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

Re: New media type...

From: Todd Fahrner <fahrner@pobox.com>
Date: Wed, 24 Sep 1997 14:29:55 -0700
Message-Id: <v03102806b04f2ee452ea@[206.245.203.103]>
To: MegaZone <megazone@livingston.com>, www-style@w3.org
At 11:50 AM -0700 9/24/97, MegaZone wrote:
>I've been editing a book on HTML and the author includes many sections of
>the particulars of WebTV...
>
>And I got to thinking.  There are legitimate display concerns.  TVs do not
>have the resolution of PC monitors, and the color representations vary.
>
>Wouldn't it be nice to have a style sheet espressly for TV based systems?
>
>How about 'tv' as a media type alongside screen, print, aural, et al?

You're right, of course. But think about it: when do you stop? Handhelds,
cars (road-noise compensated aural and heads-up), sides of blimps, bank
teller machines - all have unique rendering requirements. And "print" -
print what? Business cards or billboards? 1200-dpi CMYK or 300-dpi laser?
Is all the work going into the printing extensions not to be reusable in a
paged mode of "screen" display?

I think it would be far better not to name devices but to describe them in
terms of general properties, both physical characteristics and typical
modes of use. Thus the range of optimizable devices is unconstrained by
committee-delimited namespace.

Some time ago, I wrote:

---------------------------------------

The currently proposed media types Print, Screen, Overhead, Braille, and
Aural are too narrowly and expediently framed, in my opinion. If they are
adopted, I doubt that anybody will get around to developing and
implementing support for other, no less meaningful media types.

Some suggested revisions, with comments:

PAGED  /* replaces PRINT */
a paged or indexed rendering, such as for print or non-scrolling screen
presentation.

CONTINUOUS  /* replaces SCREEN */
a continuous, unbroken presentation, typically for scrolling computer
screens or other stream-based medium like speech.

I think paged and continuous media are more natural and useful containers
than "print" and "screen" for the problems of rendering information into
human-readable form.


OVERHEAD
I would express this as

	@media paged & overhead

or equivalent syntax; there are also scrolling overhead presentations, like
ad[vert]s on the flanks of blimps:

	@media continuous & overhead

I have my doubts about the importance of "overhead" as a top-level (or even
secondary-level) label. It seems to me that the essential differentiator
between overhead and, say, printed presentation is that the former is
typically *social* while the latter is typically *personal.* Personal and
social presentation each imply a certain proximity to the rendering device
or surface (personal=close; social=farther), as well as unique modes of
interaction (social=more push; personal=more pull).

So cinema (and blimps) are:

	@media continuous & social

While an automated teller machine display is:

	@media paged & personal

And WebTV is:

	@media paged & social

(because scrolling sucks on TV, with a beer-proof remote no less)

Is this too academic?


BRAILLE
I would express a braille tactile feedback device as

	@media continuous & tactile

"Tactile" would always imply "personal," except when the rendering device
is a loudspeaker at a discotheque, or maybe some sort of multi-user sex toy
(same thing).

A braille printer would be

	@media paged & tactile


AURAL
I would express a normal speech synthesizer as

	@media continuous & aural

Paged aural synthesis might imply selective rendering of (SGML) containers,
rather than whole documents. This would facilitate, for instance, synthesis
of ordered list items (stop and wait for queue), or glossary terms, or a
karaoke rendering of a play.

   * * *

You get the idea.


Todd Fahrner  mailto:fahrner@pobox.com  http://www.verso.com/

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, 24 September 1997 17:19:01 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Monday, 27 April 2009 13:53:51 GMT