- From: David J Woolley <djw@bts.co.uk>
- Date: Tue, 10 Nov 1998 18:23:31 +0100
- To: www-amaya@w3.org
> - Using the Greek characters in plain text (poor example: beta-test) > To be covered by ??? > The greek letter entities α etc. These work on IE4 and at least display literally on NS4. Or, the corresponding Unicode &#hhhh; codes, which work correctly on IE4 and display as a substitute character on NS4. Or, of course, you can use UTF8 as your character set in the HTTP headers, or even the 16 bit Unicode character set. UTF8 is the variable length Unicode coding, used by Java etc, which has an identity mapping onto true (7 bit) ASCII. Font hacks only work on browsers that have soft fonts and actually know the font being used. Note that HTML 4 is nominally in Unicode. The conceptual logic is that the transfer character set (from the HTTP headers) is converted to Unicode, then &...; entities are resolved into Unicode. Finally this is approximated in terms of the characters available in the selected fonts (but only for display purposes), or, in practice, a similar font if the primary font doesn't contain the character. This does mean that the transfer character set needs to support a substantial part of pure ASCII. -- David Woolley - Office: David Woolley <djw@bts.co.uk> BTS Home: <david@djwhome.demon.co.uk> Wallington TQ 2887 6421 England 51 21' 44" N, 00 09' 01" W (WGS 84)
Received on Tuesday, 10 November 1998 13:33:19 UTC