Way
back when, when plain text was very very plain, arguments
about not including text styling seemed reasonable. But with
the inclusion of numerous emoji as James mentioned, it seems
odd to be protesting a few characters that would enhance
“plain text” considerably. Plain text editors today support
bold, italic, and other styles as a fundamental requirement
for usability. More text editors support styling than
support bidi or interlinear annotation.
If
there was support for the handful of text features used by
most plain text editors (bold, italic, strikethrough,
underline, superscript, subscript, et al) (perhaps using
more generalized names such as emphasis, stress, deleted…)
then many of the redundant (bold, italic, …) characters in
Unicode would not have been needed. HTML seemed to do very
well with a very few styling elements. HTML is of course
rich text, but I am just demonstrating that a very small
number of control characters would bring plain text into the
modern state of text editing. Editors that don’t have the
capability for bolding, underlining, etc. could ignore these
controls or convert them to another convention.
As
James requested, it would also provide interoperability.
Arguments
about all of the conventions that Unicode does not support
doesn’t seem compelling to me, as it seems increasingly
random as what is accepted and what isn’t, or at least the
rationales seem inconsistent.
A
case in point is the addition of the “SS” character which
made implementation complex with little benefit.
Interlinear
annotation is perhaps another example.
I
don’t want to enter into a debate about why these deserved
inclusion. I am only saying they seem less useful than some
other cases which seem deserving.
**And
right now, Dr. Strangelove style, my right hand is
restraining my other hand from typing on the keyboard, to
avoid saying anything about emoji.**
Ken
distinguished numerous variations of stress, which of course
have their place, representations and uses. But perhaps for
plain text we only need a way to indicate “stress here”,
leave it to the text editor to have some form of rendering.
For more distinctions the user needs to use rich text.
Surely there is an 80/20 rule that motivates a solution
rather than letting the one percent prevent a capability
that 99% would enjoy.
(Yes
I mixed metaphors. I feel an Occupy Unicode movement coming
on. J )
I
don’t see how adding a few text style controls would be a
burden to most implementers. Given ideographic variation
sequences, skin tones, hair styles, and the other
requirements for proper Unicode support, arguing against a
few text styling capabilities seems very last century. (Or
at least 1990s…) And it might save having to add a few more
bold, italic, superscript, et al compatibility characters…
tex