From: Peter Kirk (firstname.lastname@example.org)
Date: Sat Mar 06 2004 - 12:18:34 EST
On 06/03/2004 08:15, Ernest Cline wrote:
>... I think that
>LATIN SMALL LETTER TH WITH STRIKETHROUGH does not
>meet the guidelines of Annex H of the Principles and Procedures.
I assume that you are referring to
http://std.dkuug.dk/JTC1/SC2/WG2/docs/n2652r.pdf; or actually a previous
version as your point numbering does not match this version of the
document. Bear in mind that this is a WG2 document and not a Unicode
>Why? First of all, let me say that it does come close to meeting them,
>but not quite. First off, there is no demonstrated community of users
>that would instantly recognize this glyph and say, "I know what that
>is supposed to mean." (H.6 Point 1) Indeed, as Peter Kirk said:
>>there could well be a dictionary out there somewhere which uses
>>one of your supposedly equivalent ligatures for the voiced th and
>>another one for the unvoiced th.
I don't think my words justify the point you are trying to make with
them. My hypothetical other dictionary would have its own community of
users, largely distinct from the communities of users of the
dictionaries using the ligatures under discussion.
>That argument not only convinced me against the unification
>I proposed, it convinces me that this character truly belongs as
>a private use character. It has no well defined semantics.
>(H.6 Point 8) and no evidence that it is widespread. (H.6 Point 14).
Your reasoning is flawed because you seem to assume that a character to
be encoded must meet ALL of the criteria in the list in Annex H.6. But
the text does not state that; rather this is a list of "Some criteria
that strengthen the case for encoding".
Anyway, the character "has well defined user community / usage", the
users of the dictionary in question. It is not clear that "user" implies
those who write the character, or only those who read it. Many
historical characters have been accepted for Unicode which are not
regularly written, except in copying old texts, but are still regularly
The character also has well-defined semantics, indeed they are
explicitly defined in the dictionary. The only generally applicable test
which the character fails is that of being widespread; but it is
actually much more widespread than some characters recently (and
correctly) accepted for Unicode. But characters are not expected to meet
all these criteria.
I note also the principle in H.9 that "There are glyph changes that
cannot be absorbed quietly since the new glyph bears so little relation
to the old one that the change exceeds the implied range of glyphic
variation." Although this refers primarily to glyph change, I think it
can also be applied to synchronic glyph differences which exceed a
certain implied range.
-- Peter Kirk email@example.com (personal) firstname.lastname@example.org (work) http://www.qaya.org/
This archive was generated by hypermail 2.1.5 : Sat Mar 06 2004 - 12:53:15 EST