On 2019-11-13 3:00 AM, Asmus Freytag via Unicode wrote:
The current effort starts from an unrelated problem (Unicode not wanting to
administer emoji applications) and in my analysis, seriously puts the cart
before the horse.
But it does solve the unrelated problem.
There's nothing stopping vendors from making software which recognizes
tag character strings to reference in-line graphics. There's nothing
stopping users from employing those in-line graphics as emoji images.
It would be considered a higher level protocol which uses tag character
strings in lieu of, for example, ASCII strings like <img
src="triceratops.png">. Either way, it's rich-text expressed with
plain-text strings.
But for Unicode to provide this mechanism which "should be correctly
parsed by all conformant implementations" as well as possibly
maintaining a registry of "tag sequences known to be in use" suggests
that Unicode now considers that random images (with no symbolic meaning
other than they're pictures of something) should be exchanged as plain-text.
The QID Emoji in Unicode makes as much sense as the original emoji
inclusion. It's a natural result of the slippery slope of emoji encoding.
Emoji are open-ended but Unicode currently has barriers erected. QID
Emoji would eliminate limitations on what's supposed to be an open-ended
set. That's the problem that the current effort would resolve. In my
opinion it's better to open up a myriad of images and see which
sequences actually get used than to have vendors/enthusiasts create
images in the hope or expectation that anyone will actually use them.
