Re: Namespace vs. Scope

Jonathan Feinberg wrote:
> My students caught me off guard when they pointed out that sometimes I use 
> the word "scope", as when I'm discussing my(), and that sometimes I use the 
> word "namespace", as when I'm discussing packages.  

Talk about global variables: they are like files in a hierarchical filesystem,
in that their names, if fully qualified, have "paths"; and also analogously,
they're accessible from anywhere, at any time (as files are).
Globals (sometimes aka "package" variables) are truly global; namespaces
only exist to bring some organization.  Imagine if all your files had to
sit in a single directory...

"Scope", in non-programming contexts, has to do with visibility; and
that's one of the things that distinguishes lexical variables.  Their
visibility -- and hence accessibility -- is determined by the lexical
structure of the program.  (Of course, unlike static langauges like C,
this is not strictly true, because the data structure of a lexical can
survive beyond its "native" scope, since perl uses reference-counting.
But that's a slightly more advanced concept.)

You might want to talk about how perl variable names are just pointers, or
references, to the real variable thingy, unlike C where a name is an alias
for the memory location of the variable.

-- 
John Porter

So take a pointed stick and touch Piggy's eyes
He's gonna turn and leave you a big surprise