24336 2014-01-20 18:18:42 +0000 Encoding names should match what people actually call them 2014-01-20 23:30:22 +0000 1 1 1 Unclassified WHATWG Encoding unspecified PC Linux RESOLVED WONTFIX P2 normal Unsorted 1 geoffers+w3cbugs annevk mike www-international sideshowbarker+encodingspec oldest_to_newest 98782 0 geoffers+w3cbugs 2014-01-20 18:18:42 +0000 http://gsnedders.html5.org/web-encoding-names/results.html shows what document.characterSet returns in current versions of browsers. Notably, Firefox and Chrome both return the uppercased names for many of these. (IE returns them all lowercase except "GB18030"; ZombieOpera returns them all lowercase) Googling these encoding names it becomes clear that almost everyone refers to "UTF-8", "ISO-8859-n", etc. (uppercased), and as there is no interop here currently, and the proposed behaviour matches Firefox/Chrome, it would seem better to just give them their names that are in common usage. As such, I propose to change the names to the following (thereby changing case only): - UTF-8 - IBM866 - ISO-8859-n - ISO-8859-8-I - KOI8-R - KOI8-U - HZ-GB-2312 - Big5 - EUC-JP - ISO-2022-JP - Shift_JIS - EUC-KR - UTF-16BE - UTF-16LE 98796 1 annevk 2014-01-20 23:30:22 +0000 I value more that now you can predict what characterSet returns. With your proposed change you need to know that windows-1252 is not spelled Windows-1252. And that gb18030 is an exception.