Re: Rationale for the warning "Same colors for color and background-color in two contexts"?

9.10.2011 17:43, David Dorward wrote:

> On 9 Oct 2011, at 14:58, Andreas Prilop wrote:
>> On Fri, 7 Oct 2011, Yves Lafon wrote:
>>
>>> it's a "be aware that unwanted things might happen" warning
>>
>> There is not a single warning for
>
> <snip>
>
>> Why not?
>
> Presumably because the problems are either blindingly obvious with
 > the most basic testing (which the same foreground/background colour
 > issue sometimes isn't),

If there are cases where "Same colors for color and background-color in 
two contexts" is a real problem (and I have yet to see one), then I 
would expect it to be more blindingly obvious with the most basic 
testing than most issues. After all, if you set an element's background 
to white and another element's color to white, then _if_ the latter is 
rendered on top of the former _and_ it has no backgound of its own, then 
don't you think even the most basic testing will show that there's a 
problem?

> or because they are things that are hard to test for.

I guess the "Same colors for color and background-color in two contexts" 
warnings are issued mostly because it's easy test for in the validator. 
But I don't think it's a good enough reason, from the viewpoint of 
people who use the validator. Besides, programmatic testing for colors 
being very close to each other isn't dramatically more difficult than 
testing for equality of colors.

-- 
Yucca, http://www.cs.tut.fi/~jkorpela/

Received on Sunday, 9 October 2011 14:53:33 UTC