L2/15-182


Title: Suggested Responses to Suggestions re Cyrillic in L2/15-014

Author: Ken Whistler

Date: July 20, 2015

Action: For consideration by the UTC


Background

I received action item 142-A003, to suggest responses to sections
1 through 6 of L2/15-014, which comprises a set of detailed recommendations
for annotations to various Old Church Slavonic characters in the
Unicode names list, as well as some suggested updates to the Cyrillic
section of the core specification.

Some of the names list annotations were partly implemented for
the Unicode 8.0 names list. The suggestions below sketch out what
I think would be appropriate moving forward for Unicode 9.0.
The suggestions are schematic, and would need to be forwarded to
the editorial committee to work out the details -- particularly for
the text updates to the core specification. (Note that no recommendation
is actually needed for section 1 of L2/15-014, which was just the
general introduction of that document.)

Details

1. Section 2, U+0479 Digraph Uk

The recommendations for names list annotation were already partially
implemented as of Unicode 8.0.

I suggest that in addition to those changes, the new annotations
be corrected from "digraph uk" to "digraph onik" for explicitness.
Also, the names list for the run of problematical historic letters
for uk and omega could benefit from a further subhead and/or
notice in the names list to separate it off from the less
problematical characters.

Rather than attempt further clarification in individual annotations
to 0479/047A, beyond the recommended alternative spellings already
there, I suggest that any further discussion be added as a paragraph
regarding these characters in the core specification.

2. Section 2.1, U+047C and U+047D

The recommendations for names list annotation were already partially
implemented as of Unicode 8.0. There is no need to replicate
extensive annotations for both the uppercase and lowercase letters
of a case pair like this.

I suggest, however, that a further explanation of the identity and
use of the "beautiful omega" be added in the core specification,
based on the information provided here.

3. Section 3, U+0484, Combining Palatalization

The recommendations for the annotation and change in xrefs for
U+0484 seem reasonable, and should be done for the Unicode 9.0
names list.

However, the situation is fairly complex, and deserves more discussion
than the names list annotation can do justice to. So for this, I
also suggest a small discussion be added as a paragraph to
the core specification, based on the information in L2/15-014.

Regarding the suggested change of Script value for U+0484
to sc=Inherited, I don't think this is appropriate. U+0484
should stay sc=Cyrillic. Instead, a value of {Cyrl Glag} should
be added to ScriptExtensions.txt for Unicode 9.0 to reflect
common use of this combining mark in both scripts.

4. Section 4, U+04A4 and U+04A5 Cyrillic Ligature En Ghe

L2/15-014 is recommending that U+04A4/U+04A5 also be annotated
as used for the Cyrillic letter soft en. This seems reasonable
to me, as opting to encode a separate character with basically
an identical appearance would not be a good idea here. So this
annotation should be added to the Unicode 9.0 names list -- although
it does not need to be cloned to both capital and small letters.

5. Section 5, U+2DF5 combining Cyrillic Letter Es-Te

The detailed annotation recommended in L2/15-014 would not be
correct, as it does not reflect the outcome of the discussion
of L2/15-002 in the UTC. An annotation reflecting the outcome
of the UTC decision was already added to the 8.0 names list,
so this item should be considered closed.

6. Section 6, U+0483 Combining Cyrillic Titlo

The section of L2/15-014 provides extensions discussion of the titlo
and recommends various small updates to the core specification text.
I suggest that this section be remanded to the editorial committee
to come up with a draft which fixes the specific problems identified
in the text, and perhaps uses this information to write a more extended
discussion of the titlo-related issues for the core specification.