On 64 bit computers, you can install both 32 and 64 bit versions of Maple. And the 32-bit version still provides the Classic GUI (fortunately). So, you could try. There is even a (hackish) way to use the 64bit Linux Maple kernel with the 32bit Linux Maple Classic interface.

Have you checked whether your window manager intercepting trick works with the Classic GUI? At least, this experiment could give a further clue on whether the problem lays with java.

It happens that the procedure F, defined on the fly, has a single line that tries to select the second component of the output of a call to `evala/Divide/doit`, without checking whether it has at least two. And this output has only one, namely [false], hence the error:

> stopat(`evala/Content/doit`,27):
> radnormal(Z);
memory used=41.8MB, alloc=36.1MB, time=2.21
proc (coef, g, rofs, alginds, inds) option `Copyright (c) 1997 Waterloo Maple Inc. 
All rights reserved.`; `evala/Divide/doit`(coef,g,1,rofs,alginds,inds)[2] end proc
`evala/Content/doit`:
  27*      coefs := map(F,coefs,g,rofs,alginds,inds);
DBG> stopat(F)
[`evala/Content/doit`]
`evala/Content/doit`:
  27*      coefs := map(F,coefs,g,rofs,alginds,inds);
DBG> cont
F:
   1*  `evala/Divide/doit`(coef,g,1,rofs,alginds,inds)[2]
DBG> `evala/Divide/doit`(coef,g,1,rofs,alginds,inds)
[false]
F:
   1*  `evala/Divide/doit`(coef,g,1,rofs,alginds,inds)[2]
DBG> quit

Maple results by country depend on the subject filter. Indeed, when I set it to Programming, Canada appears first, but, e.g. when Mathematics is chosen, Chile appears on the top. With these filters, all the the search terms look on topic (no trees, leafs, etc), so that these partial results seem more reliable to me.

The same answers occur because up to Maple 13, RealDomain:-limit is just a small wrapper for top level limit:

kernelopts(version);
           Maple 13.02, IBM INTEL LINUX, Oct 12 2009, Build ID 436951
showstat(RealDomain:-limit);
RealDomain:-limit := proc()
   1   _EnvSolveOverReals := true;
   2   NumericEventHandler(('real_to_complex') = r2c_handler);
   3   clean(`assuming`([limit(args)],['real']))
end proc

While, since Maple 14 it became a grown up procedure that makes computations of its own, as Preben has described.

@pagan 

In my view, providing a good programming environment is not a high priority of Maplesoft management at present. Perhaps, persistent lobbying could help making a change :)

In this regards, I have to say that I enjoy mdcs a lot.

@pagan 

In my view, providing a good programming environment is not a high priority of Maplesoft management at present. Perhaps, persistent lobbying could help making a change :)

In this regards, I have to say that I enjoy mdcs a lot.

I would start checking xfonts here (base, 75dpi and 100dpi).

I would start checking xfonts here (base, 75dpi and 100dpi).

Do you get error messages on the console when starting Classic?

Surely it won't work if this is a 64-bit Maple installation.

TypeMK is Maple's internal form for typeset expressions, i.e. 2-D display math, in the Standard GUI. As such, it should not have a role in computations. But problems do arise e.g. when the 2-D input is processed by the 2-D parser, which is known to be quite buggy.

So, if you are using 2-D input, this is a likely cause for that output. If so, my advice is using 1-D input instead.

TypeMK is Maple's internal form for typeset expressions, i.e. 2-D display math, in the Standard GUI. As such, it should not have a role in computations. But problems do arise e.g. when the 2-D input is processed by the 2-D parser, which is known to be quite buggy.

So, if you are using 2-D input, this is a likely cause for that output. If so, my advice is using 1-D input instead.

@icegood 

This is a problem with Standard GUI "Export as". No such problem with the CLI or the Classic GUI.

@icegood 

This is a problem with Standard GUI "Export as". No such problem with the CLI or the Classic GUI.