Discussion:
Mitigating USENET spam clientside
(too old to reply)
tom
2020-04-24 03:27:54 UTC
Permalink
My USENET provider doesn't have the ability to remove spam messages and
I'm getting rather annoying by all the spam that is obviously spam
because it hits every single newgroup with the same message. It really
seems like ops should be able to catch those kinds of things.

Anyways, I wanted to ask if there was a better way to metigate USENET
spam in Claws-Mail besides manually making filtering rules by hand.

Thanks ahead of time for any tips
--
_______________________________________
/ He who always plows a straight furrow \
\ is in a rut. /
---------------------------------------
\
\
/\ /\
//\\_//\\ ____
\_ _/ / /
/ * * \ /^^^]
\_\O/_/ [ ]
/ \_ [ /
\ \_ / /
[ [ / \/ _/
_[ [ \ /_/
John McCue
2020-04-24 12:35:10 UTC
Permalink
Post by tom
My USENET provider doesn't have the ability to remove spam messages and
I'm getting rather annoying by all the spam that is obviously spam
because it hits every single newgroup with the same message. It really
seems like ops should be able to catch those kinds of things.
Looks like you are using 'eternal-september', it may
be best to ask this in one of their groups.
Post by tom
Anyways, I wanted to ask if there was a better way to metigate USENET
spam in Claws-Mail besides manually making filtering rules by hand.
I do not know of one. But I also noticed a uptick
in spam, but I think eternal-september is pretty good
dealing with it. My guess is the latest batch snuck
through and ES needs time to adjust their filters.
Post by tom
Thanks ahead of time for any tips
John
Grant Taylor
2020-04-24 15:17:26 UTC
Permalink
Post by tom
My USENET provider doesn't have the ability to remove spam messages
and I'm getting rather annoying by all the spam that is obviously
spam because it hits every single newgroup with the same message. It
really seems like ops should be able to catch those kinds of things.
SPAM on Usenet has a slightly different meaning than SPAM in email.

Usenet is more about abuse of the Usenet (NNTP) network than it is about
message content. This is more about the size of messages, binary
content where it shouldn't be, number of newsgroups, number of followup
groups, ratio of number of quoted lines to number of unquoted lines.

Usenet tends to not do anything with message content.

Some news servers / organizations will try to detect duplicate content,
but not all do.

An undesired message typically doesn't constitute SPAM in Usenet as it
typically does in email.
Post by tom
Anyways, I wanted to ask if there was a better way to metigate USENET
spam in Claws-Mail besides manually making filtering rules by hand.
I don't know about Claws-Mail.

You might be able to detect a pattern (posting host, etc) and filter
things based on that. But that's still creating a rule.

I'm not aware of much that can be done from a client point of view.
There are a few things that server operators can do. But they tend to
focus on the metadata about a message than the actual message data.
Post by tom
Thanks ahead of time for any tips
Work with your news server operator and see what they can do to help or
suggest you to do as a client.
--
Grant. . . .
unix || die
Lew Pitcher
2020-04-24 16:19:08 UTC
Permalink
Post by tom
My USENET provider doesn't have the ability to remove spam messages and
I'm getting rather annoying by all the spam that is obviously spam
because it hits every single newgroup with the same message. It really
seems like ops should be able to catch those kinds of things.
Anyways, I wanted to ask if there was a better way to metigate USENET
spam in Claws-Mail besides manually making filtering rules by hand.
Thanks ahead of time for any tips
I have no tested solution for you, but a suggestion instead.

In the past, I've run leafnode (https://sourceforge.net/projects/leafnode/)
as an intermediary NNTP service (a "caching news proxy", as the homepage
calls it).

Leafnode provides rudementary filtering using regular expressions, and (if
you feel up to it) you probably can hook in a spam filter such as
SpamAssassin or CleanFeed.

Hope this helps.
--
Lew Pitcher
"In Skills, We Trust"
Loading...