Re: [xml-dev] maps

[Amelia A Lewis <amyzing@talsever.com>]

On Sun, 2002-08-04 at 13:43, Jeff Rafter wrote:
> I think that there are root types somewhere in there... I don't know how you
> would express it without a notion for numeric values though. This sounds
> like a double edged complaint-- "They didn't give me enough" "They gave me
> too much". If they had given all of the types possible it would have just
> been more fuel for the fire.

Both are true.

Because the specification requires full support of an inconsistent
collection of datatypes, it has given nearly everyone too much.  That
is, it has given database programmers types that they can't use,
strongly-typed languages types that they can't use, cartographers types
that they can't use ... but must nonetheless support.

Because the specification prohibits the creation of alternate type
libraries, it has given everyone too little.  Not all database types are
supported, not all reasonable strongly-typed-language types are
supported, and as Simon notes, there's inadequate support for
cartography (or typography, or numerous other things).

I realize that there is a wide consensus in the XML/web community that
optional bits are bad.  However, this, it seems to me, is a
counter-case.  There is no such thing as a universal type system.

I think that XML Schema part 2 should not have been a collection of
types and facets, but a language for the creation of datatype libraries,
and ideally a mechanism for registering those datatype libraries as
well.  But I've ranted on the topic, at rather more length, elsewhere
(see xml.com, if you're interested).

Amy!
-- 
Amelia A. Lewis       amyzing@talsever.com      alicorn@mindspring.com
Money can't buy happiness, but poverty can't buy *anything*.