Time and time again, the use of any character in a post subject line or heading beyond the US-ASCII repertoire causes such characters to be incorrectly escaped. The entire site has to be UTF-8-compliant.
Typophile
MetaPhile: Character encoding in post headings is borked
I agree that this needs to be fixed. &3049; in place of a character makes the headings hard to read.
- Lex
Yeß!
This is actually (mostly) due to the sIFR Flash thing that formats the headings in a custom font. It doesn't seem to be able to handle anything outside simple ASCII. The fact that often when you post something (including topic titles) some auto-formatting takes place, converting dumb quotes to smart quotes and so forth, which sIFR can't properly display. I don't think this is inherent to the way sIFR works. Maybe it is.
sIFR can handle characters beyond ASCII, see its How to use and Tricks pages.
I am sure that the fonts featured in the header do contain an ellipsis, an em-dash etc., and also offer a decent language support.
However – IIRC –, Typophile uses something different than sIFR (see Christian Robertson’s comment here).
Coding Flash to handle characters beyond ASCII is no biggie.
The script errantly converts high-order characters to entities, then (again errantly) deletes leading ampersand (also number sign, if hex – I can’t tell) and trailing semicolon. It then renders the remaining numbers.
Captured from web.archive.org on 29 Jul 2014. view this capture