This is an archived snapshot of W3C's public bugzilla bug tracker, decommissioned in April 2019. Please see the home page for more details.
For text (or text-like) MIME types, the character encoding is crucially important. Right now, the sniffing algorithm mostly just throws it away, even when the information is known. The spec should better handle the cases where a charset parameter is present in the supplied MIME type and when the charset parameter can be set according to the BOM. See also: bug 22009.
As I mentioned elsewhere, I'm not convinced we should change much to the status quo.