"sum" quotes its arguments -- what's up with that?

C Y wrote:

>--- Richard Fateman <fateman@cs.berkeley.edu> wrote:
>
>  
>
>>I think that many of the problems, with integrate, sum etc.
>>  would disappear if one insisted
>>on a notation that was better.
>>
>>For example, borrowing from lisp or scheme...
>>
>>instead of sum(i,i,1,3)
>>
>>we could have
>>    sum ( lambda([i],i), 1..3).
>>
>>Thus it is clear where i is bound.
>>    
>>
>
>Except then we will have a lot of new users who won't be able to
>understand why they have to type lambda while doing a summation.
>

In that case, you could tell people to write....

sum(f, i,1,3)    and we would automatically translate it to

sum(lambda([i],f),1..3).

That's essentially what the sum program does now, except that if   f is 
a variable  and not the
index of summation, it is evaluated once.  I think that is what it does, 
anyway.

Is it right all the time? no.
Is there another way to do it that is always right?  I don't know of 
one, though the
lambda idea helps.


>
>I think rather than trying to come up with a notational workaround we
>should put on our radar screen the sorting out of scope issues in
>Maxima.  That seems like a perfect issue to tackle in 5.9.3
>development, since it will (hopefully) kill a lot of the reported bugs
>and unexpected behavior in things like this.  
>
>CY
>
>
>		
>__________________________________ 
>Do you Yahoo!? 
>Yahoo! Mail - now with 250MB free storage. Learn more.
>http://info.mail.yahoo.com/mail_250
>  
>