[
Lists Home |
Date Index |
Thread Index
]
- From: Richard Lanyon <rgl@decisionsoft.com>
- To: xml-dev@lists.xml.org
- Date: Thu, 23 Nov 2000 11:47:32 +0000 (GMT)
On Wed, 22 Nov 2000, Bullard, Claude L (Len) wrote:
> From: Roger L. Costello [mailto:costello@mitre.org]
> >[2] A big question ... what is "semantics"?
> It is a point of view. It is organized by contract or
> by behavior. Something means what you say it means.
Absolutely. I think Wittgenstein said something along the lines of
"the meaning [semantics] of a word is its use in language".
Now, schemas constrain the syntax of a document, and this will have a
knock-on effect on how a particular document/element/whatever can be
used (there wouldn't be much point otherwise). However, there are some
sorts of semantics that can't be so easily constrained this way.
Let's use a concrete example - the familiar Address element. By
having an Address element/type, you constrain its contents to being a
few lines long and having an alphanumeric code at the end. This makes
it quite good for use in storing addresses and quite difficult for
other information. So, just by restricting the syntax you have
made the element particularly relevant for some uses, and thus given
it some "meaning" or "semantics".
But the syntactic restrictions have no effect on whether this is a
BillTo or a ShipTo address - there's nothing XML Schemas can do to
stop you putting BillTo information in a ShipTo element. This latter
level of "meaning" will, in the XML world, still have to be handled by
contract.
--
Richard Lanyon (Software Engineer) | "The medium is the message"
XML Script development, | - Marshall McLuhan
DecisionSoft Ltd. |
|