Observations and rants

From: Mark E. Shoulson via Unicode <unicode_at_unicode.org>
Date: Wed, 17 Jan 2018 21:30:57 -0500
I've been keeping up with the Document Register and following some of the discussions therein, especially regarding Emoji, and I remain surprised with some of the bizarre suggestions.  Not random people suggesting weird emoji, but things from the ESC and UTC!

A lot can be summed up in the simple question, "Why doesn't anyone listen to Charlotte Buff?"  She has written, from what I have seen, pretty cogently and clearly on various minefields that emoji decisions are likely to be treading, and has been quite the opposite of "entitled" or demanding of strange special characters.  For the most part, she has warned *against* encoding too much, lest covering some special cases open up demand for covering all of them.  What is the UTC thinking, that hair variation can be reasonably covered, in initial cases, by RED/CURLY/NONE/WHITE?  Are these supposed to be orthogonal?  In which case, the instant question is how to handle hair that is red and curly.  If they can be combined, how do we handle CURLY+NONE?  Why are the more common colors (brown, black, blond) left for future study?

Regarding https://www.unicode.org/L2/L2018/18027-wg2-fdbk-response.pdf, I also don't really see what the point of SOFTBALL is as distinct from BASEBALL, especially since softballs I've seen are often the same color as baseballs, and can be distinguished mainly by size—and neither size _nor color_ is something reliably encoded in a Unicode character (characters with colors in their names notwithstanding.  Glyphs are canonically foreground-vs-background markings, from what I can tell.)  A similar issue of color might apply to NAZAR, which, without blue coloring, looks mostly like some form of target.

MAGNET kind of has to be horseshoe-shaped.  Otherwise it's just a cylinder.  The horseshoe shape is indeed already obsolete, as it was used mainly to keep old alnico magnets from demagnetizing themselves, and modern magnets (especially NdFeB magnets) have no need of such gentling.  But cartoons for decades represented magnets with the distinctive horseshoe shape (with the ends marked off, so it doesn't look like an actual shoe-for-a-horse), and any kid in the appropriate culture would instantly recognize such a shape as a magnet.  (Cultural specificity of emoji is a longer rant I may inflict on this list.  Briefly, nothing about emoji is "culturally neutral" and it's ridiculous to make them so.)  Even though technology changes, our symbols often remain.  Telephones are no longer shaped like U+1F4DE TELEPHONE RECEIVER or U+1F57D RIGHT HAND TELEPHONE RECEIVER or U+260E BLACK TELEPHONE or U+1F57E WHITE TOUCHTONE TELEPHONE, but the symbols are widely recognized.  ERLENMEYER FLASK is certainly a strong and recognizable science symbol.

This circles back again to some of Charlotte Buff's points.  A "bride" is a culturally-recognizable symbol, with specific graphic cues (bridal gown, which is unlike other dresses in use; veil) that we associate with a bride and weddings, specifically.  The "male counterpart" of a bride is not a "man in tuxedo."  A man in a tuxedo is a man in a tuxedo, and might be the head waiter.  It happens that Western culture lacks recognizable cues that signify "man about to be married"; that symbol just isn't representable the way "bride" is.  Show any European or American (and probably Japanese) kid a picture of a bride, and they'll say it's a bride, no matter what other pictures you showed them before.  Show them a picture of a groom—without any other context—and they might recognize it as a waiter, or a maitre-d', or a person going to a fancy ball...  Much the same argument regarding kid-recognizability can be made with respect to Santa Claus/Father Christmas vs "Mother Christmas."  Any time of the year, in any context, any kid will recognize Santa Claus.  Mrs. Claus, on the other hand, shown out of context, would be a real stumper.  I could easily see someone guessing she was Mother Goose.  (Actually the whole "show it to some random kids" test is not a bad idea for judging the sensibility of proposed emoji.)

There are a bunch of cultural visual cues we recognize for a variety of things, some of which aren't yet emoji (or emoji sequences) but likely could/should be.  Certainly something like BURGLAR or THIEF would be recognizable by the telltale mask and hat (person + mask might be a good sequence—if we had a "mask" emoji, which we don't and likely should); CONVICT by the striped clothes.  (Not all the things we like to talk about are things we approve of.)

OK, enough rambling for now.  Back to your usual discussions.

~mark
Received on Wed Jan 17 2018 - 20:31:34 CST

This archive was generated by hypermail 2.2.0 : Wed Jan 17 2018 - 20:31:35 CST