>I'm not familiar with Truetype internals, so my question might fail >the issue, but why do there have to be multiple encoding mappings for >Unicode at all? This seems like a little bit chaotic to me. Why does >this font not just contain a single set of information for Unicode usage?
Backwards compatibility. So the font could still be used on a 1 byte system, it might have a format 0 cmap. For a system which only implements the BMP, a format 4 cmap might be all it supports. Of course its easy to see now that a format 12 cmap is the best, but it might not work on all systems which use true-type fonts. -- Linux-UTF8: i18n of Linux on all levels Archive: http://mail.nl.linux.org/linux-utf8/
