...I can also see some advantages in this. The current PUA system allows private interchange of characters which are completely distinct from existing characters. But there are many characters which people might like to interchange privately which are in fact variant forms of characters already encoded, and which share the identity and all or most of the character properties of already encoded characters, differing perhaps only in their glyph or in some minor property. If these are distinguished from the existing characters only by Private Variation Selectors (PVS), which would be default ignorable characters, they can retain all of the existing character's properties except for those which a private implementation chooses to override.
It would be nice tho, if there was something like a Private Variation Selector.
...
If a text encoded in this way is processed or displayed by a regular implementation with no understanding of the PVS, it should be processed meaningfully except that the property changes will be ignored; and it should be displayed meaningfully although with the regular glyph for the character. This is a consequence of the PVS being default ignorable.
I would expect that a rendering engine would be able to ignore the PVS when standard fonts are being used; but to render the <character, PVS> sequence with the alternate glyph if the font specifies the glyph as a kind of ligature for this character sequence.
I would also expect that the existence of such a feature would be a useful way of deflecting a significant number of less well thought out new character proposals (including some which are being suggested for Hebrew).
-- Peter Kirk [EMAIL PROTECTED] (personal) [EMAIL PROTECTED] (work) http://www.qaya.org/
