This page is a compilation of formal public feedback received so far. See Feedback for further information on this issue, how to discuss it, and how to provide feedback.
Date/Time: Tue Apr 7 18:37:41 CDT 2015
Name: Asmus Freytag
Report Type: Error Report
Opt Subject: Proposal for UAX#24 Script Name
I propose a new section 2.9.x in UAX#24 that describes in more detail and
rigor how the script extension property is assigned.
< start of proposed section >
Section 2.9.x Script_Extensions Property Values
This section describes formal construction and constraints on the
Script_Extensions property values. In the following the Script_Extensions
property is abbreviated /scx/.
1. Each code point has exactly one non-empty scx set associated with it.
That is, the property value for the scx property always consists of a set.
The empty set is not allowed; the scx value for unassigned and non-character
code points is { Unknown }.
2. The elements of the scx set consist of an unordered list of unique values
of the Script property (sc values).
The scx values { Latn Grek } and { Grek Latn } are identical; for ease of
comparison the sets may be sorted and listed in alphabetical order.
3. For the purposes of creating scx sets, the following sc values, and only
these are considered /implicit/: /Common/, /Inherited/, and /Unknown/. All
other sc values are considered explicit.
4. An scx set either contains a single implicit sc value or one or more explicit sc values.
Examples:...
5. If Script(c) is explicit, then that value must be an element of the scx set as well.
Examples...
The number of values for the Script property grows every time a new script is
added to the Unicode Standard. Implementations are advised to allow for
growth. On the other hand, it is unlikely that any scx value would
individually list even a majority of existing scripts. It is for such cases of
widespread use of characters across scripts that the sc value /Common/ was
created.
There are no formal rules for when an sc value is added to the scx set for a
code point; whether to document that a character is used with multiple scripts
via the Script_Extensions property remains a judgment call, and is always
based on the best available information to the Unicode Technical Committee.
< end of proposed section >
I further propose that the Unicode Consortium *not* duplicate this information
in the stability policies. If considered absolutely necessary, my preferred
suggestion would be to cite the section and rules by reference and declare
those which must be formally stabilized.
My strong suspicion is that it is more important to document the script
extension property than to make formal guarantees.
One thing which should be considered is whether scripts can be delisted from
an scx set as more information becomes available. My strong inclination is to
answer that in the positive. (This would get an item 7 in the proposed
section).
Date/Time: Wed Apr 8 14:24:18 CDT 2015
Name: Asmus Freytag
Report Type: Error Report
Opt Subject: Editorial nit: UAX#24
Ed Note: This has already been fixed in the draft.
This is a minor editorial nit on UAX#24 Table 4. Examples of East Asian Symbols with Common Script Values ==> Table 4. Examples of East Asian Symbols with Script Value = common This affects other table headers and surrounding text as well. Reason: "common values" has a whole set of meanings beyond "value = common"... and even if it has the latter meaning it is not obvious without careful reading that "common" here is not the same a "commonly occurring" or "shared".