W3C home > Mailing lists > Public > www-style@w3.org > August 2013

Re: [css-syntax] Defining "character"

From: Zack Weinberg <zackw@panix.com>
Date: Mon, 12 Aug 2013 06:52:01 -0700
Message-ID: <CAKCAbMgBvxytgnkjKUQKsVdfv3QL6vUeYWsQu6rbFGtnvTmdJA@mail.gmail.com>
To: Simon Sapin <simon.sapin@exyr.org>
Cc: www-style list <www-style@w3.org>
> 2. Have a normative definition near the beginning like this: "Within this
specification, the ambiguous term <dfn>character</dfn> is used as a
friendlier synonym for Unicode code point." This is what CSS Text does,
except that there "character" means grapheme cluster rather than code point.

I prefer this option. I think there should also be a note reminding the
reader that a "code point" is a single entity with a value from U+0000
through U+10FFFF, i.e. surrogate pairs do not appear after conversion from
the input encoding. (This needs clarification because the terms "code
point" and "code unit" are confusingly similar, and because JavaScript uses
code units in this context (iirc).)

> 1. feels verbose, but 2. means two different definitions of the same term
in two different spec, which definitely smells bad.

The contexts are sufficiently distinct that I don't think 2. will be a
source of confusion.

zw
Received on Monday, 12 August 2013 13:52:38 UTC

This archive was generated by hypermail 2.4.0 : Friday, 25 March 2022 10:08:33 UTC