RE: Engmagate?

From: Whistler, Ken <ken.whistler_at_sap.com>
Date: Fri, 13 Dec 2013 19:32:19 +0000

Well, inconceivable? No. Inadvisable? yes.

First of all, such “comments” are not actually “comments”—they are the result of a fairly cumbersome and drawn-out process of adding *normative* standardized variation sequences to the standard.

Second – although this is a nit – FE0E and FE0F would not be used for standardized variation sequences of this type, because of their special pre-emption for the emoji variants. But FE00, FE01, etc., are, of course available.

More importantly, I consider it a *very* bad precedent to attempt to start down the road of mixing standardized variants with case pairing. That is almost guaranteed to create permanent implementation problems that would surely be worse than the perceived problem it would be attempting to solve. Case mapping could fail under circumstances mysterious to end users. Likewise, string matches would fail under mysterious circumstances, where implementations doing multi-level weighting and/or simply stripping variation selectors would match, but the same strings doing binary compares would fail, where they might not have before.

Do such problems afflict the existing standardized variation sequences? Well, yes, to a certain extent. But to date they have not been simultaneously pulled into the maelstrom of case mapping and case folding. Mixing the two is just a recipe for a big mess, IMO.

--Ken


Would it be inconceivable to add comments such as:

014A LATIN CAPITAL LETTER ENG
* glyph may also have appearance of large form of the small letter
~ 014A FE0E N form
~ 014A FE0F n form
?
Received on Fri Dec 13 2013 - 13:35:29 CST

This archive was generated by hypermail 2.2.0 : Fri Dec 13 2013 - 13:35:31 CST