- From: Jonathan Kew <jfkthame@gmail.com>
- Date: Sun, 02 Nov 2014 01:17:52 +0000
- To: Brad Kemper <brad.kemper@gmail.com>
- CC: www-style list <www-style@w3.org>
On 30/10/14 22:06, Brad Kemper wrote: > >> On Oct 30, 2014, at 7:09 AM, Jonathan Kew <jfkthame@gmail.com> >> wrote: >> >> On 30/10/14 13:36, Brad Kemper wrote: >> >>>> I'm guessing the 'mandatory <infinite>' for Safari is just a >>>> bug not really a feature. >>> >>> It is kind of crazy to have that for everything, but I think >>> 'mandatory infinite' would be useful for icon/picture fonts. Or, >>> if not 'infinite', then some sort of keyword to make it fall back >>> to a generic glyph (such as a square), instead of whatever letter >>> the icon is mapped to. . >>> >> >> What's this about "whatever letter the icon is mapped to"?! >> >> Icon glyphs should not be mapped to letters. They should be mapped >> to the appropriate Unicode symbols/dingbats/emoji/whatever (which >> means applying browser font fallback is perfectly reasonable); or >> if the icon concerned is not encoded in Unicode, then to Private >> Use Area codepoints (for which browsers shouldn't be attempting >> font fallback at all; see CSS3-Fonts[1]). > > That sounds like a fine idea. Do most dingbat fonts do that these > days? > Some - I'd be tempted to say "many", but have not attempted a survey - widely-used ones do. No doubt there are also some that assign icons or dingbats to arbitrary ASCII character codes, but that should simply be considered broken; and if such fonts perform less-than-optimally with regard to font loading, fallback, etc...... well, too bad. Authors who care about the quality of their pages and the user experience they provide should learn not to use them. JK
Received on Sunday, 2 November 2014 01:18:21 UTC