lucy24 - 8:49 pm on Aug 25, 2012 (gmt 0)
<begin topic drift>
Ooh, what fun. On my screen, the Unicode Replacement Character (the black diamond) turns into � (i-umlaut, inverted question mark, one-half) because my browser has already decided the page is in Latin-1, and it can't change its mind halfway.
It means that your copy of FF defaults to UTF-8 encoding while your other browsers default to Latin-1. The difference only becomes noticeable when someone enters a non-ASCII character.
:: detour to confirm ::
Ay-yup. The browser doesn't know what to do with A0 alone-- one-byte characters stop at 7F-- so it shows each one as EF BF BD-- which unpack in the other direction to your three Latin-1 characters.
And that's why so many people insist on using entities for anything outside the vanilla ASCII range. A simple <charset> declaration will...
n/m. Been there already.
</end topic drift>
... sort of, because every time I see "chlooe dot com" I wonder if the site owners really wanted to say Chloe with dieresis-- which, ahem, I won't attempt here, see above-- but something got garbled in transit.
Is that a French server farm, or is somone running a robot from the middle of a "real" ISP?
Footnote: And then, for reasons I won't try to go into, my Latin-1 preview turned into UTF-8 as soon as I posted for real. I believe this has happened before. It confuses me horribly. Especially the part where it toggled back in the other direction when I added this paragraph.