somebody is flooding comp.lang.c with Thai spam

[Lynn McGuire]

Somebody is flooding comp.lang.c with Thai spam. Is there any way to
get rid of this ?

Thanks,
Lynn McGuire

[Dan Purgert]

On 2023-10-03, Lynn McGuire wrote:
> Somebody is flooding comp.lang.c with Thai spam. Is there any way to
> get rid of this ?

killfile the sender (and/or googlegroups in general).


--
|_|O|_|
|_|_|O| Github: https://github.com/dpurgert
|O|O|O| PGP: DDAB 23FB 19FA 7D85 1CC1 E067 6D65 70E5 4CE7 2860

[John McCue]

Dan Purgert <d...@djph.net> wrote:
> On 2023-10-03, Lynn McGuire wrote:
>> Somebody is flooding comp.lang.c with Thai spam. Is there any way to
>> get rid of this ?
>
> killfile the sender (and/or googlegroups in general).

Killing posts from googlegroups works great. I think
people should start doing that.

I use tin, and for the tin people here, all you need to
do is add these 4 lines to your ~/.tin/filter

group=*
case=0
score=kill
msgid_last=*<*@googlegroups.com>*

I mention that because tin changed their filtering a bit
and that is the only solution I could get working.

--
[t]csh(1) - "An elegant shell, for a more... civilized age."
- Paraphrasing Star Wars

[yeti]

Is there a chance that a petition to de-peer Google might succeed?

--
Fake signature.

[Daniel65]

Dan Purgert wrote on 4/10/2023 8:51 am:
> On 2023-10-03, Lynn McGuire wrote:
>> Somebody is flooding comp.lang.c with Thai spam. Is there any way to
>> get rid of this ?
>
> killfile the sender (and/or googlegroups in general).
>

Lynn posted from gmail.com

If she killfiles googlegroups will she, in effect, be killfiling herself??
--
Daniel

[Jack]

Lynn used a gmail address. She didn't use Google Groups to post. So the
answer is NO. Killfiling Google Groups means killfiling anybody using
"User-Agent: G2/1.0".

Hope this helps.

[Kaz Kylheku]

No. Using a From: identity with a Gmail address isn't related Google Groups.

A kill pattern targetting "@gmail.com" in the From header is possible,
but has nothing to do with blocking Google Groups.

That pattern will block numerous legitimate, non-spamming users that
don't use Google Groups, since it is a popular mail domain.

--
TXR Programming Language: http://nongnu.org/txr
Cygnal: Cygwin Native Application Library: http://kylheku.com/cygnal
Mastodon: @Kazi...@mstdn.ca
NOTE: If you use Google Groups, I don't see you, unless you're whitelisted.

[Marco Moock]

Am 04.10.2023 um 03:05:02 Uhr schrieb yeti:

> Is there a chance that a petition to de-peer Google might succeed?

I don't see one.
Although, you can encourage NNTP servers operators to block posts from
Google Groups at all at least in some groups that are unreadable.

[Dan Purgert]

Shouldn't -- I have googlegroups killed, and yet I saw Lynn's message.

In my case, I kill the googlegroups messages based on source message-id.
The rule (slrn) is essentially this (some comments / extra data injected
by slrn when it creates the rule has been removed for brevity)

%BOS
[*]
% ignore all from google-groups
Score: =-9999
Message-ID: .*@googlegroups\.com.*$
%EOS

One can also opt to kill them via injecting User Agent (IIRC it's
something along the lines of "GG/1.0")

[Daniel65]

Daniel65 wrote on 4/10/23 2:58 pm:

Thank you to the three of you who sorted my mis-understanding out!
--
Daniel

[Anton Shepelev]

Marco Moock

> Although, you can encourage NNTP servers operators to
> block posts from Google Groups at all at least in some
> groups that are unreadable.

An NNTP server may provide two kinds of service: a regular
one, and one with GG cut off (perhaps, except a white-list,
if the admin will bother maintaining it). The user will
then choose the second service by, for example, adding the
`NoGG:' prefix to the username with which he autheticates,
so that authenticating with NoGG:<username> gives him a GG-
free Usenet. Perhaps a non-standard port can be used for
the same purpose. Is there a simple (if hackish) solution
to filter out GG serverside for users who ask for it,
without modifying the software?

--
() ascii ribbon campaign -- against html e-mail
/\ www.asciiribbon.org -- against proprietary attachments

[Daniel65]

Anton Shepelev wrote on 5/10/23 12:35 am:

> Marco Moock
>
>> Although, you can encourage NNTP servers operators to
>> block posts from Google Groups at all at least in some
>> groups that are unreadable.
>
> An NNTP server may provide two kinds of service: a regular
> one, and one with GG cut off (perhaps, except a white-list,
> if the admin will bother maintaining it). The user will
> then choose the second service by, for example, adding the
> `NoGG:' prefix to the username with which he autheticates,
> so that authenticating with NoGG:<username> gives him a GG-
> free Usenet. Perhaps a non-standard port can be used for
> the same purpose. Is there a simple (if hackish) solution
> to filter out GG serverside for users who ask for it,
> without modifying the software?
>

Must be possible cause in one of my other UseNet groups, one guy, who
runs his own ISP business, has apparently filtered GoogleGroups from his
newsfeed .... and is asking those GG users who he likes to get
themselves another UseNet service so he can still see them!!

--
Daniel

[Marco Moock]

Am 05.10.2023 schrieb Daniel65 <dani...@nomail.afraid.org>:

> Must be possible cause in one of my other UseNet groups, one guy, who
> runs his own ISP business, has apparently filtered GoogleGroups from
> his newsfeed .... and is asking those GG users who he likes to get
> themselves another UseNet service so he can still see them!!

Groups that are flooded cannot be read with Google Groups anymore
anyway, so those users stop using those groups or get another
newsserver and newsreader with filter rules.

[Ray Banana]

* Daniel65 wrote:
> Anton Shepelev wrote on 5/10/23 12:35 am:

>> An NNTP server may provide two kinds of service: a regular
>> one, and one with GG cut off (perhaps, except a white-list,
>> if the admin will bother maintaining it). The user will
>> then choose the second service by, for example, adding the
>> `NoGG:' prefix to the username with which he autheticates,
>> so that authenticating with NoGG:<username> gives him a GG-
>> free Usenet. Perhaps a non-standard port can be used for
>> the same purpose. Is there a simple (if hackish) solution
>> to filter out GG serverside for users who ask for it,
>> without modifying the software?
>>
> Must be possible cause in one of my other UseNet groups, one guy, who
> runs his own ISP business, has apparently filtered GoogleGroups from his
> newsfeed .... and is asking those GG users who he likes to get
> themselves another UseNet service so he can still see them!!

Excluding all posts from GG from a newsserver is dead easy,
letting a user choose between GG free and GG versions would
require massive changes to the OVER, NEWNEWS, and ARTICLE
commands plus modifications to the OVERVIEW database.

Currently, I'm running all articles originating from GG
through a SpamAssassin server with a Usenet-specific
ruleset and reject all articles marked as spam.

The current flood of articles in Thai (ISO-639-1,ISO-639-2
and ISO-639-3) is, however, giving me headaches, especially as
they are in UTF-8 and base64 encoded.

The rejected articles are sent to a batcher that creates
NoCeM messages for the rejected articles and posts them
to news.lists.filters, so other newsservers can use them
to clean up their own servers.

On the same code path, I could just as well bluntly reject
every post from GG.

--
For every complex problem there is an answer that is clear, simple, and wrong.
H.L. Mencken

http://www.eternal-september.org

[Anton Shepelev]

Ray Banana:

> Excluding all posts from GG from a newsserver is dead
> easy, letting a user choose between GG free and GG
> versions would require massive changes to the OVER,
> NEWNEWS, and ARTICLE commands plus modifications to the
> OVERVIEW database.

Sounds like a big job. Or: deploy a separarate instance of
the server with the user dataabase shared with the original
and the article database synced from the original through a
filter...

> Currently, I'm running all articles originating from GG
> through a SpamAssassin server with a Usenet-specific
> ruleset and reject all articles marked as spam.

Makred as SPAM via GG? I haved marked some forty of them in
comp.lang.c yesterday, and coincidentally the flooding ook a
break several hours later.

> The current flood of articles in Thai (ISO-639-1,ISO-639-2
> and ISO-639-3) is, however, giving me headaches

Thanks for your service to the Free world, Ray!

[yeti]

Ray Banana <ray...@raybanana.net> writes:

> Currently, I'm running all articles originating from GG
> through a SpamAssassin server with a Usenet-specific
> ruleset and reject all articles marked as spam.

\o/

That may explain why I get no new food for my newest c.l.c score file
experiment?

I have looked at the Unicode block for Thai to find the numerical order
of the Thai chars and tried ...

| (("from"
| ("[ก-๛]" -1000 nil r))
| ("subject"
| ("[ก-๛]" -1000 nil r)))

... and it looks fine with the locally cached articles.

--
Take Back Control! -- Mesh The Planet!
I do not play Nethack, I do play GNUS! o;-)
Solid facts do not need 1001 pictures.

[yeti]

Ray Banana <ray...@raybanana.net> writes:

> Currently, I'm running all articles originating from GG
> through a SpamAssassin server with a Usenet-specific
> ruleset and reject all articles marked as spam.

[Jon Ribbens]

On 2023-10-05, Ray Banana <ray...@raybanana.net> wrote:
> Excluding all posts from GG from a newsserver is dead easy,
> letting a user choose between GG free and GG versions would
> require massive changes to the OVER, NEWNEWS, and ARTICLE
> commands plus modifications to the OVERVIEW database.
>
> Currently, I'm running all articles originating from GG
> through a SpamAssassin server with a Usenet-specific
> ruleset and reject all articles marked as spam.
>
> The current flood of articles in Thai (ISO-639-1,ISO-639-2
> and ISO-639-3) is, however, giving me headaches, especially as
> they are in UTF-8 and base64 encoded.

Would "reject all base64-encoded articles" not be a reasonable filter?

[A. Filip]

Lynn McGuire <lynnmc...@gmail.com> wrote:
> Somebody is flooding comp.lang.c with Thai spam. Is there any way to
> get rid of this ?

In similar case of pl.soc.polityka the spam was send via Google-Groups.
Reporting the spam via Google-Groups may force Google to move its heavy
ass after a few days of regular reports.

--
A. Filip

[Ray Banana]

* Jon Ribbens wrote:
>> The current flood of articles in Thai (ISO-639-1,ISO-639-2
>> and ISO-639-3) is, however, giving me headaches, especially as
>> they are in UTF-8 and base64 encoded.
>
> Would "reject all base64-encoded articles" not be a reasonable filter?

Been there, done that (in a test system). Too many false positives.

But I'm getting closer:

X-Spam-Status: No, score=9.0 required=10.0 tests=BAYES_99,BAYES_999,
CONTENT_BASE64,CONTENT_UTF8 autolearn=no autolearn_force=no
version=3.4.6

And the Bayes filter is now kicking in, too. It took more than 10 days
of learning before it finally started to show up in the spam reports.

--
Пу́тін — хуйло́
http://www.eternal-september.org

[Ray Banana]

I could inundate their abuse mailbox with ~5000 spam reports per day ;-),
but I suspect they would just blacklist my mail server.

[Adam H. Kerman]

Ray Banana <ray...@raybanana.net> wrote:

>>. . .

>Excluding all posts from GG from a newsserver is dead easy,
>letting a user choose between GG free and GG versions would
>require massive changes to the OVER, NEWNEWS, and ARTICLE
>commands plus modifications to the OVERVIEW database.

>Currently, I'm running all articles originating from GG
>through a SpamAssassin server with a Usenet-specific
>ruleset and reject all articles marked as spam.

>The current flood of articles in Thai (ISO-639-1,ISO-639-2
>and ISO-639-3) is, however, giving me headaches, especially as
>they are in UTF-8 and base64 encoded.

>The rejected articles are sent to a batcher that creates
>NoCeM messages for the rejected articles and posts them
>to news.lists.filters, so other newsservers can use them
>to clean up their own servers.

>On the same code path, I could just as well bluntly reject
>every post from GG.

I didn't realize how very much work you were doing till you mentioned
Spam Assassin the other day. Is it really necessary to decode the base64
first, or are you just relying on headers?

In any event, thank you.

You are doing this work to avoid rejecting the handful of non-flooded
articles from Google Groups.

[Adam H. Kerman]

A. Filip <an...@pseudonim.pl> wrote:

>>. . .

>In similar case of pl.soc.polityka the spam was send via Google-Groups.
>Reporting the spam via Google-Groups may force Google to move its heavy
>ass after a few days of regular reports.

Google does nothing with spam reports. Nothing. Don't bother.

[A. Filip]

Ignoring spam reports _completely_ may create long term legal risk for
any firm under US jurisdiction, doesn't it? In such case big+ google
may be a juicy+ target for punitive damages.

Long term perspective/risks "may create difference".

--
A. Filip

[Kaz Kylheku]

On 2023-10-05, Daniel65 <dani...@nomail.afraid.org> wrote:
> Must be possible cause in one of my other UseNet groups, one guy, who
> runs his own ISP business, has apparently filtered GoogleGroups from his
> newsfeed .... and is asking those GG users who he likes to get
> themselves another UseNet service so he can still see them!!

He could just look up the anonymized posting account strings from the
headers of the users he likes, and whitelist them.

[Kaz Kylheku]

On 2023-10-05, Ray Banana <ray...@raybanana.net> wrote:

> Excluding all posts from GG from a newsserver is dead easy,
> letting a user choose between GG free and GG versions would
> require massive changes to the OVER, NEWNEWS, and ARTICLE
> commands plus modifications to the OVERVIEW database.

This is a non-starter. Filtering GG for some users but not others
is much, much better done by the client, where a one-liner
entry in a kill file does the job.

Especially if you're going to do a blanket removal job with no
whitelisting.

Now if I had to implement the feature, I would not touch the
semantics of those commands with a ten foot editor.

I'd have it so that the server binds an alternative socket on
a different port. Clients would have to be capable of choosing
a non-default NNTP port.

Connections on that port would be flagged as not wanting to see
Google Groups posts.

The article processing logic would then somehow pretend, for the tagged
connections, that GG articles don't exist.

[Adam H. Kerman]

A. Filip <an...@pseudonim.pl> wrote:
>"Adam H. Kerman" <a...@chinet.com> wrote:
>>A. Filip <an...@pseudonim.pl> wrote:

>>>>. . .

>>>In similar case of pl.soc.polityka the spam was send via Google-Groups.
>>>Reporting the spam via Google-Groups may force Google to move its heavy
>>>ass after a few days of regular reports.

>>Google does nothing with spam reports. Nothing. Don't bother.

>Ignoring spam reports _completely_ may create long term legal risk for
>any firm under US jurisdiction, doesn't it?

No

>In such case big+ google may be a juicy+ target for punitive damages.

What are my ordinary damages?

[Lynn McGuire]

Excellent ! And thank you very much !

Lynn

[Ray Banana]

Thus spake Kaz Kylheku <864-11...@kylheku.com>

> This is a non-starter. Filtering GG for some users but not others
> is much, much better done by the client, where a one-liner
> entry in a kill file does the job.

I agree. But some users seem to have difficulties configuring
their newsreaders or their newsreaders are lacking the
required configuration options.

> Especially if you're going to do a blanket removal job with no
> whitelisting.

> Now if I had to implement the feature, I would not touch the
> semantics of those commands with a ten foot editor.

Neither would I. I'm using inn-CURRENT and would have to apply
my own patches every time I update the software.

> I'd have it so that the server binds an alternative socket on
> a different port. Clients would have to be capable of choosing
> a non-default NNTP port.
> Connections on that port would be flagged as not wanting to see
> Google Groups posts.

That would require a second newsserver (if you use INN, it's different
for Diablo, as it has a separate overview server).

> The article processing logic would then somehow pretend, for the tagged
> connections, that GG articles don't exist.

Again, this will not work for INN, as the user connects to the server
BEFORE they authenticate. ISTR that alphanet.ch offered such a choice
in their web frontend, as the user never directly connected to the
server. The appropriate backend was chosen in the web client *before*
the web client connected to the backend.

[Marco Moock]

Am 05.10.2023 schrieb "A. Filip" <an...@pseudonim.pl>:

> Ignoring spam reports _completely_ may create long term legal risk for
> any firm under US jurisdiction, doesn't it? In such case big+ google
> may be a juicy+ target for punitive damages.

Is there a case in that happened?

[Don Vito Martinelli]

Adam H. Kerman wrote:
> A. Filip <an...@pseudonim.pl> wrote:
>> "Adam H. Kerman" <a...@chinet.com> wrote:
>>> A. Filip <an...@pseudonim.pl> wrote:
>
>>>>> . . .
>
>>>> In similar case of pl.soc.polityka the spam was send via Google-Groups.
>>>> Reporting the spam via Google-Groups may force Google to move its heavy
>>>> ass after a few days of regular reports.
>
>>> Google does nothing with spam reports. Nothing. Don't bother.
>
>> Ignoring spam reports _completely_ may create long term legal risk for
>> any firm under US jurisdiction, doesn't it?
>
> No
>
>> In such case big+ google may be a juicy+ target for punitive damages.
>
> What are my ordinary damages?
>
>> Long term perspective/risks "may create difference".

Luckily that (hopefully) does not apply to eternal-september.test - it
is also being spammed to death at present.

[Ray Banana]

Thus spake Don Vito Martinelli <hyperspac...@vogon.gov.invalid>

> Luckily that (hopefully) does not apply to eternal-september.test - it
> is also being spammed to death at present.

No, it's finally used for the purpose it was created for ;-)

[Daniel65]

Adam H. Kerman wrote on 6/10/23 3:24 am:

This is all Ray does to earn his ZERO Euros from us all! ;-P (although
I've donated 20 Euro for each of the last few years!!

https://www.paypal.com/donate?token=_LMHRwmkO7vYeulNItMx8kuRADZEjDriNU91jFyxljBsiLwvNHpCl_97Avy_UwrTFxIlZinqsZnC3Jx6&locale.x=<?php>print

--
Daniel

[Daniel65]

Ray Banana wrote on 5/10/23 11:26 pm:

> * A. Filip wrote:
>> Lynn McGuire <lynnmc...@gmail.com> wrote:
>>> Somebody is flooding comp.lang.c with Thai spam. Is there any
>>> way to get rid of this ?
>>
>> In similar case of pl.soc.polityka the spam was send via
>> Google-Groups. Reporting the spam via Google-Groups may force
>> Google to move its heavy ass after a few days of regular reports.
>
> I could inundate their abuse mailbox with ~5000 spam reports per day
> ;-), but I suspect they would just blacklist my mail server.
>

Hmm! If your server is busy firing off those ~5000 spam reports each
day, what might be the impact on my receiving the messages that I want??
A couple of seconds?? ;-P
--
Daniel

[Ray Banana]

Thus spake Daniel65 <dani...@nomail.afraid.org>

>> You are doing this work to avoid rejecting the handful of non-flooded
>> articles from Google Groups.

No, I'm actually doing this because I would like to learn more about
the capabilities of the more recent versions of SpamAssassin.
After having used it in several mail servers for decades, I'm finally
getting to leverage its full potential.

> This is all Ray does to earn his ZERO Euros from us all! ;-P (although
> I've donated 20 Euro for each of the last few years!!

Thank you, you are really generous ;-)

[Ray Banana]

* Daniel65 wrote:
> Ray Banana wrote on 5/10/23 11:26 pm:

>> I could inundate their abuse mailbox with ~5000 spam reports per day
>> ;-), but I suspect they would just blacklist my mail server.
>>
> Hmm! If your server is busy firing off those ~5000 spam reports each
> day, what might be the impact on my receiving the messages that I want??
> A couple of seconds?? ;-P

None at all. It's a different server. And I used to send ~15.000 mails
to all registered E-S users in just a couple of minutes back in the days
when Usenet was still thriving (and E-S was called Motzarella or even
Banana).

[Daniel65]

Ray Banana wrote on 6/10/23 8:23 pm:

Ah, O.K., good! ;-)
--
Daniel

[Daniel65]

Ray Banana wrote on 6/10/23 7:37 pm:

Back a few years ago, me giving 20 Euros WAS very generous ....
Nowadays, not so much!!

--
Daniel

[Anton Shepelev]

Kaz Kylheku:

> Now if I had to implement the feature, I would not touch
> the semantics of those commands with a ten foot editor.
>
> I'd have it so that the server binds an alternative socket
> on a different port. Clients would have to be capable of
> choosing a non-default NNTP port.

Not all are, though. Another option from the top of my
balding head would be to provide two versions of each group,
so that comp.lang.c.no_gg would be GG-free (except
whiltelisted posters, if any). Ugly, if at all techinally
feasible...

> Connections on that port would be flagged as not wanting
> to see Google Groups posts.
>
> The article processing logic would then somehow pretend,
> for the tagged connections, that GG articles don't exist.

I was proposing the same thing.

As for whitelisting, it must be a distributed, /community/
job, where verified humans shall review "pending" GG posts
and mark senders as SPAMmers or legitimate posters.

[Retro Guy]

On Fri, 06 Oct 2023 10:37:13 +0200
Ray Banana <ray...@raybanana.net> wrote:

> Thus spake Daniel65 <dani...@nomail.afraid.org>
>
> >> You are doing this work to avoid rejecting the handful of non-flooded
> >> articles from Google Groups.
>
> No, I'm actually doing this because I would like to learn more about
> the capabilities of the more recent versions of SpamAssassin.
> After having used it in several mail servers for decades, I'm finally
> getting to leverage its full potential.

I've been doing the same for i2pn2.org for the past 6 weeks or so and I
agree, it's a very interesting challenge using spamassassin for Usenet.

I've learned a lot more than I used to know about creative rules, and it's
a good learning experience. Enjoy!

Also, I was happy to see e-s taking on the spam also. The more servers that
attack the problem, the more pleasant Usenet will be for the users. You're
doing a great job!

--
Retro Guy

[Sn!pe]

Yes indeed. Your users thank you both,
and the others fighting the good fight too.

--
^Ï^. Sn!pe <https://youtu.be/_kqytf31a8E>

My pet rock Gordon just is.

[Ray Banana]

* Retro Guy wrote:

> I've learned a lot more than I used to know about creative rules, and it's
> a good learning experience. Enjoy!

I can now generate automatic rules from Google spam, filtering on the Google
posting-account in Injection-Info.

> Also, I was happy to see e-s taking on the spam also. The more servers that
> attack the problem, the more pleasant Usenet will be for the users. You're
> doing a great job!

BTW, I just looked at the Google Groups web interface and it looks as if
comp.lang.c had vanished from Google Groups. Can anyone confirm this?

eternal-september.test is still available, so I can test as much as I want.

[Andy Burns]

Ray Banana wrote:

> BTW, I just looked at the Google Groups web interface and it looks as if
> comp.lang.c had vanished from Google Groups. Can anyone confirm this?

No it's still there, receiving a heavy amount of the Thai spam, about an
hour ago ...

[Retro Guy]

On Fri, 6 Oct 2023 14:19:39 -0000 (UTC)
Ray Banana <ray...@raybanana.net> wrote:

> * Retro Guy wrote:
>
> > I've learned a lot more than I used to know about creative rules, and it's
> > a good learning experience. Enjoy!
>
> I can now generate automatic rules from Google spam, filtering on the Google
> posting-account in Injection-Info.

Nice. I am doing similar. My spamassassin script (the way I connect cleanfeed to
spamc) allows grepping whatever from the messages and automatically working with
the info. Automation is the way to go as manually handling each spam can be
overwhelming, and not really necessary.

I'm also using 'seek-phrases-in-corpus' from spamassassin. It does a pretty good
job of comparing a dir of spam with a dir of ham and finding patterns only in the
spam. I'm currently leaning heavily toward patterns in spam and it seems to be
working well.

--
Retro Guy

[Richard Damon]

Google Groups has at times (temporarily) dropped Usenet groups that
triggered too many spam alerts. Their software tries to contact the list
"moderator" to warn them of the problem, and when it can find it, it
suspends the list on Google Groups. Often they get around to putting it
back.

(Not realizing THEY are the source of the spam).

[candycanearter07]

On 10/6/23 18:36, Richard Damon wrote:
> (Not realizing THEY are the source of the spam).

Oh the irony
--
user <candycane> is generated from /dev/urandom

[Andy Burns]

Richard Damon wrote:

> Google Groups has at times (temporarily) dropped Usenet groups that
> triggered too many spam alerts. Their software tries to contact the list
> "moderator" to warn them of the problem, and when it can find it, it
> suspends the list on Google Groups. Often they get around to putting it
> back.

They can't get their heads around the difference between a googlegroup
and a usenet group, and the fact that there isn't anyone "in charge" of
unmoderated groups, when they remove a group from googlegroups (e.g.
uk.telecom) access to search the archives is lost.

> (Not realizing THEY are the source of the spam).

s/realizing/caring

[Anton Shepelev]

Ray Banana:

> Excluding all posts from GG from a newsserver is dead
> easy, letting a user choose between GG free and GG
> versions would require massive changes to the OVER,
> NEWNEWS, and ARTICLE commands plus modifications to the
> OVERVIEW database.

Tiny amounts of SPAM are leaking into comp.lang.c :

<d1ef8d1e-3fdd-4484...@googlegroups.com>

<54d3cfc5-bd8c-44e4...@googlegroups.com>
<d61dd7c6-7afe-4fa4...@googlegroups.com>
<702d5b9d-9488-490a...@googlegroups.com>
<ca58ba3e-c84c-40ad...@googlegroups.com>

[Nuno Silva]

On 2023-10-07, Anton Shepelev wrote:

> Tiny amounts of SPAM are leaking into comp.lang.c :

comp.text.tex seems to be a target too now :-\

--
Nuno Silva

[Daniel65]

Anton Shepelev wrote on 7/10/23 9:10 pm:

> Ray Banana:
>
>> Excluding all posts from GG from a newsserver is dead
>> easy, letting a user choose between GG free and GG
>> versions would require massive changes to the OVER,
>> NEWNEWS, and ARTICLE commands plus modifications to the
>> OVERVIEW database.
>
> Tiny amounts of SPAM are leaking into comp.lang.c :
>
> <d1ef8d1e-3fdd-4484...@googlegroups.com>
>
> <54d3cfc5-bd8c-44e4...@googlegroups.com>
> <d61dd7c6-7afe-4fa4...@googlegroups.com>
> <702d5b9d-9488-490a...@googlegroups.com>
> <ca58ba3e-c84c-40ad...@googlegroups.com>
>

Hmm!! In one of my other NGs, the chief poster there (some might,
rightly, suggest the "chief SPAMMER there") is bitching about the
spammers on comp.lang.c too!!

Finally, he's right about something!!
--
Daniel

[Adam H. Kerman]

Andy Burns <use...@andyburns.uk> wrote:
>Richard Damon wrote:

>>Google Groups has at times (temporarily) dropped Usenet groups that
>>triggered too many spam alerts. Their software tries to contact the list
>>"moderator" to warn them of the problem, and when it can find it, it
>>suspends the list on Google Groups. Often they get around to putting it
>>back.

>They can't get their heads around the difference between a googlegroup
>and a usenet group, and the fact that there isn't anyone "in charge" of
>unmoderated groups, when they remove a group from googlegroups (e.g.
>uk.telecom) access to search the archives is lost.

The interface doesn't add a Newsgroups header when posting to a Google
Group, does it? Seems like a readily detectable difference to me.

>>(Not realizing THEY are the source of the spam).

>s/realizing/caring

Right

[Andy Burns]

Adam H. Kerman wrote:

> The interface doesn't add a Newsgroups header when posting to a Google
> Group, does it? Seems like a readily detectable difference to me.

Apparently it does

> X-Received: by 2002:ae9:e217:0:b0:774:299d:9a21 with SMTP id c23-20020ae9e217000000b00774299d9a21mr75519qkc.9.1696528672931;
> Thu, 05 Oct 2023 10:57:52 -0700 (PDT)
> X-Received: by 2002:a05:6830:1b67:b0:6be:ed66:5ab0 with SMTP id
> d7-20020a0568301b6700b006beed665ab0mr1688693ote.6.1696528672656; Thu, 05 Oct
> 2023 10:57:52 -0700 (PDT)
> Path: uni-berlin.de!fu-berlin.de!news-out.google.com!nntp.google.com!postnews.google.com!google-groups.googlegroups.com!not-for-mail
> Newsgroups: uk.d-i-y
> Date: Thu, 5 Oct 2023 10:57:52 -0700 (PDT)
> In-Reply-To: <ufmr1t$11ug5$1...@dont-email.me>
> Injection-Info: google-groups.googlegroups.com; posting-host=2001:8b0:fb4e:0:2a36:65a2:2f89:51f7;
> posting-account=de11ZAoAAACBQRb2jWnaIkHYK2q9mRvs
> NNTP-Posting-Host: 2001:8b0:fb4e:0:2a36:65a2:2f89:51f7
> References: <FIOcnZ9guMA2gDb5...@brightview.co.uk>
> <ufm41o$teru$1...@dont-email.me> <ufm7m2$u0ma$1...@dont-email.me>
> <ufme25$upme$2...@dont-email.me> <ufmr1t$11ug5$1...@dont-email.me>
> User-Agent: G2/1.0
> MIME-Version: 1.0
> Message-ID: <e381b1f7-8c30-44f7...@googlegroups.com>
> Subject: Re: Time for a re-re-think about PSTN/POTS? - Update
> From: John Walliker <jrwal...@gmail.com>
> Injection-Date: Thu, 05 Oct 2023 17:57:52 +0000

[Kaz Kylheku]

Well, that will be a shitshow; those people have a fit
at a mere "overfull \hbox".

--
TXR Programming Language: http://nongnu.org/txr
Cygnal: Cygwin Native Application Library: http://kylheku.com/cygnal
Mastodon: @Kazi...@mstdn.ca
NOTE: If you use Google Groups, I don't see you, unless you're whitelisted.

[Richard Damon]

On 10/7/23 12:54 PM, Adam H. Kerman wrote:
> The interface doesn't add a Newsgroups header when posting to a Google
> Group, does it? Seems like a readily detectable difference to me.

You can't propigate a message on USENET without a Newsgroup header, it
is basically the equivalent of the "To:" line of an email (without the
CC: and BCC: alternatives).

My guess is it gets added "near the edge", in the NNTP connection, but
much of the GG processing doesn't understand the difference.

[Adam H. Kerman]

There's no To header!

How does that work as an email message?

[Adam H. Kerman]

Richard Damon <news.x.ri...@xoxy.net> wrote:
>On 10/7/23 12:54 PM, Adam H. Kerman wrote:

>>The interface doesn't add a Newsgroups header when posting to a Google
>>Group, does it? Seems like a readily detectable difference to me.

>You can't propigate a message on USENET without a Newsgroup header, it
>is basically the equivalent of the "To:" line of an email (without the
>CC: and BCC: alternatives).

Newsgroups header, not Newsgroup

A Google Group is a mailing list.

>My guess is it gets added "near the edge", in the NNTP connection, but
>much of the GG processing doesn't understand the difference.

What NNTP connection? It's not a newsgroup.

[Anton Shepelev]

The spammers is laughing at us?

----------------------------------------------------------------------
Subject: hee heee
From: nu02...@gmail.com:
Newsgroups: comp.lang.c
Message-ID: <9f2af305-8f71-4748...@googlegroups.com>

adasd

[candycanearter07]

On 10/7/23 14:06, Richard Damon wrote:
> You can't propigate a message on USENET without a Newsgroup header, it
> is basically the equivalent of the "To:" line of an email (without the
> CC: and BCC: alternatives).

IMO, the Followup-To field sorta acts as a reverse CC, as in not adding
the group to the Followup-To makes it the CC.

[Richard Damon]

On 10/7/23 5:31 PM, candycanearter07 wrote:
> On 10/7/23 14:06, Richard Damon wrote:
>> You can't propigate a message on USENET without a Newsgroup header, it
>> is basically the equivalent of the "To:" line of an email (without the
>> CC: and BCC: alternatives).
>
> IMO, the Followup-To field sorta acts as a reverse CC, as in not adding
> the group to the Followup-To makes it the CC.

Followup-To is the equivalent of Reply-To for email, it points where you
would prefer to have replies (called followups for usenet) go to

[Richard Damon]

On 10/7/23 4:39 PM, Adam H. Kerman wrote:
> Richard Damon <news.x.ri...@xoxy.net> wrote:
>> On 10/7/23 12:54 PM, Adam H. Kerman wrote:
>
>>> The interface doesn't add a Newsgroups header when posting to a Google
>>> Group, does it? Seems like a readily detectable difference to me.
>
>> You can't propigate a message on USENET without a Newsgroup header, it
>> is basically the equivalent of the "To:" line of an email (without the
>> CC: and BCC: alternatives).
>
> Newsgroups header, not Newsgroup
>
> A Google Group is a mailing list.

Unless it is a Usenet group gatewayed, the it is both. (or all three, as
Google Groups are also Forum Based interactions, as you don't need to
acutally use email either.

I would have to see if it is even actually possible to "subscribe" to
getting email deliveries for messages gatewayed from usenet on Google Groups

>
>> My guess is it gets added "near the edge", in the NNTP connection, but
>> much of the GG processing doesn't understand the difference.
>
> What NNTP connection? It's not a newsgroup.

Google Groups DOES have Usenet newsgroups on it, via its gateway.

[Adam H. Kerman]

It's nothing of the kind. Reply-To may or may not be used for "reply to
group". It might be used for personal communication. Therefore it is NOT
equivalent.

Followup-To is used to express one's personal hypocrisy. "I'm posting
off topic, deliberately, but I'm directing YOU to post somewhere else so
I don't have to read your crappy followup."

Followup-To is meant to annoy everyone else. Its use is always a troll.

[Adam H. Kerman]

Richard Damon <news.x.ri...@xoxy.net> wrote:
>On 10/7/23 4:39 PM, Adam H. Kerman wrote:
>>Richard Damon <news.x.ri...@xoxy.net> wrote:
>>>On 10/7/23 12:54 PM, Adam H. Kerman wrote:

>>>>The interface doesn't add a Newsgroups header when posting to a Google
>>>>Group, does it? Seems like a readily detectable difference to me.

>>>You can't propigate a message on USENET without a Newsgroup header, it
>>>is basically the equivalent of the "To:" line of an email (without the
>>>CC: and BCC: alternatives).

>>Newsgroups header, not Newsgroup

>>A Google Group is a mailing list.

>Unless it is a Usenet group gatewayed, the it is both. (or all three, as
>Google Groups are also Forum Based interactions, as you don't need to
>acutally use email either.

Again: Not discussing a newsgroup but a mailing list. The complaint,
which was in the part of the quote that you didn't retain, concerned
Google's seeming inability to distinguish its reaction to a spam report
whether it's a newsgroup or a mailing list.

>I would have to see if it is even actually possible to "subscribe" to
>getting email deliveries for messages gatewayed from usenet on Google Groups

Not discussing a gated newsgroup.

>>>My guess is it gets added "near the edge", in the NNTP connection, but
>>>much of the GG processing doesn't understand the difference.

>>What NNTP connection? It's not a newsgroup.

>Google Groups DOES have Usenet newsgroups on it, via its gateway.

We are not discussing newsgroups offered on Google. We are discussing a
complaint that Google treats a spam report received on a newsgroup by
attempting to complain to the Google Group owner, even though no
newsgroup has the concept of "owner" comparable to "mailing list owner".
If Google doesn't get an answer from the nonexistent owner, then the
archive of News articles is removed and made inaccessible.

I then pointed out the obvious that because the set of headers are
different and therefore it's pretty damn straightforward to figure out
if the "group" subject of a spam complaint is Usenet or a mailing list.

We are discussion Google Groups which are not newsgroups. I know it's
confusing because Google uses the same term for unrelated concepts, but
come on, I've been saying throughout "mailing list" so that should have
been clear to you from context.

[Richard Damon]

On 10/7/23 6:08 PM, Adam H. Kerman wrote:
> Richard Damon <news.x.ri...@xoxy.net> wrote:
>> On 10/7/23 4:39 PM, Adam H. Kerman wrote:
>>> Richard Damon <news.x.ri...@xoxy.net> wrote:
>>>> On 10/7/23 12:54 PM, Adam H. Kerman wrote:
>
>>>>> The interface doesn't add a Newsgroups header when posting to a Google
>>>>> Group, does it? Seems like a readily detectable difference to me.
>
>>>> You can't propigate a message on USENET without a Newsgroup header, it
>>>> is basically the equivalent of the "To:" line of an email (without the
>>>> CC: and BCC: alternatives).
>
>>> Newsgroups header, not Newsgroup
>
>>> A Google Group is a mailing list.
>
>> Unless it is a Usenet group gatewayed, the it is both. (or all three, as
>> Google Groups are also Forum Based interactions, as you don't need to
>> acutally use email either.
>
> Again: Not discussing a newsgroup but a mailing list. The complaint,
> which was in the part of the quote that you didn't retain, concerned
> Google's seeming inability to distinguish its reaction to a spam report
> whether it's a newsgroup or a mailing list.

Right, (I started that part). The issue is that the algorithm for
dropping "spam overrun Google Group Mailing Lists" doesn't understand
that some of the Google Groups aren't actually Mailing lists, and don't
have a "moderator" to contact to warn that they need to handle their
spamming problem.

>
>> I would have to see if it is even actually possible to "subscribe" to
>> getting email deliveries for messages gatewayed from usenet on Google Groups
>
> Not discussing a gated newsgroup.

comp.lang.c isn't a gated newsgroup? Maybe YOU lost the context

>
>>>> My guess is it gets added "near the edge", in the NNTP connection, but
>>>> much of the GG processing doesn't understand the difference.
>
>>> What NNTP connection? It's not a newsgroup.
>
>> Google Groups DOES have Usenet newsgroups on it, via its gateway.
>
> We are not discussing newsgroups offered on Google. We are discussing a
> complaint that Google treats a spam report received on a newsgroup by
> attempting to complain to the Google Group owner, even though no
> newsgroup has the concept of "owner" comparable to "mailing list owner".
> If Google doesn't get an answer from the nonexistent owner, then the
> archive of News articles is removed and made inaccessible.
>
> I then pointed out the obvious that because the set of headers are
> different and therefore it's pretty damn straightforward to figure out
> if the "group" subject of a spam complaint is Usenet or a mailing list.
>
> We are discussion Google Groups which are not newsgroups. I know it's
> confusing because Google uses the same term for unrelated concepts, but
> come on, I've been saying throughout "mailing list" so that should have
> been clear to you from context.

Read what you just said. We are discussing that Google (incorrectly)
treats its Usenet gated newsgroups the same as its actual mailing lists.
It gets spam reports for a Google Group that happens to be a Gated
Usenet newsgroup, and at some point it can trigger the warning to the
"moderator" (which doesn't exist) and when it can't contact them and get
a response, it considers it "dead" and kills the archived feed, because
it doesn't understand that the gated newsgroup isn't actually a mailing
list, and no moderator can be made resposible for the spam.

This is particularly ironic, when the majority of the spam was actually
injected by Google Groups, so THEY are the ones that should have taken
responsibility.

[candycanearter07]

On 10/7/23 16:56, Adam H. Kerman wrote:
> Followup-To is used to express one's personal hypocrisy. "I'm posting
> off topic, deliberately, but I'm directing YOU to post somewhere else so
> I don't have to read your crappy followup."
>
> Followup-To is meant to annoy everyone else. Its use is always a troll.

Huh. I always saw Followup-To as a way to remove a group from a
conversation, by setting the Followup-To to not include one of the
groups it was posted to. It also could be used for moderated
announcement-type groups, where replying to an announcement posts to a
talk group (look at news.announce.newgroups)

also what is alt.dev.null?

[Richard Damon]

/dev/null is a "standard" device that just eats data sent to it

Followup-To: alt.dev.null was a 'joke' line to say that followups to the
message would be ignored.

I believe alt.dev.null was just created by someone to make such a joke,
as a place to send data that was intended to just be ignored. Since
there were no 'rules' about creating alt groups, you just needed to get
system admins to carry the group (and many would if a control message
had been sent, and anyone could send such a control message) it would be
hard to do that.

[Adam H. Kerman]

My little joke demonstrating that setting Followup-To is a troll.

It's a self-moderated newsgroup where everything is off topic.

[candycanearter07]

On 10/7/23 20:10, Richard Damon wrote:
> /dev/null is a "standard" device that just eats data sent to it
>
> Followup-To: alt.dev.null was a 'joke' line to say that followups to the
> message would be ignored.

I meant, is alt.dev.null a real group or does it just not actually post
messages sent to it

[Adam H. Kerman]

candycanearter07 <n...@thanks.net> wrote:
>On 10/7/23 20:10, Richard Damon wrote:

>>/dev/null is a "standard" device that just eats data sent to it

/dev/nul

>>Followup-To: alt.dev.null was a 'joke' line to say that followups to the
>>message would be ignored.

>I meant, is alt.dev.null a real group or does it just not actually post
>messages sent to it

It's real. On rare occasion, it's used for discussion.

[candycanearter07]

On 10/8/23 02:17, Adam H. Kerman wrote:
> candycanearter07 <n...@thanks.net> wrote:
>> On 10/7/23 20:10, Richard Damon wrote:
>
>>> /dev/null is a "standard" device that just eats data sent to it
>
> /dev/nul

$ ls /dev/nul*

/dev/null

[Adam H. Kerman]

Adam H. Kerman <a...@chinet.com> wrote:
>candycanearter07 <n...@thanks.net> wrote:
>>On 10/7/23 20:10, Richard Damon wrote:

>>>/dev/null is a "standard" device that just eats data sent to it

>/dev/nul

/dev/null

Never mind

[Andy Burns]

Adam H. Kerman wrote:

> There's no To header!
> How does that work as an email message?

Why does it need to? Maybe google are just showing some of the frantic
paddling that goes on under the waterline, it might just use the
envelope header to let it reach some sort of mail->news gateway, then
post it to usenet ... I don't know

[Andy Burns]

Adam H. Kerman wrote:

> We are not discussing newsgroups offered on Google.

I thought "we" were, I certainly was, sorry for any confusion ...

[Sn!pe]

Adam H. Kerman <a...@chinet.com> wrote:

Yes.*

It's nicknamed "the bit-bucket" because, generally speaking,
posts to there just silently disappear.

* Vs jr gryy gurz ubj, gura jr jbhyq unir gb xvyy gurz.

--
^Ï^. Sn!pe <https://youtu.be/_kqytf31a8E>

My pet rock Gordon just is.

[Anton Shepelev]

A new batch of Thai SPAM is materialising in comp.lang.c right now,
and E.-S. does not fail to see it.

[Adam H. Kerman]

Andy Burns <use...@andyburns.uk> wrote:
>Adam H. Kerman wrote:

>>There's no To header!
>>How does that work as an email message?

>Why does it need to?

A Google Group (not a newsgroup) is literally an email mailing list. If
the example you gave isn't a newsgroup, then the messages are
distributed to subscribers, at least among those not reading it through
the Web interface.

>Maybe google are just showing some of the frantic
>paddling that goes on under the waterline, it might just use the
>envelope header to let it reach some sort of mail->news gateway, then
>post it to usenet ... I don't know

Maybe email messages are stored on a News server, then offered through
the common Web interface. Still you'd think it would store them with the
required headers for email.

[Adam H. Kerman]

Google uses the term "Google Groups" confusingly as it applies to both
the common Web interface between mailing lists and newsgroups, and to the
mailing lists themselves. There was a complaint that mailing lists were
being named into Usenet namespace, which confused Google as to how spam
reports were being handled, as Google was sending spam reports to the
"Google Group owner". Well, mailing lists have "owners" but that's not
a newsgroup concept.

(Fluffy owns Usenet.)

Your example was uk.d-i-y. I see now that's a newsgroup and not a Google
Group.

[Richard Damon]

On 10/8/23 1:49 AM, candycanearter07 wrote:
> On 10/7/23 20:10, Richard Damon wrote:
>> /dev/null is a "standard" device that just eats data sent to it
>>
>> Followup-To: alt.dev.null was a 'joke' line to say that followups to
>> the message would be ignored.
>
> I meant, is alt.dev.null a real group or does it just not actually post
> messages sent to it

It is a real newsgroup. I think it is self-moderated so most messages
just sent to it disappear, but if you know how to "approve" messages,
you can post to it, and it has some traffic.

[Anton Shepelev]

[Followup-To: alt.free.newsservers]

Retro Guy:

> I've been doing the same for i2pn2.org for the past 6
> weeks or so and I agree, it's a very interesting challenge
> using spamassassin for Usenet.

Your server seems really good at filtering this recent Thai
SPAM. I have checked CLC and several other groups -- and
there is very little or no SPAM at all, whereas normal GG
articles are present. Great job!

If I propose your server as the only (due to software
limitations) peer for a Usenet-to-FidoNet gateway I use,
will it be getting the same messages as a newsclient does,
that is already cleared of SPAM?

[Adam H. Kerman]

Anton Shepelev <anto...@gmail.moc> wrote:

>[Followup-To: alt.free.newsservers]

No, I'm not fucking splitting the fucking thread. What is it with
everybody? If you want to post in a different newsgroup, START A NEW
THREAD OF YOUR OWN. It's better to keep an existing thread together
without changing crossposts and without setting Followup-To. As a
reader, I want to be able to readily find precursor articles and
articles in other branches of the thread tree in the newsgroup I'm
currently looking at.

>Retro Guy:

>>I've been doing the same for i2pn2.org for the past 6
>>weeks or so and I agree, it's a very interesting challenge
>>using spamassassin for Usenet.

>Your server seems really good at filtering this recent Thai
>SPAM. I have checked CLC and several other groups -- and
>there is very little or no SPAM at all, whereas normal GG
>articles are present. Great job!

>If I propose your server as the only (due to software
>limitations) peer for a Usenet-to-FidoNet gateway I use,
>will it be getting the same messages as a newsclient does,
>that is already cleared of SPAM?

Shouldn't you run your own News server with multiple peers that is
itself the Usenet-to-FidoNet gateway?

Then you can process Ray's NoCeMs as it appears that Ray is interested
in publishing them indefinetely or until Google Groups stops originating
Usenet spam.

[Anton Shepelev]

Adam H. Kerman to Anton Shepelev:

> > >[Followup-To: alt.free.newsservers]
>
> No, I'm not fucking splitting the fucking thread. What is
> it with everybody? If you want to post in a different
> newsgroup, START A NEW THREAD OF YOUR OWN. It's better to
> keep an existing thread together without changing
> crossposts and without setting Followup-To.

What am I to do if a discussion gradually divagates into an
off-topic suject as it did here? We were discussing Ray's
handling of SPAM, but then another server's admin chimes in,
and suddently I have a question for him that is promted by
connected with this discussion? I felt it imortant to let
readers here know that i2pn2 is successfully filtering this
SPAM, and also let them know where the discussion is
continued. Is not the standard Followup-To: header created
for this purpose of gracul migration of a tread due to
subject drift?

Shall I keep drop a follow-up notice in the message body,
and post an entirely new thread in the the on-topic group?

> As a reader, I want to be able to readily find precursor
> articles and articles in other branches of the thread tree
> in the newsgroup I'm currently looking at.

Yet other readers may be annoyned by discussion of other
Usenet servers in a group about E.-S.

> > If I propose your server as the only (due to software
> > limitations) peer for a Usenet-to-FidoNet gateway I use,
> > will it be getting the same messages as a newsclient
> > does, that is already cleared of SPAM?
>
> Shouldn't you run your own News server with multiple peers
> that is itself the Usenet-to-FidoNet gateway?
>
> Then you can process Ray's NoCeMs as it appears that Ray
> is interested in publishing them indefinetely or until
> Google Groups stops originating Usenet spam.

That is an option, but with my skills it will be a
tremedously difficult task. I am glad to have convinced the
administrator of the gateway with a single Usenet peer, to
swtich to i2pn2 (not from E.-S., mind you).

[Adam H. Kerman]

Anton Shepelev <anto...@gmail.moc> wrote:
>Adam H. Kerman to Anton Shepelev:

No, I wasn't addressing you personally. I was addressing Usenet readers
generally.

>>>>[Followup-To: alt.free.newsservers]

>>No, I'm not fucking splitting the fucking thread. What is
>>it with everybody? If you want to post in a different
>>newsgroup, START A NEW THREAD OF YOUR OWN. It's better to
>>keep an existing thread together without changing
>>crossposts and without setting Followup-To.

>What am I to do if a discussion gradually divagates into an
>off-topic suject as it did here?

START A NEW THREAD. START IT IN A DIFFERENT NEWSGROUP IF YOU THINK IT'S
OFF TOPIC IN THE GROUP YOU A READING.

This Usenet shit isn't difficult. If you think the followup YOU are
about to post is off topic, then DON'T post it. Start a new thread where
you think it's on topic.

Don't screw up ongoing threads. There's no need to inflict the middle of
a thread upon the readers of another newsgroup when the right solution
couldn't be any simpler: Start a new thread.

>. . .

[Sn!pe]

Adam H. Kerman <a...@chinet.com> wrote:

A suggestion: why not change the subject thus -- New Subject (Was: Old
Subject); crosspost it to the new group and set Followup-To that group?

That would effectively begin a new thread with a new Subject in the new
group but maintains the References to the old thread in the old group
for those who want to see what went before.

Those who don't want to follow the branched thread with the new Subject
can simply ignore it because the old discussion would be off-topic in
the new group.

[Adam H. Kerman]

I'm literally arguing AGAINST that very thing. DO NOT add a newsgroup in
the middle of a thread, 'cuz the people reading the thread in the added
newsgroup won't see the precursor articles. The people reading earlier
articles in the thread won't see followups.

It's a typical tactic of topic cops. It's always hypocritical.

>That would effectively begin a new thread with a new Subject in the new
>group but maintains the References to the old thread in the old group
>for those who want to see what went before.

There aren't going to be all that many newsreaders designed to be able
to fetch the precursor articles from a different newsgroup, nor would
any newsreader even know how to find a followup article posted to a different
newsgroup.

>Those who don't want to follow the branched thread with the new Subject
>can simply ignore it because the old discussion would be off-topic in
>the new group.

It's really not a nice thing to do. Starting new threads where they'd be
on topic is always the right choice.

Yads behaves like this. That's the very best argument that it's the wrong
thing to do.

[Sn!pe]

So, how about this: Omit the initial crosspost; post a followup into
the new group only; set Subject: as above -- New Subject (Was: Old
Subject). Note at the head of the new article (the aforementioned
followup) that the old thread is in the old group.

From the viewpoint of the old group, the branch does not exist but the
new (sub)thread with the new topic in the new group maintains continuity
with the original thread.

I note that this thread has somewhat drifted away from the original
topic... ;-)

[Anton Shepelev]

Sn!pe:

> From the viewpoint of the old group, the branch does not
> exist but the new (sub)thread with the new topic in the
> new group maintains continuity with the original thread.

Which means that readers in the old group will not be
notified that the conversation has moved. The entire point
of Followup-To: is to notify readers of the migration of the
sub-thread, that they may subscribe to the new group to
follow! Therefore, if Adam bans Followup-To:, the only
option is to make two posts:

a. In the old group, a notice that the subthread is
redirected to the new group.

b. In the new group, the continuation article replying
the original.

And believe (tell if I am wrong!) that FollowU-To was
provided exactly for such situations!

[Adam H. Kerman]

At least you're not posting off topic.

But topic cops -- I know I'm posting topic but I'm demanding that you
post your followup in a completely different newsgroup -- don't do that.

>From the viewpoint of the old group, the branch does not exist but the
>new (sub)thread with the new topic in the new group maintains continuity
>with the original thread.

>I note that this thread has somewhat drifted away from the original
>topic... ;-)

Topic drift is a feature of Usenet.

[Adam H. Kerman]

Anton Shepelev <anto...@gmail.moc> wrote:
>Sn!pe:

>>From the viewpoint of the old group, the branch does not
>>exist but the new (sub)thread with the new topic in the
>>new group maintains continuity with the original thread.

>Which means that readers in the old group will not be
>notified that the conversation has moved. The entire point
>of Followup-To: is to notify readers of the migration of the
>sub-thread, that they may subscribe to the new group to
>follow!

"Topic cop" isn't an actual Usenet position that can demand that
everybody posting in the thread post elsewhere.

>. . .

[Sn!pe]

Anton Shepelev <anto...@gmail.moc> wrote:

> Sn!pe:
>
> > From the viewpoint of the old group, the branch does not
> > exist but the new (sub)thread with the new topic in the
> > new group maintains continuity with the original thread.
>
> Which means that readers in the old group will not be
> notified that the conversation has moved. The entire point
> of Followup-To: is to notify readers of the migration of the
> sub-thread, that they may subscribe to the new group to
> follow! Therefore, if Adam bans Followup-To:, the only
> option is to make two posts:
>
> a. In the old group, a notice that the subthread is
> redirected to the new group.
>
> b. In the new group, the continuation article replying
> the original.
>
> And believe (tell if I am wrong!) that FollowU-To was
> provided exactly for such situations!
>

Alas, it appears that your position and Adam's are irreconcilable.
Whatever, this is a philosophical question of only passing interest
for me, so I'll leave you guys to your debate.

[Sn!pe]

Adam H. Kerman <a...@chinet.com> wrote:

[...]

> > I note that this thread has somewhat
> > drifted away from the original topic... ;-)
>
> Topic drift is a feature of Usenet.

Ain't that the truth! :-)

Whatever, I'm out of this now.

[yeti]

snip...@gmail.com (Sn!pe) writes:

> Adam H. Kerman <a...@chinet.com> wrote:
>
> [...]
>
>> > I note that this thread has somewhat
>> > drifted away from the original topic... ;-)
>>
>> Topic drift is a feature of Usenet.
>
> Ain't that the truth! :-)
>
> Whatever, I'm out of this now.

What a pity that this drifted subthread doesn't have an own subtopic so
that we could downscore only this part... :.(

<biiig> Sigh! </biiig>

--
|rom The Future. +++ Breaking News From The Future. +++ Breaking News F|
| The USoA are switching to the binary number system because |
| having more than 1+1 distinct digits is far too woke. |
|+ #MABA + #makeAmericaBinaryAgain + #USA + #USoA + #woke + #MABA + #ma|

[Adam H. Kerman]

yeti <ye...@tilde.institute> wrote:
>snip...@gmail.com (Sn!pe) writes:
>>Adam H. Kerman <a...@chinet.com> wrote:

>>[...]

>>>>I note that this thread has somewhat
>>>>drifted away from the original topic... ;-)

>>>Topic drift is a feature of Usenet.

>>Ain't that the truth! :-)

>>Whatever, I'm out of this now.

>What a pity that this drifted subthread doesn't have an own subtopic so
>that we could downscore only this part... :.(

><biiig> Sigh! </biiig>

You could, like, kill file me and kill all followups to me. Real
newsreaders are capable of this.

[Daniel65]

Adam H. Kerman wrote on 9/10/23 9:18 am:

<Snip>

> Yads behaves like this. That's the very best agument that it's the
> wrong thing to do.
>
Gee Whiz, Adam, you had me totally confused by your mention, out of the
blue, of "Yads" so I had to read back through the thread to see where
Yads, a.k.a. the asswipe, had been in this thread .... but then I
realised you were just using him as an example of what a (very poor) ISP
could do!! ;-)
--
Daniel

[Sn!pe]

Adam H. Kerman <a...@chinet.com> wrote:

> yeti <ye...@tilde.institute> wrote:
> >snip...@gmail.com (Sn!pe) writes:
> >>Adam H. Kerman <a...@chinet.com> wrote:
>
> >>[...]
>
> >>>>I note that this thread has somewhat
> >>>>drifted away from the original topic... ;-)
>
> >>>Topic drift is a feature of Usenet.
>
> >>Ain't that the truth! :-)
>
> >>Whatever, I'm out of this now.
>
> >What a pity that this drifted subthread doesn't have an own subtopic so
> >that we could downscore only this part... :.(
>
> ><biiig> Sigh! </biiig>
>
> You could, like, kill file me and kill all followups to me. Real
> newsreaders are capable of this.

MacSOUP can kill subthreads as well as whole threads;
also it can auto-kill (sub)threads on many criteria.

--
^Ď^. Sn!pe <https://youtu.be/_kqytf31a8E>

[Dan Purgert]

On 2023-10-08, Anton Shepelev wrote:
> Sn!pe:
>
>> From the viewpoint of the old group, the branch does not
>> exist but the new (sub)thread with the new topic in the
>> new group maintains continuity with the original thread.
>
> Which means that readers in the old group will not be
> notified that the conversation has moved. The entire point
> of Followup-To: is to notify readers of the migration of the
> sub-thread, that they may subscribe to the new group to
> follow! Therefore, if Adam bans Followup-To:, the only
> option is to make two posts:
>
> a. In the old group, a notice that the subthread is
> redirected to the new group.
>
> b. In the new group, the continuation article replying
> the original.
>
> And believe (tell if I am wrong!) that FollowU-To was
> provided exactly for such situations!

It is my understanding that setting a followup-to is somewhat of an
indicator of where one would like to see replies to their posts, in the
case of a cross posted article.

For example -> posting to misc.electronics.design and m.e.repair because
to repair "something", maybe I needed to build an equivalent circuit
because some part doesn't exist anymore. Set the followup-to as a hope
to minimize the number of off-topic posts in "design"...

--
|_|O|_|
|_|_|O| Github: https://github.com/dpurgert
|O|O|O| PGP: DDAB 23FB 19FA 7D85 1CC1 E067 6D65 70E5 4CE7 2860

[Adam H. Kerman]

No offense intended toward our other universal bad examples whose names
I didn't drop.

[Ray Banana]

* Sn!pe wrote:

> Retro Guy <retr...@i2pn2.org> wrote:
>
>> I've been doing the same for i2pn2.org for the past 6 weeks or so and I
>> agree, it's a very interesting challenge using spamassassin for Usenet.
>>

>> I've learned a lot more than I used to know about creative rules, and it's
>> a good learning experience. Enjoy!
>>
>> Also, I was happy to see e-s taking on the spam also. The more servers that
>> attack the problem, the more pleasant Usenet will be for the users. You're
>> doing a great job!
>
> Yes indeed. Your users thank you both,
> and the others fighting the good fight too.

And I may add, that E-S is now processing the NoCeMs issued by i2pn2
and usenet.ovh to bundle efforts to fight the spam flood.
Hopefully, Blueworldhosting will be joining soon, too.


--
Пу́тін — хуйло́
http://www.eternal-september.org

[Anton Shepelev]

Michael S:

> And I may add, that E-S is now processing the NoCeMs
> issued by i2pn2 and usenet.ovh to bundle efforts to fight
> the spam flood. Hopefully, Blueworldhosting will be
> joining soon, too.

This is good tidings, Ray!

[Anton Shepelev]

The oldest message in comp.lang.c that I now see is dated
2023.10.23 16:45. Is it because the SPAM has pushed most
normal messages off the retention queue?

[Ray Banana]

Thus spake Anton Shepelev <anton.txt@g{oogle}mail.com>

> The oldest message in comp.lang.c that I now see is dated
> 2023.10.23 16:45. Is it because the SPAM has pushed most
> normal messages off the retention queue?

The oldest article in c.l.c I see on the server is from
August 5, *2013*, so there is not much damage done by
The Thaidal Wave(TM).Your newsreader seems only load a limited
number of articles by default.

[Anton Shepelev]

Ray Banana to Anton Shepelev:

> > The oldest message in comp.lang.c that I now see is
> > dated 2023.10.23 16:45. Is it because the SPAM has
> > pushed most normal messages off the retention queue?
>
> The oldest article in c.l.c I see on the server is from
> August 5, *2013*, so there is not much damage done by The
> Thaidal Wave(TM).Your newsreader seems only load a limited
> number of articles by default.

Yes, but I set this limit to 8192. The deleted SPAM
articles seem to be counted, wherewhy the apparent
discrepancy.