W3C home > Mailing lists > Public > www-svg@w3.org > November 2004

Re: CSS style sheets in SVG enable high and low contrast

From: Jim Ley <jim@jibbering.com>
Date: Mon, 29 Nov 2004 14:50:59 -0000
Message-ID: <013401c4d622$d75e9a70$8c909bd9@Snufkin>
To: <www-svg@w3.org>

"Jonathan Chetwynd" <j.chetwynd@btinternet.com>
> I dont agree at all, the style sheets referred to can be applied by the 
> user, they could more usefully be applied across a greater variety of 
> sites if another semantic layer were available that more fully described 
> GUIs.

You cannot change the colour of graphical elements unless you _know_ that 
the colour isn't relevant (consider the key to a simple graph, you need the 
polyline and the text both to remain the same colour, this isn't possible in 
CSS)  If we develop some description language that can define these sort of 
elements, then that would as you say be a great benefit, however it would 
not be usable with user CSS stylesheets alone.  So even with this advance 
that doesn't exist, or even being worked on, user stylesheets wouldn't help 
any.

User stylesheets as a repair for specific SVG sites, perhaps that's useful, 
but it's an incredibly expensive activity, and applies to such a small 
number of cases and is achievable by using DOM methods to inject styles.

> afaik the common user agents do support stylesheets, it is the changing of 
> stylesheets that remains problematical.

Which means they only support a subset, and especially once you move to the 
mobile focussed UA's there's nothing more than style="..." as an alternative 
to attributes (why they did that I don't know, I particularly dislike this 
deliberate partial implementation of specifications, but I can understand 
the motivation.

> Please provide any evidence that supports your statement: "the exact same 
> can be achieved in other ways"

at it's simplest, simply link to another version of the page, there's no 
reason to have all that mess - like we discussed on #svg with the 
systemLanguage detection (with Vincent), there's actually little reason to 
do all the changes in a single document, alterernate representations make a 
lot of sense, a lack of reliance on javascript is certainly one well worth 
considering.

It is also possible using SMIL animation, again which avoids script, with 
its attendant advantages, but instead requires SMIL, or if you wish you can 
use other javascript.

none of it's interesting though, it's all simple to author, and has to be 
done by the author, it can't be done by the user.  In the case of contrast 
there is research into automated methods to achieve it, however as you've 
noted they're not very good, they're also not achievable with CSS user 
stylesheets, so will need a new Access Technology (I believe some of it can 
be achieved with filters, but am waiting on hearing back from some 
accessibility people I've asked.) - if it can a bookmarklet could do it in 
ASV+IE, or MozSVG when it gets filters.

> Please note that WAI have recently also published draft documents, but 
> they dont refer to SVG.

Even I know of work on the SVG Accessibility document, it's as mature as the 
scripting one, which is currently too simplistic to really be useful.

Jim. 
Received on Monday, 29 November 2004 14:51:07 UTC

This archive was generated by hypermail 2.3.1 : Wednesday, 5 February 2014 07:14:53 UTC