Sn!pe <snipeco.1@gmail.com> wrote:
>Adam H. Kerman <ahk@chinet.com> wrote:
>>Sn!pe <snipeco.1@gmail.com> wrote:
>>>Marco Moock <mo01@posteo.de> wrote:
>>>>Am 05.09.2023 um 00:17:06 Uhr schrieb Leonardk:

>>>>>You helped put him in the spotlight by your efforts to restrict the
>>>>>free speech of others. Your personal likes or dislikes are not
>>>>>representative of the conversing community at large. You
>>>>>deliberately antagonized others who then went after the wrong people.

>>>>Is there anybody here who likes spam, address forgery and trollposts?
>>>>This is the content many people block and good NNTP server operators
>>>>ban such users.

>>>>I like freedom of speech, but spam and name forgery are not part of it.

>>>>I haven't seen many post originating from alphanet (only the operator
>>>>himself IIRC, but I don't read fr.*).
>>>>I don't see how he has been involved in having trolls on his server.

>>>>>You are attempting to ruin paganini. Why Ivo listens to you is
>>>>>mystifying as he seems to be a reasonable and bright fellow.

>>>>I simply asked questions - nothing more. Using a killfile to block
>>>>entire servers is the user's choice.

>>>>>It is people like you who kill Usenet, not those who use it.

>>>>No, nobody wants spammers, name forgers or trolls.
>>>>Server like Mixmin or aioe were on the killfile of many, many users in
>>>>de.*, because some people massively abused it for trollposts
>>>>crossposted to non-related groups.

>>>This conversation illustrates the difference between the authoritarian
>>>and libertarian viewpoints and risks generating more heat than light.

>>>IMO the choice of whether to read or to ignore another's posts should
>>>rest with the reader. It requires competence with a killfile but surely
>>>that is far better than censorship. For censored discussions, we might
>>>as well be in a web-forum.

>>I don't think so. It depends on the nature of the off-topic post.

>>1) Cancellable spam should be dealt with by spam countermeasures
>>implemented server-wide. This CANNOT be dealt with by kill file. If the
>>injecting News site takes no measures to prevent further cancellable
>>spam from being sent into Usenet, their peers need to seriously consider
>>whether de-peering is necessary.

>>2) Forgery must be dealt with at the injecting server. It's not
>>possible for the reader to deal with by kill file. They should be TOSsed
>>immediately and not allowed back.

>>3) Constantly-morphing trolls should be dealt with at the server level.
>>We get assholes that morph repeatedly in the same thread just to be
>>annoying. The News administrator should warn them. If they won't behave,
>>then TOS 'em.
>>I can be kill filed readily because I never morph. That's not true of
>>morphing trolls.

>>4) The constant posters of hate-filled articles might be TOSsed but for
>>other reasons. More typically, they themselves aren't writing the root
>>article but infringing upon copyright, reposting it from the Web. They
>>are crossposting or multiposting or both. That should probably be dealt
>>with at the injecting server, but that requires a lot of intervention.
>>If they morph or multi-post, again, that's really not possible to deal
>>with using a kill file. Cross posting can be dealt with using ordinary
>>kill file techniques if one has a good newsreader. If one doesn't want
>>to read political articles in one's non-political newsgroup, then kill
>>crossposts to *politics* and other known political newsgroups.

>>5) Trolling without crossposting, multi-posting, or copyright
>>infringement, without constant morphing, sure, leave that up to the user
>>to kill file.

>>There is a unique poster who forges others, re-injecting articles
>>written by others, AND commits abuse by pre-loading injection headers from
>>the original article. This blatantly violates RFCs but the injecting
>>server doesn't prevent it, and other servers are reluctant to junk these
>>articles in Cleanfeed or its equivalent. Despite not prohibiting this
>>abuse, no one wants to de-peer this News site. I know what to look for
>>but most Usenet readers wouldn't. There's nothing to kill file.

>>No, I've never agreed with Marco Moock, but I don't agree with you that
>>it's entirely hands off and that News administrators don't need to take
>>active measures to prevent large categories of abuse.

>All fair comment, I can find nothing in your post to disagree with. It
>has just occurred to me that as I use Eternal-September with its nicely
>sanitised feed, I'm probably not aware of the full extent of this
>problem. I'm certainly grateful for the efforts of News admins in
>cleaning the feed, I'm sure it isn't a trivial task.

You know, I looked through articles that I junked in one active
newsgroup I participate in. It gets lots of crossposting because the
political trolls know there is participation. Just this morning, my kill
file junk two dozen articles. The group would be unreadable without kill
filing. This is mainly junking unwanted crossposts. I'm not doing
anything about multi-posting, so I'm seeing quite a bit of drug spam,
which I junk manually. Much of that is through News sites that Google
Groups and nobody else peer with.

Adam H. Kerman <ahk@chinet.com> wrote:
[...]
> >>No, I've never agreed with Marco Moock, but I don't agree with you that
> >>it's entirely hands off and that News administrators don't need to take
> >>active measures to prevent large categories of abuse.
>
> >All fair comment, I can find nothing in your post to disagree with. It
> >has just occurred to me that as I use Eternal-September with its nicely
> >sanitised feed, I'm probably not aware of the full extent of this
> >problem. I'm certainly grateful for the efforts of News admins in
> >cleaning the feed, I'm sure it isn't a trivial task.
>
> You know, I looked through articles that I junked in one active
> newsgroup I participate in. It gets lots of crossposting because the
> political trolls know there is participation. Just this morning, my kill
> file junk two dozen articles. The group would be unreadable without kill
> filing. This is mainly junking unwanted crossposts. I'm not doing
> anything about multi-posting, so I'm seeing quite a bit of drug spam,
> which I junk manually. Much of that is through News sites that Google
> Groups and nobody else peer with.

It does seem that Google originates much of the spam that spoils Usenet.
My strategy is to to killfile all posts from:-

User-Agent: G2/1.0

- and to whitelist the few interesting posters who use Google. It may
be that I miss their initial posts so I rely on other readers finding
them interesting enough to respond to them, then I notice them.
--
^Ï^. Sn!pe <https://youtu.be/_kqytf31a8E>

My pet rock Gordon just is.

On Wed, 6 Sep 2023 00:20:52 +0100,
Sn!pe <snipeco.2@gmail.com> wrote:

> - and to whitelist the few interesting posters who use Google.
> It may be that I miss their initial posts so I rely on other
> readers finding them interesting enough to respond to them,
> then I notice them.

I like this strategy and use something roughly similar--but rather
than killfiling anyone, I use slrn's scoring function to highlight
the proven good posters (and interesting subjects). From there I
can identify additional posters worth reading. I know which
posters to skip, and read either all remaining posts, or just the
highlighted ones--depending on how much time I have.

--
Ted Heise <theise@panix.com> West Lafayette, IN, USA

Ted Heise wrote:
> Sn!pe <snipeco.2@gmail.com> wrote:
> >
> > - and to whitelist the few interesting posters who use Google.
> > It may be that I miss their initial posts so I rely on other
> > readers finding them interesting enough to respond to them,
> > then I notice them.
>
> I like this strategy and use something roughly similar--but rather
> than killfiling anyone, I use slrn's scoring function to highlight
> the proven good posters (and interesting subjects). From there I
> can identify additional posters worth reading. I know which
> posters to skip, and read either all remaining posts, or just the
> highlighted ones--depending on how much time I have.

I use the whitelist approach too, e.g. anything from Google gets
a score of -1 by default (and is therefore automatically marked as
read). Such articles are still accessible, if I explicitly want to
read them.

If I see some person who post something via Google, that I want to
read, this person gets a score of at least +1 and is no longer
automatically marked as read. If name morphing occurs, the negative
default score immediately will do its job again.

This scheme does no longer work for flooding with thousands of junk
articles. But my news provider cancels such spam.

On Tue, 5 Sep 2023 18:23:01 -0000 (UTC)
"Adam H. Kerman" <ahk@chinet.com> wrote:
> I'm not doing
> anything about multi-posting, so I'm seeing quite a bit of drug spam,
> which I junk manually. Much of that is through News sites that Google
> Groups and nobody else peer with.

I've never seen a googlegroups spam message which wasn't injected through
googlegroups. Do you happen to have a Message-ID of one ? In fact , I
wonder if googlegroups even forwards to other servers messages which did
not originate on googlegroups ; obviously it still makes available through
its web interface articles it received from other servers.

Spiros Bousbouras <spibou@gmail.com> wrote:
>On Tue, 5 Sep 2023 18:23:01 -0000 (UTC)
>"Adam H. Kerman" <ahk@chinet.com> wrote:

>>I'm not doing
>>anything about multi-posting, so I'm seeing quite a bit of drug spam,
>>which I junk manually. Much of that is through News sites that Google
>>Groups and nobody else peer with.

>I've never seen a googlegroups spam message which wasn't injected through
>googlegroups.

Sorry, I'm wrong. The drug spam isn't coming from peering News sites.
No other News site would allow them to be users.

>Do you happen to have a Message-ID of one ? In fact , I
>wonder if googlegroups even forwards to other servers messages which did
>not originate on googlegroups ; obviously it still makes available through
>its web interface articles it received from other servers.

Am 10.09.2023 um 20:33:31 Uhr schrieb Spiros Bousbouras:

> I've never seen a googlegroups spam message which wasn't injected
> through googlegroups. Do you happen to have a Message-ID of one ? In
> fact , I wonder if googlegroups even forwards to other servers
> messages which did not originate on googlegroups ; obviously it still
> makes available through its web interface articles it received from
> other servers.

IIRC Google Groups has only a small amount of peers and that makes it
very unlikely that post from other servers will be transferred over
Google and then to other servers because the probability that these
servers already have the messages is high.

On Sep 11, 2023 at 12:26:14 AM CDT, "Marco Moock" <mm+usenet-es@dorfdsl.de>
wrote:

> Am 10.09.2023 um 20:33:31 Uhr schrieb Spiros Bousbouras:
>
>> I've never seen a googlegroups spam message which wasn't injected
>> through googlegroups. Do you happen to have a Message-ID of one ? In
>> fact , I wonder if googlegroups even forwards to other servers
>> messages which did not originate on googlegroups ; obviously it still
>> makes available through its web interface articles it received from
>> other servers.
>
> IIRC Google Groups has only a small amount of peers and that makes it
> very unlikely that post from other servers will be transferred over
> Google and then to other servers because the probability that these
> servers already have the messages is high.

I can't say with 100% certainty how their NNTP infrastructure is setup, but I
have never seen an article pass through Google's infrastructure that did not
originate from Google Groups.

I know they peer with most of the commercial Usenet providers and a telecom or
two still running Usenet services in Europe, but it doesn't look like they
"exchange" articles from other sites with peers, they accept inbound but only
propagate articles that originate from their platform to peers.

Jesse Rehmer <jesse.rehmer@blueworldhosting.com> wrote:
>Sep 11, 2023 at 12:26:14 AM CDT, Marco Moock <mm+usenet-es@dorfdsl.de>:
>>Am 10.09.2023 um 20:33:31 Uhr schrieb Spiros Bousbouras:

>>>I've never seen a googlegroups spam message which wasn't injected
>>>through googlegroups. Do you happen to have a Message-ID of one ? In
>>>fact , I wonder if googlegroups even forwards to other servers
>>>messages which did not originate on googlegroups ; obviously it still
>>>makes available through its web interface articles it received from
>>>other servers.

>>IIRC Google Groups has only a small amount of peers and that makes it
>>very unlikely that post from other servers will be transferred over
>>Google and then to other servers because the probability that these
>>servers already have the messages is high.

>I can't say with 100% certainty how their NNTP infrastructure is setup, but I
>have never seen an article pass through Google's infrastructure that did not
>originate from Google Groups.

>I know they peer with most of the commercial Usenet providers and a telecom or
>two still running Usenet services in Europe, but it doesn't look like they
>"exchange" articles from other sites with peers, they accept inbound but only
>propagate articles that originate from their platform to peers.

I already withdrew my comment that the drug spam originated with sites
that Google Groups was peering with. I was wrong. Let's say instead that
Google Groups accepted drug spam from users on hosts that no responsible
News site would have tolerated for very long.

On Mon, 11 Sep 2023 14:10:19 -0000 (UTC)
"Adam H. Kerman" <ahk@chinet.com> wrote:
> Jesse Rehmer <jesse.rehmer@blueworldhosting.com> wrote:
> >I can't say with 100% certainty how their NNTP infrastructure is setup, but I
> >have never seen an article pass through Google's infrastructure that did not
> >originate from Google Groups.
>
> >I know they peer with most of the commercial Usenet providers and a telecom or
> >two still running Usenet services in Europe, but it doesn't look like they
> >"exchange" articles from other sites with peers, they accept inbound but only
> >propagate articles that originate from their platform to peers.
>
> I already withdrew my comment that the drug spam originated with sites
> that Google Groups was peering with. I was wrong. Let's say instead that
> Google Groups accepted drug spam from users on hosts that no responsible
> News site would have tolerated for very long.

It isn't about your original comment anymore , I asked because I'm curious in
general and not just for spam posts but all of them. Like Jesse , I've never
seen an article which passed through googlegroups and did not originate from
googlegroups.

Is the Path: header field a reliable way to tell ? That's where I tend to
look and if *googlegroups.com appears at all then it's the penultimate
component , right before the !not-for-mail part. Perhaps some server
administrator who reads this group can run a script and see if there are
messages arriving at their server which don't meet this description.

Spiros Bousbouras <spibou@gmail.com> wrote:
>Mon, 11 Sep 2023 14:10:19 -0000 (UTC) Adam H. Kerman <ahk@chinet.com>:
>>Jesse Rehmer <jesse.rehmer@blueworldhosting.com> wrote:

>>>I can't say with 100% certainty how their NNTP infrastructure is setup,
>>>but I have never seen an article pass through Google's infrastructure
>>>that did not originate from Google Groups.

>>>I know they peer with most of the commercial Usenet providers and a
>>>telecom or two still running Usenet services in Europe, but it doesn't
>>>look like they "exchange" articles from other sites with peers, they
>>>accept inbound but only propagate articles that originate from their
>>>platform to peers.

>>I already withdrew my comment that the drug spam originated with sites
>>that Google Groups was peering with. I was wrong. Let's say instead that
>>Google Groups accepted drug spam from users on hosts that no responsible
>>News site would have tolerated for very long.

>It isn't about your original comment anymore , I asked because I'm curious in
>general and not just for spam posts but all of them. Like Jesse , I've never
>seen an article which passed through googlegroups and did not originate from
>googlegroups.

>Is the Path: header field a reliable way to tell ? That's where I tend to
>look and if *googlegroups.com appears at all then it's the penultimate
>component , right before the !not-for-mail part. Perhaps some server
>administrator who reads this group can run a script and see if there are
>messages arriving at their server which don't meet this description.

Are you suggesting that Google Groups truncates Path? Anything is
possible.

On Mon, 11 Sep 2023 17:15:20 -0000 (UTC)
"Adam H. Kerman" <ahk@chinet.com> wrote:
> Spiros Bousbouras <spibou@gmail.com> wrote:
> >Mon, 11 Sep 2023 14:10:19 -0000 (UTC) Adam H. Kerman <ahk@chinet.com>:
> >>Jesse Rehmer <jesse.rehmer@blueworldhosting.com> wrote:
>
> >>>I can't say with 100% certainty how their NNTP infrastructure is setup,
> >>>but I have never seen an article pass through Google's infrastructure
> >>>that did not originate from Google Groups.
>
> >>>I know they peer with most of the commercial Usenet providers and a
> >>>telecom or two still running Usenet services in Europe, but it doesn't
> >>>look like they "exchange" articles from other sites with peers, they
> >>>accept inbound but only propagate articles that originate from their
> >>>platform to peers.
>
> >>I already withdrew my comment that the drug spam originated with sites
> >>that Google Groups was peering with. I was wrong. Let's say instead that
> >>Google Groups accepted drug spam from users on hosts that no responsible
> >>News site would have tolerated for very long.
>
> >It isn't about your original comment anymore , I asked because I'm curious in
> >general and not just for spam posts but all of them. Like Jesse , I've never
> >seen an article which passed through googlegroups and did not originate from
> >googlegroups.
>
> >Is the Path: header field a reliable way to tell ? That's where I tend to
> >look and if *googlegroups.com appears at all then it's the penultimate
> >component , right before the !not-for-mail part. Perhaps some server
> >administrator who reads this group can run a script and see if there are
> >messages arriving at their server which don't meet this description.
>
> Are you suggesting that Google Groups truncates Path? Anything is
> possible.

No , I'm not suggesting anything , I simply asked a question. To rephrase it ,
if in Path: a *googlegroups.com pattern appears not as the final or
penultimate component , does it mean that the message passed through
googlegroups but was not posted through googlegroups ?

Le 12/09/2023 à 22:52, Spiros Bousbouras a écrit :

> No , I'm not suggesting anything , I simply asked a question. To rephrase it ,
> if in Path: a *googlegroups.com pattern appears not as the final or
> penultimate component , does it mean that the message passed through
> googlegroups but was not posted through googlegroups ?

Has anybody got a direct outwards feed with Googlegroups ? I think this is
the question. And I don't know what they're doing because they don't even
show headers anymore on their crappy interface.