Re: Revised N2586R

From: Philippe Verdy (
Date: Mon Jun 23 2003 - 13:06:40 EDT

  • Next message: Michael Everson: "Re: Revised N2586R"

    On Monday, June 23, 2003 2:54 PM, Michael Everson <> wrote:
    > > It wouldn't be hard to provide a comparable descriptive paragraph
    > > that began with an image of the Stars and Stripes, but I don't think
    > > we'd want to encode the US flag as a character.
    > That would be a logo.

    Most probably not: such an image of a country flag without its colors is not meaningful, this is just a form, a contour, which, if assigned to a character with a representative glyph, could be colored with yellow and red stripes, and would not have the semantic of the same flag, or could be seen as a caricature.

    All flags are meaningful only with a minimum of recognizable colors which have an history and meaning. Some logos too, but not all. Also a flag can be exposed by people mostly at will under some conditions attached to respect. A logo is copyrighted and is a piece of art with a owner that has some exclusive rights on it, like glyphs in a font (in most countries except glyphs created in US). So a flag is really a colored image, not a logo, not a glyph and thus not a character either...

    Even its proportions and design are well defined, unlike many glyphs associated to characters, which accept a lot of variations without loosing their character semantic. On the opposite, a Christian Cross or a Muslum Moon, qualifies as a character, because a representative glyph will accept many variations, without loosing its meaning as a religious symbol. Same thing for common symbols encodable as characters like a heart symbol for card games, a king symbol for chess, etc... These symbols represent real concepts that may have corresponding words when used in a sentence, and used in various languages (so they can be part of a formal "script"). On the opposite, the various forms of bullets or arrows in Dingbats, are probably excessive, as they can be swapped without loosing their semantic. These should have been unified as characters, with possible glyph variants.

    -- Philippe.

    This archive was generated by hypermail 2.1.5 : Mon Jun 23 2003 - 13:51:21 EDT