W3C home > Mailing lists > Public > w3c-wai-ig@w3.org > July to September 2002

Re: Preloading: CSS and attribute.

From: C. Bottelier <c.bottelier@iradis.org>
Date: Sat, 28 Sep 2002 23:23:50 +0200
Message-Id: <5.1.0.14.0.20020928231110.00a8a4c0@mail2.iradis.org>
To: "Micho" <MichoKest@terra.es>
Cc: www-html@w3.org;, www-style@w3.org;, w3c-wai-ig@w3.org

At 10:56 PM 9/28/02, Micho wrote:
>So, the thread went more or less this way:
>
>+ <section> preloading control was proposed to have some kind of control 
>of how data is displayed while it's loading: show nothing until complete 
>or show while loading .
>+ This proposal was declared merely presentational, but anyway other 
>people fund it somehow useful for better designing and to help the 
>clearness of xthml files, by standarizing preload (section visibility 
>control) and removing unportable scripts.
>+ The first proposed way (Micho) was to use an attribute for the section tag.
>+ Tantek mentioned an alternative way of achieving this by the use of CSS.
>
>So, by analyzing the possible uses of this functionality, I've decided to 
>expand it a little more:
>
>a) Instead of talking of preloading, we should talk about visibility 
>control: visible while loading, visible when loaded or not visible.
>b) Visibility control should be avaible for section and layer tags.

It would probably be better defined as: for all
replaced elements, and all block level elements,
which contain one or more replaced elements.

>c) Implementation methods could be CSS through classes or XHTML attributes.
>d) Althought CSS would be the most well structured method, the attribute 
>way would undoubtely be also handy and easier to remember for coding 'by 
>heart' (this is a quite subjective note).

When imnplemented as an CSS property it could be forced
to be ignored. This would enable it for special users need,
devices, or other software, which use an user agent (that
does or does not implement the property) to get their data.

When implemented as an (X)HTML attribute, there would be
no way to disable it's effect for an user agent that implements the 
attribute, and by this can cause accessibility issues.

>So, things seem pretty clear, why not reach a final point in this thread?

The proposed CSS property of Tantek would seem to
be the thing, now only the *complete* definition
for it and agreement of *all* parties.

Christian Bottelier
Received on Saturday, 28 September 2002 17:16:55 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Tuesday, 19 July 2011 18:14:06 GMT