Gabriel Sechan wrote:
From: Tracy R Reed <[EMAIL PROTECTED]>
XML and s-expressions are nearly interchangeable, very interesting! I
definitely have a new appreciation for XML as well as how poor our
chosen standard implementation of it is.
Sooo yet another reason to hate the emperor's new clothes of technology?
Yes and no.
That end tag is remarkably important. It's the difference between a
file which is completely corrupt due to a lost parenthesis and a file
which is 99% recoverable. In addition, there is a reason why a whole
chunk of programmers tend to write code like:
} // Closes if(!flgFoo)
#endif // #ifdef WINDOWS
I don't tend to do so, but some people seem to find it helpful.
In addition, XML handled the encoding problems (Unicode, UTF-8, base-64,
etc.) while s-exp sat around saying "ASCII UPPERCASE RULEZ!". If you
look at the Lisps, they *still* have this bias. We use spacing *and*
capitalization in natural language for a reason, folks.
Yes, correctly formed XML and s-exp are isomorphic. It's when things
*aren't* correctly formed that the differences appear.
XML is really good for *interchange*. It's when it gets used for more
than that that it gets in trouble. I also don't like all of the extra
crap tucked into XML (namespaces, schemas/DTD's, etc.). But, then, you
don't *have* to use it, and I don't.
Maybe it's useful for the web. The verdict is still out. The existence
of CSS seems to indicate that something is missing from XML or, at the
very least, there are other needs which are orthogonal.
When XML needs semantic meaning, life gets ugly.
-a
--
[email protected]
http://www.kernel-panic.org/cgi-bin/mailman/listinfo/kplug-lpsg
