[XSL-LIST Mailing List Archive Home] [By Thread] [By Date]

The original purpose of XSL...


Subject: The original purpose of XSL...
From: Guy_Murphy@xxxxxxxxxx
Date: Mon, 23 Nov 1998 11:49:14 +0000

Hi.

Being new to the list, there is a level discourse occurring on the subject
of the two heads of XSL, namely transformation, and formatting, that I
previously wasn't aware was occurring. I feel that maybe this discourse is
moving into considering areas that aren't really within the remit of the
original design goals of XSL, or indeed XML itself.

I'll make my bias clear at this point... I'm a Web designer, and I feel
that XML/XSL where created for me. Not the printer, not the academic, me.
An inherently selfish stance I'll confess, but one which I think W3C has
lead me toward.

If we look at the design goals of XML I think it reinforces my stance,
amongst suggestions that it should be quick, easy, and human readable, we
see at the top of the list...

1. XML shall be straightforwardly usable over the Internet.

And sure enough this is where one would hope the greatest uptake of XML
will be, over the Net, delivered to the browser, as a replacement for HTML
when HTML is buckling under the strain of large Web based application, or
where large archives of document make XML the obvious choice etc., etc.

If there is no uptake of XML as a data delivery format on the Net, I think
it can be regarded as having failed it's purpose. If the print typesetters
of the world where
to adopt it, that would be great, but it will still have failed it's
initial purpose.

I think that this is of utmost importance for XSL, because I believe that
XSL inherits most of the design goals from XML. The primary uptake of
XSL has to be in the production of renderable data within the Web browser.
Anything else that it accomplishes is great, but if it fails in this
purpose, people are going to feel more than a little let down.

So while it is obviously crucial to consider other languages such as TeX,
and DSSSL, for comparative development, XSL isn't one of these languages,
and isn't meant to be.

I feel that XSL has to encompass both transformation and formatting because
 I am not convinced that the Web design community will accept two languages
in order to render XML in the browser in a flexible manner. Yes we have
CSS, so could use XML with CSS, but CSS is not an XML dialect, so hinders
the working of the two. XSL should be in my mind the one stop solution for
both transformation and formatting.

For large scale Web sites/applications, I would certainly like to see the
death of HTML. It's a bastard language, literally, ill suited to other than
 simple applications.
I think that with refinement of the current implementations, XML/XSL can do
 this. I'm not convinced that XSL split in two will facilitate this.

To sum up my position, if XML/XSL allows complicated typesetting, cool. If
it doesn't I really don't care, I neither I would hazard will the countless
 millions
of Web users that ultimately going to be it's primary users.

Of course this is only the biased opinion of one Web designer :)

Oh, and as an aside, I've uploaded an experiment in XML/XSL as implemented
in IE5b2 of the Complete Works of Shakespeare at
http://www.guy-murphy.easynet.co.uk/shakespeare/

Any comments are welcome.

Cheers
     Guy Murphy
     guy_murphy@xxxxxxxxxx



 XSL-List info and archive:  http://www.mulberrytech.com/xsl/xsl-list



Current Thread
Keywords