Re: "modern" colors Re: Changes for emacs 28

[Ricardo Wurmus]

Ergus <spacibba@aol.com> writes:

>>New users don't have to understand it from the start though, it is
>>something one can come to understand with using Emacs.  If you hover
>>with the mouse over each item, it will describe what each thing means,
>>and you can change each thing accordingly.
>>
> New users are used to know if the document has changes at least. And in
> the applications they use: filename* by default.

We have buffers that end on “*”, such as “*scratch*”, “*shell*” (M-x
shell), “*mu4e-view*”, etc.  Emacs can display graphics and/or Unicode
characters; it doesn’t need to follow a convention here other than to
provide *some* indication of a changed buffer.

The modeline already separately indicates a changed buffer with “*” (and
another “*” to indicate that the buffer is writable).  Modeline themes
such as the excellent smart-mode-line display a coloured Unicode
character instead (a vertically centered cross), which works just as
well.

>>   3) Colors: People prefer higher contrast in general 4 example: in my
>>   system when the region es enabled the default gray color is so light
>>   that I can't see it. Same applies to icon that when enabled or disabled
>>   the difference sometimes is minimal.
>>
>>Can you provide research on that people do actually prefer higher
>>contrast in general?  Your example doesn't really follow from the
>>first claim -- since that is your specific preference, not everyone
>>elses preference.
>>
> Lock back in this same thread there was a long discussion about
> that. The supporters of light colors brought some articles about
> astigmatism and so on, while the others bring different ones.
>
> Again look around and compare what you see.

None of the articles AFAIR talked about *high* contrast; many user
interface guidelines and accessibility recommendations do recommend
sufficient contrast (sometimes quantified as a minimum contrast between
two colours), but this doesn’t mean that higher contrast is always
better than lower contrast.

The Solarized colour palette, for example, aims to provide *equal*
contrast, but is agnostic on exactly how high the contrast threshold
should be (in some implementations of the theme the contrast threshold
can trivially be adjusted).

Back to the original claim: I cannot reproduce the problem in “emacs
-Q”; when I activate the region the background and foreground colours of
the text in the region are swapped, providing the same contrast as when
the region is not active.

-- 
Ricardo