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

Re: indents

From: Todd Fahrner <todd@verso.com>
Date: Wed, 16 Apr 1997 15:21:08 -0700 (PDT)
Message-Id: <v03102806af7a8d478257@[206.245.203.103]>
To: www-style@w3.org
[apologies if this message arrives more than once - having mail problems]

At 4:00 PM -0700 4/15/97, David Siegel wrote:

> Ahah. So maybe it happens that first paragraphs should have a structural
> mark-up tag in HTML, one that is STANDARD?

We don't need more HTML tags. For "throw-away" material, <p class=initial>
will do today, and it's standard. For material that might be re-used (or
re-styled), XML is the way to go. You can invent tags and declare their
significance (or not) in a DTD. CSS or DSSSL can format either.

Extra added bonus (void where prohibited): you can use
<initialpara>stuff</initialpara> RIGHT NOW. CSS browsers (IE & NS) will let
you style it as you wish, sorta (beta). Downlevel browsers (NS & IE) will
render it inline. This is not an endorsement of this approach, but it does
have implications for ease-of-transition to XML.

<HINT>I still think a P:INITIAL pseudoclass in CSS would be a harmless
enhancement to 4.0-implementations, useful for rendering legacy HTML
attractively.</HINT> I can't think of a downside, unless it would be
stalling the inevitable embrace of DSSSL. :^)

But seriously, given that CSS is clearly the next wave of client-side
style, couldn't DSSSL be deployed effectively on servers, or at "publish"
time, to generate HTML+CSS? If this is the case, then the highest priority
for CSS is not to fill it out with problem-solving ability, but to make it
a suitably rich vocabulary for DSSSL-generated formatting. Or vice-versa:
make sure DSSSL is capable of specifying the full range of formatting that
CSS can.

Jon's examples of DSSSL include references to picas and points. Does DSSSL
know about pixels? About ex? Percentages or rendering area (width and
height)? How does it compare with CSS-P in terms of granularity of control?
Is there a DSSSL analogue to cascading, or to multiple choices for, say,
fonts? Finally, can the dynamic formatting behaviors of CSS+scripting
languages (like alpha-channel manipulation) be modelled and generated in
DSSSL? How about rotated text? Fit to curve?

In short, is there a transition path from CSS to DSSSL, or is all lost
already? :^)

________________________________________
Todd Fahrner

The printed page transcends space and time. The printed page, the
infinitude of books, must be transcended. THE ELECTRO-LIBRARY.

--El Lissitzky, 1923
Received on Wednesday, 16 April 1997 17:10:11 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Monday, 27 April 2009 13:53:49 GMT