RE: Emoji: National Flag Symbols

From: Ruszlan Gaszanov (
Date: Tue Dec 23 2008 - 15:19:30 CST

Actually, a potential decision to encode flags would bring up an interesting
problem: traditionally, a plain text character is recognized by glyph shape
only, while glyph colors are more or less irrelevant to character semantics.
As a consequence, present-day text rendering systems are not designed to
enforce specific colors on particular glyphs.

A flag, however, is a different animal entirely. Admittedly, there are some
flags that have a very distinct pattern (US or UK for instance) so they
would be recognizable without proper colors applied, but for great many
colors are critical. For instance, there would be no clear way to
distinguish between monochrome versions of French, Italian or Irish flags.

Hence, in order to use "flag characters" in plain text meaningfully, we
would need new rendering systems that a able enforce specific colors on
glyphs, and that would change the nature of plain text as we know it.

So, the question we must ask ourselves here: do we really want to do this?


-----Original Message-----
From: [] On
Behalf Of Christopher Fynn
Sent: Tuesday, December 23, 2008 4:36 AM
To: Unicode List
Subject: Re: Emoji: National Flag Symbols

Kenneth Whistler wrote:

> The emoji proposal should, IMO, encode precisely 10 EMOJI COMPATIBILITY
> SYMBOL FOR FLAG OF XXX characters and we done with it. These should be
> documented as interoperability characters for mapping these
> SJIS gaiji extensions for wireless vendors. End of story.

> It should not be opened up to a general scheme for registering
> the flag of any country, current or future, in an open-ended
> matter. That is just guaranteed to be the kind of endless
> headache that Chris Fynn was warning about. The 676 positions
> for possible country codes won't suffice, nor will the justification
> that only *national* flags could be encoded as characters stand
> up. And the whole issue of flag characters is something that
> will sidetrack the entire emoji symbols proposals when it gets
> into the ISO context.

> Better than heading that route would be to cross off the
> 10 flag symbols (like the logos) as yet more emoji which cannot
> be mapped to characters at all, but which will need to be
> mapped to strings, as they are for the DoCoMo cross-mappings.

Yes - any encoding of *specific* flags opens up a barrel of worms and,
in the end, encoding almost every national flag seems inevitable. But
then, if we have coloured flags, how about e.g. nautical signal flags -
surely these better qualify as a "writing system"....

- C

> --Ken

This archive was generated by hypermail 2.1.5 : Fri Jan 02 2009 - 15:33:07 CST