W3C home > Mailing lists > Public > www-style@w3.org > August 2002

Re: What are Semantics?

From: Coises <Randy@Coises.com>
Date: Thu, 22 Aug 2002 04:38:08 -0700
To: www-style@w3.org
Message-ID: <irf9mu4evjl0r6v8suqe965mpjj8q7nrug@4ax.com>

[Mon, 19 Aug 2002 15:56:41 EDT] Svgdeveloper@aol.com:
>It seems to me that we need to be clearer about terminology since two, 
>seemingly intelligent, individuals interpret the semantic richness and 
>poverty of XML in two diametrically opposed ways.

Comparing XML to (for example) HTML in "semantic richness" is like
comparing the proverbial apples and oranges.  Despite its name, XML isn't
really a language: it's a *syntax* over which languages may be defined.

A markup language used in an XML document can be specified as a namespace.

When an XML document is viewed as *only* an XML document, it contains
virtually no machine-processable information, aside from the structure of
the document tree.  On the other hand, XML can be used to construct
documents of practically any degree of "semantic richness" --- but it's
namespace(s), not XML, that describe how that meaning is represented.

One might reasonably compare the semantic richness of HTML (or whatever)
to that of a particular XML *namespace* --- not of XML itself.


It's important to remember (AI SF notwithstanding) that only *people* can
work with meaning.  Using a style sheet, a browser can translate a
document's markup from the "semantic space" of an XML namespace to the
semantic space of formatted visual presentation.  But a *person* who
understands the namespace has to write that style sheet!

***  Any time a computer appears to do something based on "semantics,"
***  it's an illusion engineered by human beings who understand how meaning
***  is captured in the relevant representational spaces (languages).


This should make apparent the problem with so-called "generic" XML.
Indeed, one may represent a great degree of meaning in such markup; but who
else will know what it means?  You're writing in a private language.

If the XML is designed only for a program that you're writing as well, then
that's no problem.

If the XML is designed to represent a generally-available document, though,
there is a problem.  You can write a style sheet, and supply it with the
document, to describe how to present it in a browser.  Will you know how to
present it in an aural browser, or in braille?  Will you have thought of
how it should be translated on a very small screen, or a very large one, or
a printer, or for someone with limited vision?  A person may be able to
"reverse-engineer" your private language and figure out how to construct
a style sheet for his or her own needs --- if he or she has the time, and
a good knowledge of XML and CSS.

How do you react if you come upon a web page that interacts with your
browser so badly that your only hope to make sense of it is to sift through
the source?  (I'm "into" this stuff, and what I usually do is curse the web
designers for being such idiots and then move on to a different site.)
-- 
Randall Joseph Fellmy aka Randy@Coises.com
Received on Thursday, 22 August 2002 07:38:40 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Monday, 27 April 2009 13:54:15 GMT