[
Lists Home |
Date Index |
Thread Index
]
----- Original Message -----
From: Gavin Thomas Nicol
Sent: Saturday, January 18, 2003 12:18 AM
On Friday 17 January 2003 03:53 pm, Mike Plusch wrote:
> p.s. I don't know Adam Bosworth, but if he's arguing
> for better support for XML in programming languages,
> then I certainly think he's on the right track. : )
>Why?
Because Water is an example of an all-purpose language
that was designed for XML and Web Services.
>But in all honesty, when I went and looked at Water, I was quite
shocked. At
>least XEXPR is consistent. Water seems like a hodge-podge of
alternate
>syntaxes and special case rules aimed at specific operations, rather
than a
>consistent language. The fact that it doesn't even pretend to be XML,
but
>rather just pays it lip service, makes it even worse. I'd rather work
in VB,
>or LISP (despite the claims of "as easy as Basic" and "as powerful as
>lisp")...
The only people that could say that Water is inconsistent is Scheme
folks.
I have a lot of respect for Scheme, and Jonathan "Mr. Secure Scheme"
had a hand in the Water language.
Scheme had a purity and consistency for academics, but it would
never be accepted by general developers because it's not very
practical.
I've seen various people use Scheme for Web programming and
templating, but they generally embed another syntax (like HTML)
within Scheme. Water is more consistent in this way.
The developers that are using Water tell me that they really love
Water's consistency and uniformity -- much more so that either
the .NET or J2EE architectures.
I'm not sure what you mean when you say a "hodge-podge of
alternate syntaxes". XML actually has at least 9 different
syntax rules used for various values within an XML document.
Parsers are not easy to create for XML 1.0.
<!ENTITY >
<!-- --> (comments)
<?xml ....?> (first line)
&foo; (entities)
[ ] for grouping ENTITY tags
space delimited values in !ENTITY
foo:bar (namespaces)
<