Re: namespace node implementation

On Thursday, Oct 23, 2003, at 15:23 Europe/Berlin, Kay, Michael wrote:

> > and the saxon experience confirms my observation.
> > [...]
> I'm sorry, but the notation you are using is not one that I am 
> familiar with.

sorry, i'd hoped it would be a fairly easy to follow. it was only to 
illustrate agreement on the minimal properties of a first class name. 
i'll explain. in general, it is the output of simple lisp introspection 
facilities. lisp being the implementation context for the xml processor 
which produced the posted examples.

[you may have observed a similar approach in the examples i posted.
> ? (describe (name $a))
the $a is from p.bother's posts, "(name $a)" is a function application. 
"(describe ...)" prints the description of that result.
> Symbol: ||::\a
which is a datum of type symbol, an interned name,
> INTERNAL in package: #<Package "">
interned in the set of names named "",
> Print name: "a"
with a name string, or "local part", of "a"
> Value: #<Unbound>
> Function: #<Unbound>
several properties (a function a data value) which are not material to 
the discussion
> Plist: (:PREFIX "")
and an additional property, the prefix.

which is also "held" in 32 bits.]

the point of which is that, given the "32-bit name" which was 
mentioned, which is close-enough to first-class, name instances can 
serve, in themselves, to represent the information produced by decoding 
and required for encoding, and are, in themselves, a sufficient basis 
for all operations on a closed model, without recourse to the 
in-scope-namespace mechanism.

i note the mention of runtime name construction, but note also that 
such operations are known to be unsave, and are unnecessary.
...