Michael Everson a écrit :
At 12:08 -0700 2004-05-14, Peter Kirk wrote:
ell, I accepted somewhat reluctantly that Phoenician should be separately encoded because a small number of users want it to be, although a majority apparently do not want it to be.
I really don't know if those who spoke for the "majority" were really representative of a real "majority".
[PA] Is representing the majority of a community of users important ? If so, how do we know what this majority thinks ? Or, as was mentioned, these users are sometimes too conservative and then don't really know what is good for their own good in terms of script analysis and their preferences should be ignored ?
This would not be an acceptable position if Unicode intended to force all users of Phoenician to move immediately to the new script - although it would actually make much more sense to do so.
Unicode doesn't force people to do anything. (Well, apart from using smart font technology for a lot of scripts, but that's not relevant here.) Unicode makes characters available for those who wish to use them.
[PA] Surely Unicode does not make all characters available : it rejects some and unifies some. Why reject or unify if their inclusion would not pose a problem ? I somehow have the impression that the sheer presence of characters (duplicates for instance) does have an effect on users and forces certain processing (normalisation sometimes, decomposition in some cases, changing transcoding filters in other cases (what are the Coptic users having Coptic texts encoded as Greek data going to do?), changing/adding Cmap for some fonts (Coptic ones previously indexed with Greek code points ?)), etc. to achieve the desired effect.
P. A.
