W3C home > Mailing lists > Public > www-html@w3.org > February 1999

Re: feature for HTML

From: Tim Bagot <tsb@earth.li>
Date: Thu, 18 Feb 1999 13:58:14 +0000 (UTC)
To: HTML mailing list <www-html@w3.org>
Message-ID: <Pine.LNX.3.96.990218132146.464Q-100000@c32.keble.ox.ac.uk>
On Thu, 18 Feb 1999, David Meadows wrote:

> I am not sure what you mean by CSS providing "similar functionality through
> other mechanisms". Can you clarify? Because I can't see any way to reproduce
> frame-like functionality with CSS1 (nor CSS2, though I am not so familiar
> with that specification).

Fixed positioning allows certain parts of a document to remain static with
respect to the viewport, i.e. not to scroll with the rest of the document. 
This can be used, for instance, to provide a navigation bar, one of the
more popular uses of frames. More sophisticated interactions, though often
possible, tend to be a little impractical without frames, unless some sort
of scripting is employed.

> In fact, it is this point -- that frames are structural in nature -- that
> makes it surprising that they were not included in Strict HTML 4.0. I still
> haven't seen a satisfactory explanation for this...?

There are a couple of problems with frames. This is largely why they are,
to a certain extent, discouraged. There is, for example, no way of
encoding a frameset in a URI, except by having a separate frameset
document for all possible combinations of frames. Fragment identifiers
worsen the problem.

The CSS approach also has the advantage of degrading more gracefully. One
need simply put the static section in a DIV at the bottom of the document,
typically matched on the class attribute; a non-CSS browser will just
ignore the style information and put the section at the bottom, still
leaving it easily accessible. This, IMHO, is superior to having to use
NOFRAMES, partly because of the way NOFRAMES is often abused (or ignored),
and partly because it makes maintaining a site a little easier. 

Tim Bagot
Received on Thursday, 18 February 1999 09:06:00 UTC

This archive was generated by hypermail 2.3.1 : Wednesday, 7 January 2015 15:05:50 UTC