Re: [Axiom-developer] Re: help function (Waldek Hebisch)

[Gabriel Dos Reis]

On Wed, 5 Sep 2007, Martin Rubey wrote:

| Gabriel Dos Reis <address@hidden> writes:
| 
| > (1) -> ?foo x  == x + x
| >                                                                    Type: 
Void
| > (2) -> ?foo 3
| >    There are no library operations named foo 
| 
| > (2) -> <<bar x == x + x
| >                                                                    Type: 
Void
| > (3) -> <<bar 3
| >    There are no library operations named bar 
| 
| Interesting.  I'd consider these as bugs. 

Whatever they are, I believe your earlier statement that some choice
was natural needs, at minimum, some proof.

| (1) -> ? x == x^2

Noice that your claim wasn't that the single character "?" by itself
can be a name.  You clearly stated:

  But since Mathematica does not allow either of ?, <<, ! *as first
  char of an identifier*, they do not have the same problem, so using
  ")" as separator of name spaces doesn't make any sense.  *Aldor and
  Axiom allow all three*,   so ")" is a natural choice.

Emphasis is mine.

[...]

| (7) -> _<_<foo x

The character ")" can also start an Axiom identifier if is it
escaped.  From there, I don't see how the choice of ")" is natural.
Would you mind clarifying the naturalness?

-- Gaby