[
Lists Home |
Date Index |
Thread Index
]
From: "Miles Sabin" <miles@milessabin.com>
> Rick Jelliffe wrote,
> > I think we need to draw the line between different kinds of
> > "typelessness".
> > * typelessness, at an extreme where an operation occurs without
> > any checking on some kind of memory location or a pointer.
>
> No argument here.
> > I don't really see any typelessness at work in XML: it seems to be a
> > property of languages rather than data.
>
> Now I'm confused. You started with what you claimed were different kinds
> of typelessness (for all that I mostly disagree), then showed how
> they're applied them to XML, and now you're claiming there's no
> evidence of typelessness at work in XML?
No, I list different kinds of "typelessness" and end up suggesting these are
not typelessness. I think my double-quotes were not clear enough to convey
my intent, sorry. I think we agree.
I regularly use strong static typing, dynamic typing, reflection, late-bound
evaluation in programming, and I don't see why XML isn't richer with
the full bag-of-tricks available. David and Jonathan's comments are
very interesting.
I used to work for a large US company, doing support on their LISP-based
operating system. They felt at that time (c1990) that LISP would die out, but that
when you looked under the hoods of conventional programs, you would see
it full of LISPy things: lists, tagged typing, properties, generic functions,
garbage collection, and apply (if not eval). Looking at XML, Java and C#
it seems they were pretty right back then. LISP's flexible approach to
typing didn't die: it got assimlated. Only small problems can be
well expressed using a single paradigm: that is gotcha of minimalism.
So I don't see that the issue is with types versus no types. It is more between
minimal typing and maximal typing. It is a little futile: sometimes we need
the minimum sometimes we need the maximum. I think underneath the
talk against so-called typing is a lot of dislike of WXS, for a variety
of reasons. What matters is having a plurality of technically excellent
approaches available that don't tread on each other's toes.
Cheers
Rick Jelliffe
|
