- From: Mark Davis <mark.davis@jtcsv.com>
- Date: Fri, 14 Mar 2003 10:59:55 -0800
- To: "Roozbeh Pournader" <roozbeh@sharif.edu>
- Cc: "Roberto Scano - IWA/HWG" <rscano@iwa-italy.org>, <ishida@w3.org>, <w3c-i18n-ig@w3.org>, <public-i18n-geo@w3.org>, "Al Gilman" <asgilman@iamdigex.net>, <w3c-wai-gl@w3.org>
I see two edges. (a) If you want to be most standard, you simply put the character in the document. When someone prints/displays the document, they will presumably use a font that is appropriate for them (or for their target audience). Your mother wouldn't use an Urdu font to print a document; she wouldn't see Urdu glyphs for the numbers. (b) If you want the precise shape, no matter what the target environment, then you would embed an image in the document. That way you are insulated from the vagaries of the user's environment (but are locked into a fixed shape). Approach (a) is a bit like just-in-time localization, where one transmits binary numbers wherever possible. At the latest possible point, they are formated into whatever the user's locale conventions are. There are intermediate points between these. For example, no matter what the target environment, I could want a glyph appropriate for Urdu, but I don't care about the exact shape. Or I could want a glyph appropriate to a particular font (Times New Roman), but not care about the precise shape, which may vary somewhat on different machines or resolutions. In either of these latter two cases, I am dependent on the level of support offered on the recipient machine; if Urdu or Times New Roman are not installed/available, then I will get some fallback anyway. Mark ________ mark.davis@jtcsv.com IBM, MS 50-2/B11, 5600 Cottle Rd, SJ CA 95193 (408) 256-3148 fax: (408) 256-0799 ----- Original Message ----- From: "Roozbeh Pournader" <roozbeh@sharif.edu> To: "Mark Davis" <mark.davis@jtcsv.com> Cc: "Roberto Scano - IWA/HWG" <rscano@iwa-italy.org>; <ishida@w3.org>; <w3c-i18n-ig@w3.org>; <public-i18n-geo@w3.org>; "Al Gilman" <asgilman@iamdigex.net>; <w3c-wai-gl@w3.org> Sent: Friday, March 14, 2003 10:13 Subject: Re: Re[2]: FW: acronym in title... > On Fri, 14 Mar 2003, Mark Davis wrote: > > > Good point, although you are safest in that case actually annotating with > > the font, and not depending on the target system. > > That's not safe. That way your document won't be standard. How do you know > what will be the exact name of an Urdu font on a future desktop with a > very standard browser and a very standard set of fonts? > > Since OpenType allows language-specific glyph replacements, that desktop > may not even have an Urdu-specific font. It may have an Arabic font that > has an extra glyph for the Urdu 7 to use only when some software > specifically asks for text to be set in Urdu. Annonating with a font name > will not have *any* effect in this case. > > roozbeh > >
Received on Friday, 14 March 2003 14:00:17 UTC