Ain't it the truth!
But it is also practical because often you do not
remember what face is which color.
For the purists, agreed, this is non-textbook usage
because the thought is of course to set those faces to
something sensible (sensible colors) *once*, and from
that point use the faces semantically, to express
purpose - the assumption being their color values make
sense. However, it has happened to me many times that
I do *not* use them that way but instead use them as
mere (?) placeholders for face colors, and then it is
very useful to just type the face name to see the
color it represents.
Besides, in principle it doesn't conflicts supposedly
(?) proper usage to have them in cool colors, only
then that is more a cool stunt rather than an
advantage in productivity as well.
Holly would if she could. If you have a non-JavaScript
version of that I'll read it.
> If you tell me the underlying machinery, I'll point
> them to it. [Or of course you can join that
> conversation]
You ask me to tell you so THEY will understand? :)
Well, in the smallest possible picture, it is very
simple. When the face names themselves are typed in
a buffer, they appear in the same color as they would
give the entities that are associated with them.
This makes it is easier to set up such associations
because you can see on the face name what the result
will be.
What this illustrates on a somewhat bigger scale is
that Elisp is used to improve the tool which you use
to write Elisp.
On the biggest scale it is the computer exponential
success story with computers and programmers taking
turns making each other better. That is in principle
equally true with bulky compiled languages (like C),
however with Lisp (and not just this particular
example) you can feel it instantly. It is a mighty
realization, even for a might programmer.
Did that make sense? Perhaps only to people who
already understand it. Remember the words of
Saint Morpheus: "Unfortunately, no one can be told
what the Matrix is. You have to see it for yourself."