[Bug 160529] Inconsistency on color schemas on each language

[Matthew Woehlke]

(re-adding kwrite-devel as much of this is still about word completion 
specifically)

Anders Lund wrote:
> On Wednesday 16 April 2008, Matthew Woehlke wrote:
>> So... which color scheme exactly are you looking at? (Or am I misreading
>> something in the above?) In Oxygen color scheme, ActiveText is bright
>> pink. I'd say that's notably different from the near-black NormalText.
> 
> I didn't check, and didn't change it in this case. I assume oxygen, the 
> activetext being magentaish supports that if I am not mistaken.
> 
> I do not say that activetext does not differ from normaltext, I say it is not 
> standing out from a normaltext surrounding, and I believe that is wrong.

Ok. I guess it doesn't bother me that way, but this sounds like an issue 
with the Oxygen scheme rather than a code issue. I would be *more* than 
happy to entertain suggestions; I simply don't have any better ideas at 
the moment.

For word completion specifically, perhaps a manual tint (with amount > 
default==0.3) for the background but with *NormalText* foreground would 
be more suitable? ...except this is probably worse with light-on-dark 
color schemes, but you should be able to handle that by choosing 
NormalText or ActiveText based on whichever has higher 
KCU::contrastRatio; that should favor ActiveText in light-on-dark 
schemes where ActiveText is lighter than NormalText.

>> Anders Lund wrote:
>>> On Tuesday 15 April 2008, Matthew Woehlke wrote:
>>>> At any rate, FooText on FooBackground (except where Foo == Normal) is
>>>> guaranteed to be the worst contrast of any combination.
>>> Meaning, they are not meant to be used togeather.
>> Not so much that, as that you're by definition ALWAYS reducing the
>> contrast between background and text. While this is likely to be true in
>> many cases for other combinations, it may not be, whereas FooText
>> against FooBackground will always have less contrast than FooText
>> against NormalBackground.
> 
> In a way, that is just too bad.

Well, it's unavoidable, unless ActiveBackground is redefined as 
something other than NormalBackground tinted with ActiveText. Which, 
honestly, I think would be confusing (not to mention making scheme 
configuration much more complicated than it already is).

>> But ideally they should be usable together. I agree that the current
>> situation is that FooBackground tends to reduce the contrast too much,
>> hence the patch I posted last night.
> 
> I understand that, and while it's working around the problem, it definately 
> limits the usefullness of activebackground, [snip]

Why ActiveBackground in particular? Do you feel there is not a problem 
with the other alternate background roles? If not, what makes 
ActiveBackground "special"?

> But where is a background that is so 
> close to the default make much sense?

I think alternate background roles are more useful when you don't 
require them to be "in your face" (I don't think they were ever really 
meant for that) or when they aren't used against NormalBackground (in 
which case the limited contrast with NormalBackground isn't noticed). 
For example, katepart's incremental search is a good use, or changing 
the background of the URL bar for secure sites, etc.

> Up to now, the word completion is a plugin (the configurability of the colors 
> would not be with the othre colors => yelling usability experts in the far 
> horizon), and the ability to color the inserted text is new to kde 4, so I 
> cared more about having it at all than making it configurable.

Again, that's fine. I'm really not against the current situation, just 
suggesting possible future improvements.

-- 
Matthew
 > pinotree uses the large trout on tsdgeos and PutHuhn :)
 > PutHuhn runs
 > tsdgeos lights a fire and eats the trout
(with apologies to Pino Toscano, PutHuhn and Albert Astals Cid, who came 
up with this entirely on their own)