Re: name change

From: Asmus Freytag <>
Date: Tue, 29 Nov 2011 08:43:33 -0800

Well put, Jon. The role of a more complete human-readable (yet still
short) description is handled by the totality of a character's entry in
the code charts. For some characters, this includes up to a dozen lines
of notes and other information. In many cases, this information is
enough to let you pinpoint the character, that is relate it to its real
use in text, no matter the context. Where that is insufficient, the full
description includes the properties for the character from the UCD and
the discussion in the relevant chapter of the standard.

The notes and discussions in the text can be amended any time to fix any
shortcomings, the same is true for most of the UCD properties for the


On 11/29/2011 1:55 AM, Jon Hanna wrote:
> On 2011-11-23 10:38, Jeremie Hornus wrote:
>> I was thinking the ID being the code point value itself, and the "name"
>> a human readable description of it.
> They are both IDs. One is from the range of numbers from 0 to 1114111
> (10FFFF base 16), the other is from the range of strings of characters
> each in the range {U+0020, U+002D, U+0030-U+0039, U+0041-U+0056} with
> rules for equality matching that lets us compare with mixed-case forms.
> In both cases we have a value which can be compared for equality with
> another value, which refers to at most one entity, and which offers a
> reasonable guarantee of immutability (due to the stability on names
> that started this thread). They're both IDs.
> On the other hand, the name is not always very good as a human
> readable description. It's unpleasant in this context at the best of
> considered incorrect, and it very often when accurate is still
> incomplete as to the different uses made of the character in different
> contexts, all of which makes it occasionally poor as a human readable
> description (pretty much every time people have wanted to have a name
> changed, at least one person thought it failed in this regard).
> The human-readability is a convenience for those using the
> identifiers, but it's more akin to the human-readability of
> programming and markup languages than the human-readability of a
> well-written reference book.
Received on Tue Nov 29 2011 - 10:51:06 CST

This archive was generated by hypermail 2.2.0 : Tue Nov 29 2011 - 10:51:23 CST