Gnus development mailing list
 help / color / mirror / Atom feed
From: Kai Grossjohann <kai@emptydomain.de>
Subject: Re: How to use the spam.el package?
Date: Tue, 04 Nov 2003 21:39:39 +0000	[thread overview]
Message-ID: <87ptg724pw.fsf@emptyhost.emptydomain.de> (raw)
In-Reply-To: <4nu15l175o.fsf@lockgroove.bwh.harvard.edu> (Ted Zlatanov's message of "Mon, 03 Nov 2003 16:20:03 -0500")

Ted Zlatanov <tzz@lifelogs.com> writes:

> On Mon, 03 Nov 2003, kai@emptydomain.de wrote:
>> 
>> Well, spam is being processed by Bogofilter in my case.  Or, put
>> another way, the processing I do on spam is to move it to
>> INBOX.spam.
>
> OK, how would you suggest I change the docs and/or implementation?  I
> understand the problem, I'm just not sure how I can fix it.

I think it's better to change the docs.  The behavior makes sense:
moving spam to a destination group is something that can be done in
addition to other processing.

(The alternative would be to provide a list of actions, such that
moving to another group could be part of it.  But that would make the
configuration more complex for no apparent gain.)

Maybe you could extend the description of M-d.  Perhaps like this:

    Marking a message as spam means that Gnus might do two things with
    it.  First, it can analyze the message (we call it "spam
    processing"), and secondly it can move the message to another
    group (which is called the "process destination").  Read on for
    more on spam processing and process destinations.

Further down, it says "you have to collect your spam in one or more
spam groups, and set spam-junk-mailgroups appropriately".  The
following questions come up when you're naive:  How to collect the spam
in spam groups?  Of course, the answer is by hitting M-d and then
relying on the process destination to do the collecting.  But the docs
don't say :-)

And secondly, why do I *have* to set spam-junk-mailgroups?  Isn't it
enough to just process the spam?  How does spam-junk-mailgroups come
in?  Why do I need a group where all new articles get the spam mark
automatically?  Say I use the black/whitelist thing.  Then why is it
not sufficient to just blacklist the sender, then delete the spam?



I also mentioned a tutorial-type example.  It seems that you use
spam.el in the "normal" way, in the way that spam.el was intended to
be used.  So how about you describe your setup, explaining what each
configuration item does?  Something along these lines:

    I set nnmail-split-fancy like this: ....  This invokes spam-split
    which puts spam into a specific group.  This group is specified by
    the spam-split-group variable.  I have (setq spam-split-group
    "incoming-spam"), so spam-split will put spam into the
    nnml:incoming-spam group.  From time to time, I enter that group
    to check whether messages in that group are really spam.  If I
    encounter any ham, I hit M-u on it.  When exiting the group, the
    spam from nnml:incoming-spam is moved to nnml:train-spam and the
    ham is moved to nnml:train-ham.  I do this by setting the group
    parameter spam-process-destination to nnml:train-spam and
    ham-process-destination to nnml:train-ham.  I then use
    nnml:train-spam and nnml:train-ham for bogofilter training.  I set
    the spam-processor group parameter for nnml:train-spam to
    gnus-group-spam-exit-processor-bogofilter, and for the group
    nnml:train-ham I set the ham-processor group parameter to
    gnus-group-ham-exit-processor-bogofilter.  After being processed
    by bogofilter, spam from nnml:train-spam will be deleted because
    of the foo setting.  (Ed. note: which setting?)  And ham from
    nnml:train-ham will be moved to nnml:mail.misc because I set the
    group parameter ham-process-destination on nnml:train-ham
    accordingly.

In the above description, I tried to guess what your, Ted's, setup
might be.  I'm sure it's wrong.  But maybe you get the idea of what I
mean by showing the connection.

In fact, I think that lengthy paragraph is still too difficult to
understand to see the big picture.  Hm.  We need a summary explaining
just what happens.  Maybe a drawing would be nice?  But I'm really bad
at drawing...  So, maybe the above should be split into a description
what happens and then a description of the settings that are needed to
make it happen.

WDYT?

I could also describe my setup, but I'm afraid that's not really
typical.  After all, I try to make it server-based, and I guess most
users don't have their own server.

Kai




  parent reply	other threads:[~2003-11-04 21:39 UTC|newest]

Thread overview: 11+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2003-11-01 20:17 Kai Grossjohann
2003-11-03 18:37 ` Ted Zlatanov
2003-11-03 20:26   ` Kai Grossjohann
2003-11-03 21:20     ` Ted Zlatanov
2003-11-03 22:06       ` Kai Grossjohann
2003-11-04  2:36         ` Ted Zlatanov
2003-11-04 21:39       ` Kai Grossjohann [this message]
2003-11-03 20:15 ` Kai Grossjohann
2003-11-03 20:25   ` Ted Zlatanov
2003-11-03 21:04     ` Kai Grossjohann
2003-11-04 20:57       ` Kai Grossjohann

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=87ptg724pw.fsf@emptyhost.emptydomain.de \
    --to=kai@emptydomain.de \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).