rtests in Maxima problem



Thanks Richard,

One reason for factoring out the Maxima code into a separate package is to 
enable comment and possible collaboration.  I appreciate that simp:false 
is not the best approach here.  It would be an interesting project to put 
these functions on a sound CS basis.

Chris


On Thu, 2 Dec 2010, Richard Fateman wrote:

> On 12/2/2010 12:52 AM, Chris Sangwin wrote:...
>
> ...
>> 
>> The library I have written already does these sorts of things.  I know how 
>> difficult Maxima can be with simp:false.  I've battled against it quietly 
>> for the last 5 years!  And it now mostly works very well.  1000s of 
>> students are using these functions everyday and having useful feedback 
>> provided to their online quizzes.  See
>> http://web.mat.bham.ac.uk/C.J.Sangwin/Publications/2010-3-1-STACK.pdf
>
> I'm impressed by this -- it is far more sophisticated than I thought!  (Some 
> examples' English is kind of stilted. Finnish-English translation?).
>
> The major difficulty appears to be getting various systems to work together, 
> and to get material from other
> sources (e.g. MapleTA) to work with STACK, too.
>
> Still, running Maxima with the simplifier off, and then perhaps turning 
> it on in bursts, does not seem to match what you really need. Given 
> that, it is interesting that you were able to do so much.
>
> It is, I think, important to realize that one (perhaps the only?) 
> plausible "killer app" for computer algebra systems is "math education". 
> There are far more calculus or pre-calculus students than researchers 
> using advanced math.  And at least some of those researchers combine 
> unfortunate traits: obstinate+unwilling to read documentation + 
> unwilling to pay [ to support further development or commercialization 
> etc.]
>
> RJF
>
> PS, I remember seeing an interactive system to teach Lisp.  I don't know if 
> it ever worked for students,
> but I found it irritating because the solutions that I offered to the 
> questions were graded wrong
> because I wrote  more advanced solutions than the grading program 
> anticipated.
>
>
>