xml-general mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Stefano Mazzocchi <stef...@apache.org>
Subject Re: FORM is to HTML, as ? is to XML
Date Tue, 21 Dec 1999 14:58:46 GMT
COFFMAN Steven wrote:
> If the only way for a client to send information upstream to a server is via
> HTML forms, transformed from XML via XSLT, then the CGI model is still going
> to be the only game in town, regardless of whether the backend is servlet,
> or what-have-you.

Hmmm, let's talk a little about this.

Today's web technology is HTML driven: content, logic, style and
behavior mixed in a single file. Very nice for newbies (you learn HTML
and you're done), a pain for more advanced uses. (as we all know since
we're here!)

We wish to change this from a "language driven" situation to a "model
driven" one. This means that you base common foundation on the ideas and
patterns first, then on the proposed (separated!) languages that define
these patterns, and you forget about the implementatino (which comes
naturally out of this process, thru companies or open source efforts)

The XML model currently has (apologies if I'm missing something):


 XML - the syntax
 XNamespaces - the schema orthogonality model
 XSchema - the schema definition (validation)
 XInfosets + XInclude - the inclusion fragmentation model
 XLink + XPointer - the linking/anchoring model
 XPath - the item identification model


 XSLT - the transformation language


 XSLFO - the page formatting language
 SVG - the vector graphics language
 SMIL - the multimedia definition language


 X3D - the 3D definition language
 WML - wireless markup language
 VML - voice markup language
 XHTML - hypertext markup language

If you compare this with the HTML model, you see how many things HTML
simply made accessible to the user but going in the wrong direction:
filling the spec more and more with stuff.

The XML model is about context separation: each spec defines it's own
specific thing and you use it if you need it. So, you can have an XML
document with no absolute knowledge about all these things. While you
can't ingnore parts of the HTML spec and go away with it.

So, yes, I totally agree with you that we lack a language to define the
XML->Protocol mapping, something that I usually call "behavior", or
"client-side logic".

It is not clear, in my mind, if this "client-side logic" should contain
stuff like "mouseOver" or general client-side scripting or code (applets
and such).

But on the other hand, there is no reason to scream for lack of purity
if you specify this "behavior" in the final language (as in group 4):
each language may be based on a different protocol to connect back to
the server. Even for HTML, it's not mandated that you use HTTP to
implement your forms actions.

> This is disheartening because there is then no way to duplicate HTML form
> functionality in "pure" XML. 

What is "pure" XML?

> You can translate your XML query into XHTML, or
> into mixes of XML and HTML, but there's no XML for the browser to look at
> and be able to recognize that it is an opportunity to communicate back to
> the server. Am I wrong to think so?

No, you're not, just not focused. If your pure XML is something that's
well formed, well, you can't do much with that (but pretty print it and
syntax highlight it, as IE5 does).

If your pure XML is some language that contains information on how to
drive the server call-back without having indications on content, style
and blah blah, yes, the XML model lacks of something like that (maybe
WIDL? I still have to take a look at that)

On the other hand, is it something that really want to separate from a
final language?

In the future, I picture the need for something like that because FO
doesn't have it. So, to be able to use FO as a total replacer for HTML
(which is the goal), you need a way to specificy it's user driven

But for now, the use of XHTML for forms doesn't hurt me at all.

Stefano Mazzocchi      One must still have chaos in oneself to be
                          able to give birth to a dancing star.
<stefano@apache.org>                             Friedrich Nietzsche
 Come to the first official Apache Software Foundation Conference!  
------------------------- http://ApacheCon.Com ---------------------

View raw message