W3C home > Mailing lists > Public > www-style@w3.org > September 1998

Re: Properties applicable to root?

From: L. David Baron <dbaron@fas.harvard.edu>
Date: Wed, 30 Sep 1998 18:00:28 -0400 (EDT)
Message-Id: <199809302200.SAA07748@login6.fas.harvard.edu>
To: tantek@cs.stanford.edu, todd@verso.com, www-style@w3.org
>From: Tantek Celik <tantek@cs.stanford.edu>
>Seems reasonable to me.  Perhaps we need a new pseudo-class, :root, so that
>a style sheet does not have to depend on a particularly named tag being the
>"root" element.

I assume this pseudo-class would be different from the current ones in
that it can only be a selector in itself, i.e., HTML:root is invalid.
(Otherwise it could disagree with the document.) What about the fact
that HTML is defined in the HTML specs as the root of the document?  Or
does this relate mainly to XML?  It seems like something other than a
pseudo-class would be applicable here.

>Why not have the width of the root element determine the width of the
>viewport?  I think that ability would be useful to authors attempting to
>fine tune the experience of their pages.  (Imagine enabling a 'kiosk' sort
>of thing with width:100% and height:100% on the root element)
>You could also use the 'overflow' property on the root element to determine
>whether or not scrollbars show up on the viewport.

>Perhaps the border properties could apply to the border drawn on the
>viewport itself, for UAs and platforms that would support that.  Again, more
>author control over the presentation.

I don't think it should be part of CSS to move the user's browser
around the screen or change the look of the browser.  A loss of
consistency of the browser interface would make many novice users give
up.  I get mad when web sites resize my browser or create new windows,
and I think most other users do as well.  If I am working with a web
browser on one side of my screen, and another program on the other
side, I want to keep it that way, and I don't think there should be
attributes to mess with that.  I think the concept of a web page should
be as a document viewed within the browser, and what is outside the
browser should be irrelevant.

>I would loosen that to say it could take "none" or one of the "block-level"
>display types.

This doesn't seem to make sense, since the other block-level display types
involve lists and tables, and a BODY element is not valid within lists or
tables (I don't think ... it's not really a question that has been asked
before).  I think we should be careful not to propose something that would
make the page incomprehensible to a browser without CSS or a browser that
overrode the display: properties.  (This is a general problem with using
the list and table display: properties outside of a UA style sheet.)

>You could interpret float within the context of the screen/viewing area.
>For example:
>HTML { float:left }
>could float the window to the left edge of the screen, similarly with
>float:right.  See below for why this would be useful.

See above.

>You could also allow positioning of the viewport within the screen:
>HTML { position: absolute; top: 20px; left: 20px;}

See above.

>Another example: say you have two documents you want to present to the user
>side-by-side on their screen, each taking 50% of the width of the screen. 
>You could use the following two stylesheets:
>
>// stylesheet 1
>HTML { float: left; width: 50%; overflow:scroll }
>
>// stylesheet 2
>HTML { float: right; width: 50%; overflow:scroll }

This should be done with frames, not with moving browser windows around.
Users have different size screens and they should be able to look at the
documents as they want.

OK... time to shut up and get back to my homework.

David Baron

--------------------------------------------------------------------
L. David Baron         | Freshman, Harvard
dbaron@fas.harvard.edu | < http://www.fas.harvard.edu/~dbaron/ >
Webmaster, International Weather Satellite Imagery Center, etc.
--------------------------------------------------------------------
Received on Wednesday, 30 September 1998 18:00:30 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Monday, 27 April 2009 13:53:56 GMT