This is an archived snapshot of W3C's public bugzilla bug tracker, decommissioned in April 2019. Please see the home page for more details.
Currently the specification states that Accept-Charset should be set as appropriate by the user agent. However, Accept-Charset is a useless header that is on its way out. Opera already dropped it from requests and Mozilla is planning to. Alternatively maybe the specification should not say anything about Accept-Charset and Accept-Encoding except that authors do not have control over them and that content-encodings used in transmission are automatically decoded by the user agent.
Proposal for CfC: Do not say anything about Accept-Charset and Accept-Encoding in the send() method section.
Changed the specification as suggested.