W3C home > Mailing lists > Public > www-style@w3.org > February 2002

Re: CSS, css, etc. [css layout should be symmetrical]

From: Henri Sivonen <hsivonen@niksula.hut.fi>
Date: Thu, 21 Feb 2002 20:42:09 +0200
To: www-style@w3.org
Message-Id: <B627331E-26FA-11D6-9D70-003065B8CF0E@niksula.hut.fi>
On Thursday, February 21, 2002, at 06:11 , Vadim Plessky wrote:

> Obviously, Lynx doesn't support CSS.
> And some people clearly indicated that they need *simple* browser, 
> like Lynx,
> with support of key CSS features.
> If we want to replace CSS, I guess the only alternative can be (in 
> some way)
> Display Postscript (DPS).
> Obviously, you can build up translation engine which will convert 
> XML + some
> extra attributes to PostScript (DPS).

Like an XSLT transformation that outputs SVG perhaps? For mainly 
textual content, I prefer HTML + CSS with dynamic line wrapping, user 
style sheets etc. over PDF-like and paper-like online documents that 
are essentially static images with no dynamic line wrapping etc.

Obviously, "make up tags as you go" XML with no predefined semantics 
and vector graphics wouldn't mesh well with a Lynx-like simple 

> |   > I, personally, found even XHTML Basic too bloated, and would 
> prefer to
> |   > use XML when possible. (for example, on corporate Intranet, for
> |   > presentations, etc.)
> |
> |   XHTML Basic _is_ XML.
> It still have pre-defined elements.
> This should _die_,  really!

What's wrong with predefined semantics? Or rather, what's the point 
in using markup that doesn't mean anything? How would you denote a 
link? Would you use XLink (predefined semantics again) or would you 
bind the linkness to an arbitrary element using some kind of 
behavioral extension to CSS?

Henri Sivonen
Received on Thursday, 21 February 2002 13:42:47 UTC

This archive was generated by hypermail 2.3.1 : Monday, 2 May 2016 14:27:00 UTC