bug#41544: 26.3; Possible incorrect results from color-distance

[Mattias Engdegård]

1 juni 2020 kl. 18.32 skrev Eli Zaretskii <eliz@gnu.org>:

> Please remind me why do we want to make such deep and wide changes in
> our color system?  Is that just because we have a minor bug in
> color-distance?  Wouldn't it be better to just fix that one bug?

These changes are neither deep nor wide. Perhaps the confusion arose from 
remeding two almost entirely independent issues in the same discussion: the 
flaws in color-distance, taken care of by the first patch, and the contrasting 
colour computation, which was the topic of the message and patch you replied to.

Sorry about conflating the two -- I should have opened a separate bug for the 
colour contrast mess. Their only point of intersection was, in hindsight, 
rather coincidental (css-mode).

> I'm worried because the way our automatic color mapping on
> color-challenged TTYs works took some time to get right, and it worked
> well for years.  If we want to risk rocking that particular boat som
> profoundly, we had better had a very good reason, preferably several
> good reasons.

Not quite sure what you are talking about here. You previously asked about the 
exact value of TTY_SAME_COLOR_THRESHOLD. Were you unsatisfied with my answer, 
and if so, in what respect?