Re: [stack] disallowing recursive definitions

On 2/4/08 3:15 AM, "John Cowan" <[email protected]> wrote:

> Manfred Von Thun scripsit:
> 
>> > My analogy was with Unix pipes. Here each command (a molecule in my
>> > meaning) might consist of a primary command (an atom in my meaning))
>> > together with any (often optional) additional atoms. Some of these
>> > are additional parameters, some of these are command modifiers. None
>> > of these is a command by itself. The Unix pipe is a concatenation of
>> > such molecules.
> 
> But surely a molecule like "ls -l -C 2" is just syntactic sugar for
> ordinary Joy "[-l -c 2] ls", where the list is treated by the "ls"
> word as a list, not a quotation?

That is one possible notation. But if a command frequently uses no
modifiers,
then that notation  would force users to supply an empty list of modifiers,
as ³[] ls² for your example, in the most frequent uses.

My list of candidate notations, using your example of ³ls², and their
drawbacks:

1. [-l ­c 2] ls  # forces empty list of modifiers if none are wanted

2. ³la² ls  # ditto, only useful for plain flags, no good for other
parameters

3. foo | ls ­l ­c 2 | bar # unix style pipe symbol, painful if modifiers are
rare

4. (ls ­l ­c 2) # Lisp notation optional modifier list

5. ls(-l ­c 2) #  functional notation optional modifier list, currently my
favourite 

I must confess that my own list of possible modifiers for concatenative
languages
is miniscule compared with the multitude of modifiers in unix. But I am
hoping that
that is mostly a failure of imagination on my part.

If anything like this is introduced, then a decision has to be made whether
modifiers
are just for the primitives or also for user defined molecules.

  - Manfred



[Non-text portions of this message have been removed]