At 22:18 -0700 2004-05-19, John Hudson wrote:
I don't automatically accept the argument, made by Michael earlier today, that 'There is a requirement for distinction for X in plain-text'.
The Universal Character Set is supposed to contain all the scripts of the world. For generations students of writing systems have been naming and distinguishing scripts. I have such books here from the middle of the 19th century; presses in France, England, and Italy were cutting type for these scripts two hundred years before that at least. The plan is to encode the important distinguishable nodes. A different level from palaeography, as can be seen from the unification of varieties under the rubric "Phoenician". (I note that at least one person has argued reasonably for splitting Neo-Punic off of that unification.)
On what basis do we decide that X is necessary in plain-text while Y should be done with mark-up or some other 'higher level protocol'?
Wit and taste? There isn't an algorithm. -- Michael Everson * * Everson Typography * * http://www.evertype.com
