SmallNic: A simple result

Stephane Ducasse ducasse at
Sat Feb 15 19:50:54 UTC 2003

On Saturday, February 15, 2003, at 08:25 PM, Ned Konz wrote:

> On Saturday 15 February 2003 10:27 am, Bill Spight wrote:
>> Documentation that, it seems
>> to me, is quite important, but may be hard to come by, is *why* the
>> button (or, in my brief experience, *any* Morph), loses its own
>> color when added to a window.

You are right I should I added in the comment of the test that this was 
system that was doing that for consistency look. (It cost recently
3 hours of frustration, luckily ned gave me the answer).
My  point was just to show to hannes that tests can be used as snippet 
some scenario.

May be we could extract information from them to produce better 

>>  Especially in view of the fact that
>> color constancy is one of those things we learn early in life (by
>> age 4). Color constancy is so strong that it will override the
>> physical frequency of reflected light. Once we know the color of
>> something, we tend to see it that way (after age 4) even if the
>> light coming from it is a different hue. If it is not an accident
>> that Morphs violate color constancy, there should be a good reason
>> for it. What is it?
> I believe this is so that the user can simply change the color scheme
> of specific windows. When Morphs become a part of a system window,
> they get re-colored to match it.
> If specific colors were chosen for the submorphs, changing the window
> color wouldn't have the desired effect.
> In fact, if you turn on "hidden scroll bars" and turn off "alternative
> window look" and open a browser, you'll be able to see the effect of
> having both consistently colored morphs and not: the category list
> will have a colored scroll bar, while new scroll bars will be grey.
> -- 
> Ned Konz
Prof. Dr. Stéphane DUCASSE (ducasse at
  "if you knew today was your last day on earth, what would you do
  different? ... especially if, by doing something different, today
  might not be your last day on earth" Calvin&Hobbes

More information about the Squeak-dev mailing list