Wolf Eichler wrote:
> I still would not say it makes much sense to introduce a general "end
> user defines header field" feature. Apart from your valid point #2,
> *new* fields only make sense when there is a recipient who knows what
> they mean (semantic contract). As header fields are only interpreted by
> software, there is a lack of recipient, rendering the action nonsense.
>
But no programmer, even a Mozilla one, can predict what custom header
fields might be understood by other software in the future. Allowing the
user to specify custom headers allows them to deal with things like this
in the future.
E.g. the mozilla news client doesn't allow one to add "X-no-archive: yes".
And, of course, this doesn't even address those who have customised
scripts which use the headers.
--
"The world is a comedy to those who think, a tragedy
to those who feel." -- Horace Walpole
---------------------------------------------------------------
Simon Waldman, England email: [EMAIL PROTECTED]
http://www.firecloud.org.uk/simon
---------------------------------------------------------------
