W3C home > Mailing lists > Public > www-style@w3.org > March 2003

Re: Test suite submissions

From: Ian Hickson <ian@hixie.ch>
Date: Wed, 12 Mar 2003 12:17:02 -0800 (PST)
To: Jim Dabell <jim-www-style@jimdabell.com>, Paul Arzul <patricka@mkdoc.com>
Cc: "www-style@w3.org" <www-style@w3.org>
Message-ID: <Pine.LNX.4.50.0303121154540.16401-100000@dhalsim.dreamhost.com>

On Wed, 12 Mar 2003, Jim Dabell wrote:
>
> On Tuesday 11 March 2003 10:45 pm, Ian Hickson wrote:
> >    * Making the tests simple to use (green for pass, red for fail, or
> >      other techniques mentioned in the guidelines [1]).
>
> As I understand it, around 10% of all men have some form of colour-blindness,
> most of which find it difficult to distinguish between red and green.
> Wouldn't a better policy be black/white, or number of elements?

Paul Arzul wrote a similar comment.

Without wanting to offend anyone, it would be very hard to reliably test
visual UAs with any form of visual impairment. While the specific
impairment of red/green colour blindness could maybe have been generally
catered for by relying on another colour, such as blue, instead of green,
the red/green colour codes have been used to such an extent by the CSS
testing community that most CSS testers have developed instinctive
reactions to those colours, and changing them at this point would be a
greater usability hit.

In practice, in my six years of writing CSS tests, I cannot recall
receiving a single complaint from anyone who was blocked from using CSS
tests because of red/green colour blindness.


Paul continues:
>
> it easy to overcome -- here's an atomic paragraph test example: [..]

It isn't easy to write tests that don't rely on colour. In fact it's hard
enough to come up with complex tests in the first place without having an
entire category of tests be made unacceptable.

Try to take this relatively simple test:

   http://www.hixie.ch/tests/adhoc/css/box/block/margin-collapse/001.html

...and rewrite it so that it would catch all the same bugs, without
relying on colour.

Or try writing a test that doesn't depend on colour yet would catch
Mozilla's bug on:

   http://www.hixie.ch/tests/adhoc/css/box/block/margin-collapse/008.html

This isn't a trivial problem. There are hundreds if not thousands of tests
already written that would have to be rewritten.

-- 
Ian Hickson                                      )\._.,--....,'``.    fL
"meow"                                          /,   _.. \   _\  ;`._ ,.
http://index.hixie.ch/                         `._.-(,_..'--(,_..'`-.;.'
Received on Wednesday, 12 March 2003 15:16:23 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Monday, 27 April 2009 13:54:20 GMT