficult to implement or computationally too costly. While, if implemented, the advantage for the user would be significant.

Find a list of available packages in ?index,package .

Find a list of available packages in ?index,package .

I have installed most of the Maple versions since Maple V Release 2. There are several reasons to for that, including:

1. Reproducibility of my own worksheets and code, as well as those of other authors. They may work different, wrong, or directly do not work at all in newer versions.

2. New versions introduce new features and solve some bugs, but they introduce some other new (regression) bugs. Comparison of code in different versions may allow to identify bugs.

3. Recent versions are much slower in startup than pre Java versions. With them, I can start Maple and perform some quick calculation in the same time that Standard GUI is only ready for input.

As everything, this is a matter of taste...