Jim S <
j...@jimXscott.co.uk> wrote:
> In 40tude Dialog is there a way to use the 'plonks' without opening
> the Scoring and actions window every time?
Add a toolbar button to do the plonk action. Right-click on the toolbar
(message list pane), and select Configure buttons. In the customize
dialog, scroll down the Categories list to and select Message. In the
right-side list for Commands, scroll down until you find the action you
want to assign to a button, like Add [subject] PLONK rule. Drag it to
the toolbar to create a button for it.
I've never cared for using Dialog's plonk rule. It does not narrowly
focus on a particular sender, so you end up with false positives:
posters getting plonked that you did not intend to flag. In addition,
you might plonk someone, but do you really want to watch all the replies
to that person get started as a separate thread? I have not tested if
the plonk rule also works with the "Settings -> General Settings -> Misc
-> Watch/Ignore works on subthread" setting. However, flagging a posts
as ignored doesn't hide them unless you choose a default view that Hides
Ignored posts (Group -> Message Views).
Note that I also configure 40tude to download ALL of a message, not just
the overview headers. Normally you have to click on a message in the
message list pane to get the entire message to download. Until then,
all you have available for testing in your rules are the overview
headers. Those are often insufficient to focus on just the unwanted
poster. What if they used "invalid <
inv...@invalid.com>" as their
command and address fields in their From header. LOTS of posters use
those as a generic identity. If you plonk on just the From header,
you'll plonk on everyone using a generic ID, including the good posters
that want to hide in Usenet by not deciding on a unique ID. A plonk
rule on "invalid <
inv...@invalid.com>" is far too vague to target just
who you wanted to filter out. By having 40tude download all of a
message (all headers + body), your filters can test on any header. For
example, I often include the injection node in the PATH header, and that
is not an overview header, so you won't have it to test on when getting
just the overview headers. "Groups -> Default Group options ->
Retrieving -> Retrieve bodies for all ..." has Dialog retrieve all
headers plus bodies, so your filters can test on any header and even for
strings in the body of messages. Can't do that when testing on just the
overview headers. I retrieve entire messages because I found rules
based on just overview headers are often too vague resulting in lots of
false positive (which I wouldn't see until I switched to Show All view).
Another problem of using simple plonk rules which are based only on the
From header is you will be plonking posts from a good sender that a
forger is joe jobbing. Likely the forger will be using a different NNTP
server, so testing on the injection node in the PATH header targets the
forger rather than falsely targets the poster getting forged. The
forger might be using a different NNTP client than the victim, and other
headers might identify the victim or better identify the forger. When I
define a filter, and because I do not delete the unwanted message but
just hide them, I can review the performance of my filters to see which
ones need more tweaking, or add more criteria to catch a poster that
nymshifts a lot. For examples: David In Devon aka David G. Brooks aka
David B[rooks] aka BoaterDave aka David_B aka ~BD~ aka BD1 aka BD2 aka
Beady aka imbeady2 and more, and variants he uses for each; and pooh the
cat who nymshifts by trying to vary his From header where he still wants
to be seen as pooh but tries to avoid simple filters, like replacing
"oo" with "00", and adding punctuation, like hyphens. Regex works well
to catch variants, but often needs to incorporate other headers, and
often more than just the overview headers.
After deciding on hiding an unwanted poster or types of posts, and with
several posts to use for defining on what to filter, I use a regex
(regular expression) filter (Settings -> Scoring and Actions)to try to
narrowly focus on just the unwanted poster or message type. Since I'm
focusing on a particular unwanted poster or message type, I don't use
scoring, and instead use a regex filter that marks read and flags as
ignored the message, like:
!setcolor(<fg>;<bg>),ignore,markread <regex>
I set the color of the message based on the type of filter, so when I
switch to the Show All view the color will indicate which [type of]
filter got used. With the message flagged as ignored, it won't show
with the default Hide Ignored messages view. However, occasionally I
want to see if my rules have false positives, so I select Show All to
check, and tweak my filters accordingly to better target just the
unwanted posters or messages. I couldn't do that if I deleted the
unwanted messages. Also, if you delete a message, the "Watch/Ignore
works on subthread" setting won't work, and you'll see disconnected
threads for replies to the unwanted messages. Sometimes someone
mentions they got help in one of the hidden subthreads, I want to see,
so I can easily switch to the Show All view to look.
Simple plonk rules are unfocused, don't accomodate nymshifting or sock
puppeting. For example, lots of trolls, malcontents, and forgers use
mixmin (aka sewer) NNTP servers. A plonk rule won't work on those
trolls, but a PATH header rule works very well, but you need to retrieve
ALL headers to have the PATH header available for a filter. Just some
of my anonymous/remailer/mixmin filters are:
# Ignore anonymous posts through remailers.
...
!setcolor(white;red),ignore,markread Header {^Path: \S+(\bmail.?2.?news.*|\b(mixmin\.net|dizum\.net|reece\.net\.au|tioat\.net|mail2news\.nymu\.eu|gweep\.ca|bananasplit\.info|darner\.ixazon\.lan|tahina\.priv\.at|pseudonym\.nl)(!?\.?POSTED)?(!not-for-mail)?|!usenet-gateway|!sewer-output.*)$}
!setcolor(white;red),ignore,markread Message-ID {\b(remail(er)?|mix(min|master)|cypher|dizum|firenze|gilgamesh|lib erty|tatoo|ccs1|tioat\.net|rip\.ax\.lt|paranoi|reece\.net\.au)}
!setcolor(white;red),ignore,markread From {\b(remail(er)?|mix(min|master|nym)|dizum|rip\.ax\.lt|ecn\.org|mailinator\.com|paranoi(\w|-)*|test)\b}
!setcolor(white;red),ignore,markread From {^(\W|_)*(Nomen\s*Nescio|George\s*Orwell|Anne\s*Onime|Fritz\s*Wuehler|Dave\s*U.*\s*Random)\b.*<}
!setcolor(white;red),ignore,markread From {<\S+@bigappleremailer}
!setcolor(white;red),ignore,markread Header {(?-s)^Return-Path: .*(<mixmaster|\bremailer\b)}
!setcolor(white;red),ignore,markread Header {^(Mail-To-News-Contact|X-MailConverter)}
!setcolor(white;red),ignore,markread Header {(?-s)^Organization: .*Bigapple\s*Mail2News}
!setcolor(white;red),ignore,markread Header {^In-Reply-To: <\S+@.*mixmin\.net>}
...
Some test on overview headers. Some test on non-overview headers. The
PATH filters are very good to get rid of those troll sources. I might
have to review the filters posts, or see what was said in an otherwise
filtered post, and why you'll notice my filters do an ignore action
instead of delete. Also, if I don't want to see unwanted posters, I
certainly don't want to see any replies to them, and why my Dialog is
configured to apply filters to subthreads, too.