barryvan 01 December 2011 Have you checked that this issue is logged as a bug in bugzilla? Given that this is a quite a niche use (at least at present), it’s not unlikely that the Firefox developers aren’t even aware of this issue. Also, you say that “the font shouldn’t be downloaded if none of the characters within the Unicode range are present in a given page.” Is that the actual behaviour, or is that just the expected behaviour? Would an alternative fallback be to strip down the font (using, say, FontForge) so that it only includes the characters to be substituted? I understand that some fonts’ licensing and distribution methods preclude this, but for other fonts, it could potentially mean less fallback CSS and smaller font downloads.