From: Bill Davidsen (davidsen@prodigy.com)
Date: Wed Sep 06 2000 - 18:14:02 CDT
In <G0GKL0.CHz@clw.cs.man.ac.uk>,
chl@clw.cs.man.ac.uk (Charles Lindsey) wrote:
> In <yl7l8qebsv.fsf@windlord.stanford.edu> Russ Allbery <rra@stanford.edu> writes:
>
> >I'm not arguing that we should cram everything into one parameter. I'm
> >arguing that we should add an additional parameter which has the same
> >semantics as NNTP-Posting-Host, namely "if we're choosing to provide any
> >data that may be useful for filtering, this parameter will be present and
> >you can use it as a black box of data to filter on." This is what we have
> >with NNTP-Posting-Host now, and I think we should provide a parameter with
> >the same semantics.
>
> OK, but I thought my posting-host parameter DID have the same semantics as
> NNTP-Posting-Host. The only difference I am aware of is that I have made
> provision for either FQDN or IP number or Both. Is there anything in the
> existing NNTP-Posting-Host I have not included?
>
> OTOH, I make provision for further parameters, which may or may not
> provide further filtering opportunities.
On this one I would bless using a filed called NNTP-Posting-Host,
followed by FQDN or IP. I suspect that making up a new name, or adding
things which might confuse SPAM filters which do a simple string match,
would be a negative impact. I'd love to just require IP, since some
hosts don't use FQDN, but "the horse is out of the barn" as my
grandmother used to say.
And given the likely near future, we should formally bless IPv6 IP
addresses as well, because they're coming sooner than the next draft.
I really would like to get rid of the whole header, but it would be a
decade before people started filtering on a new one. Let's keep it just
the way it is, or not have it at all and use one of the field of
whatever the injector info is called today.
-- -bill davidsen (davidsen@prodigy.com) "The secret to procrastination is to put things off until the last possible moment - but no longer" -me