W3C home > Mailing lists > Public > www-style@w3.org > April 1997

RE: Cascading and scripting (was: The concept of cascading) (fwd)

From: Chris Wilson (PSD) <cwilso@MICROSOFT.com>
Date: Wed, 30 Apr 1997 13:05:53 -0700
Message-ID: <41F7F4CE3CA2CF11BC5000805F14B2A90174B58B@RED-31-MSG.dns.microsoft.com>
To: "'Paul Prescod'" <papresco@calum.csclub.uwaterloo.ca>
Cc: megazone@livingston.com, www-style@w3.org
But the example can easily be extrapolated to be more complex - let's
say when I hover over an item, I want to change its properties in
non-deterministic ways - e.g., I want to take the current time and use
that to determine the color, or I want to change ANOTHER item's
attributes - (we do this for quick TOC-type displays - list the table of
contents on one side, and then the other side contains descriptions, all
of which are hidden - when the cursor rolls over one of the TOC items,
un-hide that set of content).

It is once again probably possible to describe this as just another
pseudoclass (to use a CSS type) - but the second example above I would
argue requires scripting-like capabilities anyway, so I fail to see the
wisdom in implementing yet another scripting language (YASL) just to get
these effects.

And yes, as someone else pointed out, we do have examples that create
and destroy content on-the-fly, rendering those documents somewhat less
easy to index - but I don't really see how that is a different problem
than CGI scripts.  Indeed, the "document" is a virtual concept - it is
fast becoming an application platform.  That's not a requirement - just
because I have executables on my hard drive does not mean I can't have
text files - but it is a possibility.

I don't think this is a necessity for the vast majority of documents -
e.g., for all the examples I've mentioned so far, it would be easy to
author the document such that it could be validated, indexed, converted,
and certainly maintained (in fact, IMO it would be hard to write the
document using the technologies we've been evangelizing WITHOUT making
that statement true).  Talk to Scott Isaacs (ScottI@microsoft.com) about
how to write ultra-simple dynamic stylesheet effects that do not affect
(or even touch) the document structure - he's written a large number of
them.

	-Chris
Chris Wilson
cwilso@microsoft.com
***

> -----Original Message-----
> From:	Paul Prescod [SMTP:papresco@csclub.uwaterloo.ca]
> Sent:	Tuesday, April 29, 1997 2:06 PM
> To:	Chris Wilson (PSD)
> Cc:	megazone@livingston.com; www-style@w3.org
> Subject:	Re: Cascading and scripting (was: The concept of
> cascading) (fwd)
> 
> > BUT, I obviously feel that exposing stylesheet functions - that is,
> > presentation attributes - to script engines through the object model
> is
> > incredibly powerful and goes a long way into turning applications
> using
> > the Web platform into truly interactive experiences.  I think it's
> silly
> > to recommend that someone write J*Script just to write basic
> stylesheets
> > - but I think it's great to be able to write
> > "ONMOUSEOVER="this.style.fontWeight='bold'" to get hover effects.
> Check
> > out the IE4 demo pages, and you'll see what I mean, if it's not
> > blindingly obvious.
> 
> The example you present is entirely resaonble, elegant and has NOTHING
> to
> do with manipulating the content at runtime (adding elements, removing
> elements, etc.) You have just created a flow object that responds to
> mouse movement.  Great! This is absolutely NOT what I am afraid of.
> What
> I am afraid of is that people want to make document elements be
> created and
> destroyed at runtime. The "document" then becomes a virtual concept
> that is
> unvalidatable, unindexable, unconvertable and, in the end,
> unmaintainable.
> 
> Note also that the example that you present does not really have to be
> 
> tied to any scripting language at all. You merely need an ONMOUSEOVER
> property and a list of properties to change when the ONMOUSEOVER
> occurs.
> You could also select another element to change through a simple
> query.
> 
>  Paul Prescod
Received on Wednesday, 30 April 1997 16:06:14 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Monday, 27 April 2009 13:53:50 GMT