- From: Todd Fahrner <fahrner@pobox.com>
- Date: Mon, 9 Feb 1998 16:14:14 -0800
- To: John Udall <jsu1@cornell.edu>, www-style@w3.org
John Udall wrote (3:55 PM -0500 2/9/98): " >Why bother trying to preserve structure and " >semantics in a display format? All you need is DIVs, SPANs, tables, and " >forms. And support for "atomist" CSS - nothing too relative or " >inheritance-intensive, and preferably inline. And DHTML. Right? If you " " Wrong. Or at least, not completely right. I think it's even more completely wrong than you do. From your reply and others, though, I see that my rhetoric was too elliptical. I think preserving structure and semantics all the way up to the stylesheet interpreter is critical if stylesheets are ever to be more than simple collections of formatting attributes, suitable only for a narrow range of outputs anticipated by willful designers. If documents are to achieve true portability across a continuously variable and infinitely extensible range of output media (which I take to be the Web ideal), then stylesheets must be very highly parametric. They must marry the demands of document structure/semantics with the limitations of the rendering environment (aspect, color and physical resolution, properties of available fonts, frequency response, etc.). I'm excited by XSL as a potential container for such parametrization [sp?], but frightened to see that the first implementations are geared more toward feeding legacy renderers "empty caloric" instructions, rather than upping the ante for rendering behavior per se. Sorry if that's a muddle. I wish I had more time to develop it. " In general, I think Todd's perception here is valid. In an ideal "world it " would be nice to move to just <DIV>, <SPAN>, <TABLE> and <FORM>, plus CSS. " Eventually we might get there. That's a terrible world, though. It's print before the codex: the fixed scroll. " My disagreement is that I think that the " transition will take longer. HTML 4.0 transitional might in fact be a " destination of sorts. But if it is, it might be because it provides a " transition between the old technology and the new, rather than merely a " transition between versions of HTML standards. Funny, I thought "transitional" meant "transitional to HTML 4.0 Strict". Not to XML in the general case. Perhaps HTML 4.1 should be defined as an XML DTD. Todd Fahrner mailto:fahrner@pobox.com http://www.verso.com/agitprop/ 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 Monday, 9 February 1998 19:08:07 UTC