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

7.10.2011 18:57, Yves Lafon wrote:

> Yes, it's a "be aware that unwanted things might happen" warning,

Unwanted things might always happen.

> you can see in the advanced interface that there is a selection of "all",
> "normal" ,"most important" warnings.

But why does this rather theoretical warning belongs to "normal"? The 
warning about missing background when color is set (or vice versa) does 
not, even though it is more understandable and probably more useful.

> The choice might be clearer, or the warning might output its severity in
> the report...

I don't think that matters. What matters is whether the warning is 
useful to anyone. Are there any known cases? If yes, what would be the 
explanation that should be given about this rather mystic warning?

I strongly suggest removing this warning completely. As the second best 
option, moving it those shown only when "all" warnings are requested for 
_and_ reformulating it e.g. as follows:

Same colors for color and background-color in two contexts b and a
=>
The same color is set as content color for one element and as background 
color for another (elements b and a). This may sometimes cause problems 
if the elements are positioned over each other.

Yeah, that's long. But if something is worth saying, it's worth saying 
that most people have a fair chance of understanding what it really 
tries to say.

The main reason for removing questionable warnings or moving them to 
those not displayed by default is that most style sheets around have 
considerable real problems. To help people fix the real problems, we 
should avoid drawing their attention to issues that just aren't problems 
in any normal situation.

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

Received on Friday, 7 October 2011 16:53:53 UTC