usefor-article-07 May 2002

4.2.1.  Naming of Headers

   Despite the restrictions on header-name syntax imposed by the
   grammar, relayers and reading agents SHOULD tolerate header-names
   containing any US-ASCII printable character other than colon (":",
   US-ASCII 58).

   Whilst relaying agents MUST accept, and pass on unaltered, any non-
   variant header whose header-name is syntactically correct, and
   reading agents MUST enable them to be displayed, at least optionally,
   posting and injecting agents SHOULD NOT generate headers other than
     o headers established by this standard or any extension to it;
     o those recognized by other IETF-established standards, notably the
       Email standard [RFC 2822] and its extensions, excluding any
       explicitly deprecated for Netnews (e.g. see section 9.2.1 for the
       deprecated Disposition-Notification-To-header); or,
       alternatively, those listed in some future IANA registry of
       recognized headers;
     o experimental headers beginning with "X-" (as defined in 4.2.5.1);
     o on a provisional basis only, headers related to new protocols
       under development which are the subject of (or intended to be the
       subject of) some IETF-approved RFC (whether Informational,
       Experimental or Standards-Track).
   However, software SHOULD NOT attempt to interpret headers not
   specifically intended to be meaningful in the Netnews environment.

   Header-names are case-insensitive. There is a preferred case
   convention, which posters and posting agents Ought to use: each
   hyphen-separated "word" has its initial letter (if any) in uppercase
   and the rest in lowercase, except that some abbreviations have all
   letters uppercase (e.g. "Message-ID" and "MIME-Version"). The forms
   given in the various rules defining headers in this standard are the
   preferred forms for them. Relaying and reading agents MUST, however,
   tolerate articles not obeying this convention.