"Trei, Peter" <[EMAIL PROTECTED]> writes:
[Placing a keyword in the subject line. Good things:]
> 7. It's a proven technique, borrowed from Usenet.
And works pretty well there, it seems.
> It also has two bad properties::
>
> 1. It is butt-ugly.
Some creative use of mail filters might solve that problem. It's a
pain to do that for most people, though. The user's mailer would
accept postings with the keyword, remove it, and display it for the
user. The user's postings would automatically have the keyword tacked
onto the end of the subject line if the keyword doesn't already exist
if the mail goes to cypherpunks@anything. That's probably far more
trouble than it's worth, considering that the alternative is to have
"CP" (or whatever) at the end of the line.
> 2. Some regular list posters may refuse to use it, so you'd have
> to check the non-keyworded messages anyway (or set up
> special filtering inclusion rules for them).
True. If someone refuses to use it, I can just exempt him. It gets
more complicated when arbitrary people reply to him, though.
> PS: Quite honestly, while the current level of spam is annoying, I filter
> it by hand pretty fast - once I realized that anything from 'sparklist'
> was garbage, the rest of its messages got deleted unread in a few seconds.
Same here. What I do is just read the legitimate list traffic, skip
the trash, taunt old Jeff, and then just do a catch-up on the mail
group. My mail client uses a scoring system, so I can assign point
values to various things (subject lines in all caps, kill things
beginning with "ADV: " which has the advantage of not killing replies
to the ads, etc.). The effect is cumulative. For example, I score Tim
May up, but that's cumulative with deductions for all-caps titles and
such. And unless the score winds up really low, I still see it, but at
the end of the list.
And like I said before (since various people won't stop harping on it), I
*know* good and well that Sparklist isn't guilty of "federal crimes."
Technically, they mailbombed the list, but their only "crime" was one
of blithe idiocy when they wrote their software. I was: a) Severely
annoyed, mostly because they had their site wide open for it, and
b) Trying to get the upstreams to pay *some* kind of attention to it.
(B was an incredibly dumb tactic, I admit. I hit myself with a
cluebat afterwards and chilled.)
The annoyance was due to the fact that I got hit with similar spam both
the day before and the day the list was hit. For the most part, what I
did was copy the stuff to abuse addresses. Two results were of
particular interest:
* Celebrity-link, I think it was called, responded and claimed that I
subscribed. I told them that I didn't, copied their upstream, and got
an almost immediate unsubscription.
* Quicken (Intuit) responded and claimed that "it may take up to 8
weeks" to stop sending me spam. I responded to that (firmly but
politely) and told them that it was ridiculous (obviously), but also
suggested various ways they could keep their lists from being used
in that manner and avoid most abuse complaints about it when they
are. That email was apparently circulated around, and they basically
said that they were working on some method of fixing the problem(s).
The rest either silently removed me, or sent a message saying that I
was removed, which did surprise the hell out of me. (That occured
after Sparklist bombed the list.)
I suppose that if any of us wanted to, we could set up some script
which goes around and grabs subscriber lists, merges them, adds the
remailer list, and scores down or kills postings from people not on
these lists. That would be something done on the user end or, I
suppose, at some CDR specifically designed for it. Jim Choate
could/should add such a CDR to his list of moderated CDRs if he
wanted, and it should be noted to subscribers that that CDR is
filtering in that way.
