I think it requires a little bit of math to decide what is going on here.
The notion of greatest common divisor combines TWO notions.
Common divisor
and
Greatest.
Since the numbers in a finite field, e.g. with modulus =some prime,
all have inverses, except for 0, they are all divisors of each
other (except 0).
Of all the divisors, which should you choose? Since the notion
of order, and hence greatest, is not usually defined for finite
fields, neither is GCD.
So it is not entirely the fault of the gcd program. Perhaps your
question doesn't make sense.
Should there be an error message?
RJF
Robert Dodier wrote:
>hi fabrizio,
>
>
>
>>As a reminder I also re-report the
>>problem with "mod".
>>
>>mod is still not defined for "1".
>>
>>
>
>i think we need a little more context here. mod has recently
>(post 5.9.2) been changed; mod was renamed to polymod
>and nummod was renamed to mod. what version are you
>running? also, the global variable modulus is nil by default,
>which probably explains the error message ("NIL is not of
>type NUMBER") i see when i launch maxima (5.9.1 or 5.9.2)
>and enter mod(1). however after assigning modulus a value,
>mod(1) is OK. so maybe you can be a little more explicit
>about the sequence of operations here. is it only mod(1)
>which is problematic? is mod(2) OK or not?
>
>it would also probably help to report bugs to the sourceforge
>bug tracker in addition to posting them to the mailing list.
>
>all the best,
>robert dodier
>