On Tue, Dec 05, 2006 at 09:29:06AM -0800, William Stein wrote:
> You know, honestly, the problem of how to express do Calculus with
> a computer algebra system is not exactly a new one.  It's been
> to some degree completely and totally solved by Mathematica.
> Maybe the real discussion we should be having is how can we
> make a basic interface to SAGE for doing calculus that is very similar
> to mathematica's?   Only if there is a strong technical reason why
> something is not possible in Python, should we even be having these
> discussions.   Basically, before we continue going along the route
> suggested by you and Robert, I would like a very good justification
> for why Mathematica/Maple got it wrong (despite millions of users
> and two decades of polishing and work).  Otherwise, I don't understand
> why we don't just do something similar to Mathematica.  This reduces
> the learning curve for some people a lot, and means we spend time
> solving problems instead of coming up with a new design that is probably
> wrong anyways, then spending a lot of time testing it, only to find
> that it really doesn't work for various reasons.

William, thanks for your kind hint for me to be quiet a bit.  I believe you are 
right that I should be so.  After this e-mail and a few additions to the wiki, 
I 
will be quiet on this topic.

However, I don't agree that the suggestion of using variables from the 
polynomial ring is doing what mathematica does.  Some Examples:

Exhibit 1: My understanding is that you want f(1) to be 1 in sage, but 
mathematica is not so.
In[1]:= f=x 
Out[1]= x 
In[2]:= f(1) 
Out[2]= x

Exhibit 2:  Mathematica has a dummy variable.
In[12]:= f[x_]:=Sin[x]
In[13]:= f'[x]
Out[13]= Cos[x]
In[14]:= f'[y]
Out[14]= Cos[y]

Exhibit 3: Part of the expected second variable is the name of the independent 
variable.
In[6]:= Plot[Sin[x]]
Plot::argmu: Plot called with 1 argument; 2 or more arguments are expected.

I apologize if I've made too much noise about all this, but I think that a 
clear 
concept of a dummy variable needs to be introduced.  I have thought of no easy 
way to do that with-out a rather painful bit of preparsing.  I definitely agree 
that the preparser should be avoided if possible.  Currently, I see no evidence 
that we can have a real idea of dummy variables with-out pre-parser work.

--
Joel

--~--~---------~--~----~------------~-------~--~----~
To post to this group, send email to [email protected]
To unsubscribe from this group, send email to [EMAIL PROTECTED]
For more options, visit this group at http://groups.google.com/group/sage-devel
URLs: http://sage.scipy.org/sage/ and http://modular.math.washington.edu/sage/
-~----------~----~----~----~------~----~------~--~---

Reply via email to