Re: ASRG/Filtering List Launch: "Let the games begin!"
"Mark E. Mallett" <[email protected]> Mon, 9 Feb 2004 18:13:07 -0500
| Newsgroups | gmane.ietf.asrg.filtering |
|---|---|
| Message-ID | <[email protected]> |
On Sun, Feb 08, 2004 at 01:09:10PM -0500, Yakov Shafranovich wrote:
> John Graham-Cumming wrote:
>
> Here is a note I got from one of the SpamAssasin people on the subject:
>
> My main concerns:
>
> - making sure most MUAs out there can filter on the set of headers
> - the format not sucking :-)
>
> For reference, SpamAssassin includes four headers to cover most of the
> bases:
>
> 1. X-Spam-Status: Yes, hits=22.1 required=7.5 (and other keyword=value
> pairs)
etc-
One problem with these headers are that they aren't specific about
where they were added or what facility added them. Something I
mentioned on the main list a while back:
Personally I think that with a variety of ways of marking
scores (or in fact even having the concept of a "score"), it
would be hard to specify the a particular structure here.
Plus you might have a message passing through multiple
filters, each of which would want to place its mark on the
message without disturbing other such marks (so you'd need a
tuple of at least {agent, location, assessment})
Where agent should be the filter name and specific version number,
location should let you find out where in the process they were added
(e.g. at least an IP address of the host, but probably more than just
that), and assessment would be the result of the analysis. I suppose
one would also need the assessment type (number of '*'s, thumbs
up/down, floating point value, etc).
Also, 4 separate headers injected by one filter is a bit much, I think.
A single keyword/value oriented header with folding white space would
be better.
I think Eric Raymond had mentioned that he had some notes on the
matter- if so it would be nice to see those.
mm