Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]

Groups > sci.physics.relativity > #633520 > unrolled thread

How I deal with the enormous amount of spam

Back to article view | Back to sci.physics.relativity

Contents

  How I deal with the enormous amount of spam Tom Roberts <tjoberts137@sbcglobal.net> - 2024-01-29 18:03 -0600
    Re: How I deal with the enormous amount of spam Dlzc <turkeyheadedmutha@gmail.com> - 2024-01-29 16:52 -0800
    Re: How I deal with the enormous amount of spam pnalsing@gmail.com (palsing) - 2024-01-30 00:56 +0000
      Re: How I deal with the enormous amount of spam Dlzc <turkeyheadedmutha@gmail.com> - 2024-01-29 17:09 -0800
      Re: How I deal with the enormous amount of spam Ross Finlayson <ross.a.finlayson@gmail.com> - 2024-01-30 12:54 -0800
        Re: How I deal with the enormous amount of spam Ross Finlayson <ross.a.finlayson@gmail.com> - 2024-02-03 09:37 -0800
          Re: How I deal with the enormous amount of spam Dlzc <turkeyheadedmutha@gmail.com> - 2024-02-03 09:47 -0800
            Re: How I deal with the enormous amount of spam Ross Finlayson <ross.a.finlayson@gmail.com> - 2024-02-03 09:57 -0800
          Re: How I deal with the enormous amount of spam The Starmaker <starmaker@ix.netcom.com> - 2024-02-03 13:43 -0800
            Re: How I deal with the enormous amount of spam The Starmaker <starmaker@ix.netcom.com> - 2024-02-03 14:46 -0800
              Re: How I deal with the enormous amount of spam Ross Finlayson <ross.a.finlayson@gmail.com> - 2024-02-04 09:55 -0800
                Meta: Re: How I deal with the enormous amount of spam Ross Finlayson <ross.a.finlayson@gmail.com> - 2024-02-04 10:17 -0800
                  Re: Meta: Re: How I deal with the enormous amount of spam The Starmaker <starmaker@ix.netcom.com> - 2024-02-04 12:35 -0800
                    Re: Meta: Re: How I deal with the enormous amount of spam The Starmaker <starmaker@ix.netcom.com> - 2024-02-04 12:53 -0800
                      Re: Meta: Re: How I deal with the enormous amount of spam Ross Finlayson <ross.a.finlayson@gmail.com> - 2024-02-04 18:28 -0800
                        Re: Meta: Re: How I deal with the enormous amount of spam The Starmaker <starmaker@ix.netcom.com> - 2024-02-04 22:02 -0800
                          Re: Meta: Re: How I deal with the enormous amount of spam The Starmaker <starmaker@ix.netcom.com> - 2024-02-06 10:30 -0800
                            Re: Meta: Re: How I deal with the enormous amount of spam The Starmaker <starmaker@ix.netcom.com> - 2024-02-06 11:02 -0800
                              Re: Meta: Re: How I deal with the enormous amount of spam The Starmaker <starmaker@ix.netcom.com> - 2024-02-06 11:23 -0800
                        Re: Meta: Re: How I deal with the enormous amount of spam Ross Finlayson <ross.a.finlayson@gmail.com> - 2024-02-09 11:38 -0800
    Re: How I deal with the enormous amount of spam Nico Kozák Pásztori <oiokx@kkpk.hu> - 2024-01-30 01:48 +0000
      Re: How I deal with the enormous amount of spam Physfitfreak <Physfitfreak@gmail.com> - 2024-01-29 22:59 -0600
        Re: How I deal with the enormous amount of spam Timmie Császár Barabás <aaeei@imb.hu> - 2024-01-30 15:58 +0000
    Re: How I deal with the enormous amount of spam Richard Hertz <hertz778@gmail.com> - 2024-01-29 20:11 -0800
      Re: How I deal with the enormous amount of spam tomyee3@gmail.com (ProkaryoticCaspaseHomolog) - 2024-02-03 10:38 +0000
        Re: How I deal with the enormous amount of spam nospam@de-ster.demon.nl (J. J. Lodder) - 2024-02-03 13:54 +0100
          Re: How I deal with the enormous amount of spam tomyee3@gmail.com (ProkaryoticCaspaseHomolog) - 2024-02-06 20:03 +0000
    Re: How I deal with the enormous amount of spam nospam@de-ster.demon.nl (J. J. Lodder) - 2024-01-30 11:17 +0100
    Re: How I deal with the enormous amount of spam tomyee3@gmail.com (ProkaryoticCaspaseHomolog) - 2024-02-03 08:38 +0000
      Re: How I deal with the enormous amount of spam nospam@de-ster.demon.nl (J. J. Lodder) - 2024-02-03 13:54 +0100
        Re: How I deal with the enormous amount of spam Volney <volney@invalid.invalid> - 2024-02-03 12:48 -0500
          Re: How I deal with the enormous amount of spam Roscoe Hatukaev <te@hvhre.ru> - 2024-02-03 18:21 +0000
            Re: How I deal with the enormous amount of spam Physfitfreak <Physfitfreak@gmail.com> - 2024-02-03 14:46 -0600
          Re: How I deal with the enormous amount of spam Ramses Fenstermacher <cehfem@sfse.de> - 2024-02-03 18:51 +0000
          Re: How I deal with the enormous amount of spam nospam@de-ster.demon.nl (J. J. Lodder) - 2024-02-03 22:04 +0100
          Re: How I deal with the enormous amount of spam Dlzc <turkeyheadedmutha@gmail.com> - 2024-02-03 14:59 -0800
            Re: How I deal with the enormous amount of spam RichD <r_delaney2001@yahoo.com> - 2024-02-03 17:24 -0800
            Re: How I deal with the enormous amount of spam RichD <r_delaney2001@yahoo.com> - 2024-02-03 17:40 -0800
              Re: How I deal with the enormous amount of spam Tom Roberts <tjoberts137@sbcglobal.net> - 2024-02-04 10:25 -0600
                Re: How I deal with the enormous amount of spam Dlzc <turkeyheadedmutha@gmail.com> - 2024-02-04 08:40 -0800
            Re: How I deal with the enormous amount of spam tomyee3@gmail.com (ProkaryoticCaspaseHomolog) - 2024-02-04 14:06 +0000
              Re: How I deal with the enormous amount of spam Dlzc <turkeyheadedmutha@gmail.com> - 2024-02-04 07:54 -0800
    Re: How I deal with the enormous amount of spam Athel Cornish-Bowden <me@yahoo.com> - 2024-02-03 10:48 +0100
    Re: How I deal with the enormous amount of spam wugi <wugi@brol.invalid> - 2024-02-05 12:35 +0100
    Re: How I deal with the enormous amount of spam xip14 <xip1415926@gmail.com> - 2024-02-05 06:16 -0800
    Re: How I deal with the enormous amount of spam xip14 <xip1415926@gmail.com> - 2024-02-05 10:27 -0800

Page 1 of 3  [1] 2 3  Next page →

#633520 — How I deal with the enormous amount of spam

I use Thunderbird to read Usenet. Recently sci.physics.relativity has
been getting hundreds of spam posts each day, completely overwhelming
legitimate content. These spam posts share the property that they are
written in a non-latin script.

Thunderbird implements message filters that can mark a message Read. So
I created a filter to run on sci.physics.relativity that marks messages
Read. Then when reading the newsgroups, I simply display only unread
messages. The key to making this work is to craft the filter so it marks
messages in which the Subject matches any of a dozen characters picked
from some spam messages.

This doesn't completely eliminate the spam, but it is now only a few
messages per day.

Tom Roberts

[toc] | [next] | [standalone]

#633554

On Monday, January 29, 2024 at 6:03:55 PM UTC-6, Tom Roberts wrote:
> I use Thunderbird to read Usenet. Recently sci.physics.relativity has 
> been getting hundreds of spam posts each day, completely overwhelming 
> legitimate content. These spam posts share the property that they are 
> written in a non-latin script. 

And I use Google Groups, and wade through all the spam, and report each and every message as spam to Google.  Theoretically, this means they no longer get any search engine hits via Google, unless Google lies (and is also as incompetent as their programmers).

Only have to put up with it for four more weeks...

David A. Smith

[toc] | [prev] | [next] | [standalone]

#633562

Tom Roberts wrote:

> I use Thunderbird to read Usenet. Recently sci.physics.relativity has
> been getting hundreds of spam posts each day, completely overwhelming
> legitimate content. These spam posts share the property that they are
> written in a non-latin script.

> Thunderbird implements message filters that can mark a message Read. So
> I created a filter to run on sci.physics.relativity that marks messages
> Read. Then when reading the newsgroups, I simply display only unread
> messages. The key to making this work is to craft the filter so it marks
> messages in which the Subject matches any of a dozen characters picked
> from some spam messages.

> This doesn't completely eliminate the spam, but it is now only a few
> messages per day.

> Tom Roberts

I would like to do the same thing, so I installed Thunderbird... but setting it up to read newsgroups is beyond my paltry computer skills and is not at all intuitive. If anyone can point to an idiot-proof tutorial for doing this It would be much appreciated.

\Paul Alsing

[toc] | [prev] | [next] | [standalone]

#633568

On Monday, January 29, 2024 at 7:02:05 PM UTC-6, palsing wrote:
> I would like to do the same thing, so I installed Thunderbird... but setting it up to read newsgroups
> is beyond my paltry computer skills and is not at all intuitive. If anyone can point to an idiot-proof
> tutorial for doing this It would be much appreciated. 

Google search result:
https://kb.iu.edu/d/arrn

David A. Smith

[toc] | [prev] | [next] | [standalone]

#634101

On Monday, January 29, 2024 at 5:02:05 PM UTC-8, palsing wrote:
> Tom Roberts wrote: 
> 
> > I use Thunderbird to read Usenet. Recently sci.physics.relativity has 
> > been getting hundreds of spam posts each day, completely overwhelming 
> > legitimate content. These spam posts share the property that they are 
> > written in a non-latin script. 
> 
> > Thunderbird implements message filters that can mark a message Read. So 
> > I created a filter to run on sci.physics.relativity that marks messages 
> > Read. Then when reading the newsgroups, I simply display only unread 
> > messages. The key to making this work is to craft the filter so it marks 
> > messages in which the Subject matches any of a dozen characters picked 
> > from some spam messages. 
> 
> > This doesn't completely eliminate the spam, but it is now only a few 
> > messages per day. 
> 
> > Tom Roberts
> I would like to do the same thing, so I installed Thunderbird... but setting it up to read newsgroups is beyond my paltry computer skills and is not at all intuitive. If anyone can point to an idiot-proof tutorial for doing this It would be much appreciated. 
> 
> \Paul Alsing

Yeah, it's pretty bad, or, worse anybody's ever seen it.

I as well sort of mow the lawn a bit or mark the spam.

It seems alright if it'll be a sort of clean break:  on Feb 22 according to Google, 
Google will break its compeerage to Usenet, and furthermore make read-only 
the archives, what it has, what until then, will be as it was.

Over on sci.math I've had the idea for a while of making some brief and 
special purpose Usenet compeers, for only some few groups, or, you 
know, the _belles lettres_ of the text hierarchy.

"Meta:  a usenet server just for sci.math"
  -- https://groups.google.com/g/sci.math/c/zggff_pVEks

So, there you can read the outlook of this kind of thing, then while sort 
of simple as the protocol is simple and its implementations widespread, 
how to deal with the "signal and noise" of "exposed messaging destinations 
on the Internet", well on that thread I'm theorizing a sort of, "NOOBNB protocol", 
figuring to make an otherwise just standard Usenet compeer, and also for 
email or messaging destinations, sort of designed with the expectation that 
there will be spam, and spam and ham are hand in hand, to exclude it in simple terms.

NOOBNB:  New Old Off Bot Non Bad, Curated/Purgatory/Raw triple-feed

(That and a firmer sort of "Load Shed" or "Load Hold" at the transport layer.)

Also it would be real great if at least there was surfaced to the Internet a 
read-only view of any message by its message ID, a "URL", or as for a "URI", 
a "URN", a reliable perma-link in the IETF "news" protocol, namespace.

https://groups.google.com/g/sci.math/c/zggff_pVEks

I wonder that there's a reliable sort of long-term project that surfaces 
"news" protocol message-IDs, ....  It's a stable, standards-based protocol.


Thunderbird, "SLRN", ....  Thanks for caring.  We care.


https://groups.google.com/g/sci.physics.relativity/c/ToBo6XOymUw

[toc] | [prev] | [next] | [standalone]

#636611

On 01/30/2024 12:54 PM, Ross Finlayson wrote:
> On Monday, January 29, 2024 at 5:02:05 PM UTC-8, palsing wrote:
>> Tom Roberts wrote:
>>
>>> I use Thunderbird to read Usenet. Recently sci.physics.relativity has
>>> been getting hundreds of spam posts each day, completely overwhelming
>>> legitimate content. These spam posts share the property that they are
>>> written in a non-latin script.
>>
>>> Thunderbird implements message filters that can mark a message Read. So
>>> I created a filter to run on sci.physics.relativity that marks messages
>>> Read. Then when reading the newsgroups, I simply display only unread
>>> messages. The key to making this work is to craft the filter so it marks
>>> messages in which the Subject matches any of a dozen characters picked
>>> from some spam messages.
>>
>>> This doesn't completely eliminate the spam, but it is now only a few
>>> messages per day.
>>
>>> Tom Roberts
>> I would like to do the same thing, so I installed Thunderbird... but setting it up to read newsgroups is beyond my paltry computer skills and is not at all intuitive. If anyone can point to an idiot-proof tutorial for doing this It would be much appreciated.
>>
>> \Paul Alsing
>
> Yeah, it's pretty bad, or, worse anybody's ever seen it.
>
> I as well sort of mow the lawn a bit or mark the spam.
>
> It seems alright if it'll be a sort of clean break:  on Feb 22 according to Google,
> Google will break its compeerage to Usenet, and furthermore make read-only
> the archives, what it has, what until then, will be as it was.
>
> Over on sci.math I've had the idea for a while of making some brief and
> special purpose Usenet compeers, for only some few groups, or, you
> know, the _belles lettres_ of the text hierarchy.
>
> "Meta:  a usenet server just for sci.math"
>    -- https://groups.google.com/g/sci.math/c/zggff_pVEks
>
> So, there you can read the outlook of this kind of thing, then while sort
> of simple as the protocol is simple and its implementations widespread,
> how to deal with the "signal and noise" of "exposed messaging destinations
> on the Internet", well on that thread I'm theorizing a sort of, "NOOBNB protocol",
> figuring to make an otherwise just standard Usenet compeer, and also for
> email or messaging destinations, sort of designed with the expectation that
> there will be spam, and spam and ham are hand in hand, to exclude it in simple terms.
>
> NOOBNB:  New Old Off Bot Non Bad, Curated/Purgatory/Raw triple-feed
>
> (That and a firmer sort of "Load Shed" or "Load Hold" at the transport layer.)
>
> Also it would be real great if at least there was surfaced to the Internet a
> read-only view of any message by its message ID, a "URL", or as for a "URI",
> a "URN", a reliable perma-link in the IETF "news" protocol, namespace.
>
> https://groups.google.com/g/sci.math/c/zggff_pVEks
>
> I wonder that there's a reliable sort of long-term project that surfaces
> "news" protocol message-IDs, ....  It's a stable, standards-based protocol.
>
>
> Thunderbird, "SLRN", ....  Thanks for caring.  We care.
>
>
> https://groups.google.com/g/sci.physics.relativity/c/ToBo6XOymUw
>

One fellow reached me via e-mail and he said, hey, the Googler spam is 
outrageous, can we do anything about it?  Would you write a script to
funnel all their message-ID's into the abuse reporting?  And I was like, 
you know, about 2008 I did just that, there was a big spam flood,
and I wrote a little script to find them and extract their posting-account,
and the message-ID, and a little script to post to the posting-host,
each one of the wicked spams.

At the time that seemed to help, they sort of dried up, here there's
that basically they're not following the charter, but, it's the 
posting-account
in the message headers that indicate the origin of the post, not the
email address.  So, I wonder, given that I can extract the posting-accounts
of all the spams, how to match the posting-account to then determine
whether it's a sockpuppet-farm or what, and basically about sending them up.

[toc] | [prev] | [next] | [standalone]

#636618

On Saturday, February 3, 2024 at 11:37:28 AM UTC-6, Ross Finlayson wrote:
> On 01/30/2024 12:54 PM, Ross Finlayson wrote: 
> ... how to match the posting-account to then determine 
> whether it's a sockpuppet-farm or what, and basically
> about sending them up.

Its too late to save Google Groups front end.

They post and another account posts a follow-up on the same message within less than a minute.  It is human-based based on the pace of posting.  It IS a sock-puppet farm.

David A. Smith

[toc] | [prev] | [next] | [standalone]

#636623

On 02/03/2024 09:47 AM, Dlzc wrote:
> On Saturday, February 3, 2024 at 11:37:28 AM UTC-6, Ross Finlayson wrote:
>> On 01/30/2024 12:54 PM, Ross Finlayson wrote:
>> ... how to match the posting-account to then determine
>> whether it's a sockpuppet-farm or what, and basically
>> about sending them up.
>
> Its too late to save Google Groups front end.
>
> They post and another account posts a follow-up on the same message within less than a minute.  It is human-based based on the pace of posting.  It IS a sock-puppet farm.
>
> David A. Smith
>

Ah, the sockpuppet.

"Hi sockpuppet how are you today?"
"The guy hiding under here has his arm entirely up me, I'm not doing this."

Clearly the posts themselves are generated by a mail-merge, which
you're probably familiar with or I am as "here's a list of addresses,
and here's a little document with placeholders, the templates
get merged the substitution placeholders interpolated their
expression to the matching parameters, then print that out
the printer but be careful printing the labels in the special
stick-on label paper in the printer and you wouldn't want that
poor quality adhesive would result jamming your printer".

Anyways though it's clear that every now and then one of the
parameters goes missing and it shows "Template" or "Missing"
or whatever, it's just a usual sort of blast-fax mail-merge setup.

Yet, here it's as what appears to be a ton of different compromised
Google accounts, as to whether they're associated with the names
and emails of the posts at all, is unclear.  So, there's an idea to
make a little database matching up posting-account according to
the Google G2 injector that appears to be injecting them to Usenet,
helping illustrate that such-and-such accounts are at best compromised
and probably collusive, where of course we wouldn't want the innocents
they may be to suffer just because their account is compromised.  (Or, ....)


Or, "gee, thanks OAuth".

[toc] | [prev] | [next] | [standalone]

#636766

Ross Finlayson wrote:
> 
> On 01/30/2024 12:54 PM, Ross Finlayson wrote:
> > On Monday, January 29, 2024 at 5:02:05 PM UTC-8, palsing wrote:
> >> Tom Roberts wrote:
> >>
> >>> I use Thunderbird to read Usenet. Recently sci.physics.relativity has
> >>> been getting hundreds of spam posts each day, completely overwhelming
> >>> legitimate content. These spam posts share the property that they are
> >>> written in a non-latin script.
> >>
> >>> Thunderbird implements message filters that can mark a message Read. So
> >>> I created a filter to run on sci.physics.relativity that marks messages
> >>> Read. Then when reading the newsgroups, I simply display only unread
> >>> messages. The key to making this work is to craft the filter so it marks
> >>> messages in which the Subject matches any of a dozen characters picked
> >>> from some spam messages.
> >>
> >>> This doesn't completely eliminate the spam, but it is now only a few
> >>> messages per day.
> >>
> >>> Tom Roberts
> >> I would like to do the same thing, so I installed Thunderbird... but setting it up to read newsgroups is beyond my paltry computer skills and is not at all intuitive. If anyone can point to an idiot-proof tutorial for doing this It would be much appreciated.
> >>
> >> \Paul Alsing
> >
> > Yeah, it's pretty bad, or, worse anybody's ever seen it.
> >
> > I as well sort of mow the lawn a bit or mark the spam.
> >
> > It seems alright if it'll be a sort of clean break:  on Feb 22 according to Google,
> > Google will break its compeerage to Usenet, and furthermore make read-only
> > the archives, what it has, what until then, will be as it was.
> >
> > Over on sci.math I've had the idea for a while of making some brief and
> > special purpose Usenet compeers, for only some few groups, or, you
> > know, the _belles lettres_ of the text hierarchy.
> >
> > "Meta:  a usenet server just for sci.math"
> >    -- https://groups.google.com/g/sci.math/c/zggff_pVEks
> >
> > So, there you can read the outlook of this kind of thing, then while sort
> > of simple as the protocol is simple and its implementations widespread,
> > how to deal with the "signal and noise" of "exposed messaging destinations
> > on the Internet", well on that thread I'm theorizing a sort of, "NOOBNB protocol",
> > figuring to make an otherwise just standard Usenet compeer, and also for
> > email or messaging destinations, sort of designed with the expectation that
> > there will be spam, and spam and ham are hand in hand, to exclude it in simple terms.
> >
> > NOOBNB:  New Old Off Bot Non Bad, Curated/Purgatory/Raw triple-feed
> >
> > (That and a firmer sort of "Load Shed" or "Load Hold" at the transport layer.)
> >
> > Also it would be real great if at least there was surfaced to the Internet a
> > read-only view of any message by its message ID, a "URL", or as for a "URI",
> > a "URN", a reliable perma-link in the IETF "news" protocol, namespace.
> >
> > https://groups.google.com/g/sci.math/c/zggff_pVEks
> >
> > I wonder that there's a reliable sort of long-term project that surfaces
> > "news" protocol message-IDs, ....  It's a stable, standards-based protocol.
> >
> >
> > Thunderbird, "SLRN", ....  Thanks for caring.  We care.
> >
> >
> > https://groups.google.com/g/sci.physics.relativity/c/ToBo6XOymUw
> >
> 
> One fellow reached me via e-mail and he said, hey, the Googler spam is
> outrageous, can we do anything about it?  Would you write a script to
> funnel all their message-ID's into the abuse reporting?  And I was like,
> you know, about 2008 I did just that, there was a big spam flood,
> and I wrote a little script to find them and extract their posting-account,
> and the message-ID, and a little script to post to the posting-host,
> each one of the wicked spams.
> 
> At the time that seemed to help, they sort of dried up, here there's
> that basically they're not following the charter, but, it's the
> posting-account
> in the message headers that indicate the origin of the post, not the
> email address.  So, I wonder, given that I can extract the posting-accounts
> of all the spams, how to match the posting-account to then determine
> whether it's a sockpuppet-farm or what, and basically about sending them up.

Let me see your little script. Post it here.



-- 
The Starmaker -- To question the unquestionable, ask the unaskable,
to think the unthinkable, mention the unmentionable, say the unsayable, 
and challenge the unchallengeable.

[toc] | [prev] | [next] | [standalone]

#636813

The Starmaker wrote:
> 
> Ross Finlayson wrote:
> >
> > On 01/30/2024 12:54 PM, Ross Finlayson wrote:
> > > On Monday, January 29, 2024 at 5:02:05 PM UTC-8, palsing wrote:
> > >> Tom Roberts wrote:
> > >>
> > >>> I use Thunderbird to read Usenet. Recently sci.physics.relativity has
> > >>> been getting hundreds of spam posts each day, completely overwhelming
> > >>> legitimate content. These spam posts share the property that they are
> > >>> written in a non-latin script.
> > >>
> > >>> Thunderbird implements message filters that can mark a message Read. So
> > >>> I created a filter to run on sci.physics.relativity that marks messages
> > >>> Read. Then when reading the newsgroups, I simply display only unread
> > >>> messages. The key to making this work is to craft the filter so it marks
> > >>> messages in which the Subject matches any of a dozen characters picked
> > >>> from some spam messages.
> > >>
> > >>> This doesn't completely eliminate the spam, but it is now only a few
> > >>> messages per day.
> > >>
> > >>> Tom Roberts
> > >> I would like to do the same thing, so I installed Thunderbird... but setting it up to read newsgroups is beyond my paltry computer skills and is not at all intuitive. If anyone can point to an idiot-proof tutorial for doing this It would be much appreciated.
> > >>
> > >> \Paul Alsing
> > >
> > > Yeah, it's pretty bad, or, worse anybody's ever seen it.
> > >
> > > I as well sort of mow the lawn a bit or mark the spam.
> > >
> > > It seems alright if it'll be a sort of clean break:  on Feb 22 according to Google,
> > > Google will break its compeerage to Usenet, and furthermore make read-only
> > > the archives, what it has, what until then, will be as it was.
> > >
> > > Over on sci.math I've had the idea for a while of making some brief and
> > > special purpose Usenet compeers, for only some few groups, or, you
> > > know, the _belles lettres_ of the text hierarchy.
> > >
> > > "Meta:  a usenet server just for sci.math"
> > >    -- https://groups.google.com/g/sci.math/c/zggff_pVEks
> > >
> > > So, there you can read the outlook of this kind of thing, then while sort
> > > of simple as the protocol is simple and its implementations widespread,
> > > how to deal with the "signal and noise" of "exposed messaging destinations
> > > on the Internet", well on that thread I'm theorizing a sort of, "NOOBNB protocol",
> > > figuring to make an otherwise just standard Usenet compeer, and also for
> > > email or messaging destinations, sort of designed with the expectation that
> > > there will be spam, and spam and ham are hand in hand, to exclude it in simple terms.
> > >
> > > NOOBNB:  New Old Off Bot Non Bad, Curated/Purgatory/Raw triple-feed
> > >
> > > (That and a firmer sort of "Load Shed" or "Load Hold" at the transport layer.)
> > >
> > > Also it would be real great if at least there was surfaced to the Internet a
> > > read-only view of any message by its message ID, a "URL", or as for a "URI",
> > > a "URN", a reliable perma-link in the IETF "news" protocol, namespace.
> > >
> > > https://groups.google.com/g/sci.math/c/zggff_pVEks
> > >
> > > I wonder that there's a reliable sort of long-term project that surfaces
> > > "news" protocol message-IDs, ....  It's a stable, standards-based protocol.
> > >
> > >
> > > Thunderbird, "SLRN", ....  Thanks for caring.  We care.
> > >
> > >
> > > https://groups.google.com/g/sci.physics.relativity/c/ToBo6XOymUw
> > >
> >
> > One fellow reached me via e-mail and he said, hey, the Googler spam is
> > outrageous, can we do anything about it?  Would you write a script to
> > funnel all their message-ID's into the abuse reporting?  And I was like,
> > you know, about 2008 I did just that, there was a big spam flood,
> > and I wrote a little script to find them and extract their posting-account,
> > and the message-ID, and a little script to post to the posting-host,
> > each one of the wicked spams.
> >
> > At the time that seemed to help, they sort of dried up, here there's
> > that basically they're not following the charter, but, it's the
> > posting-account
> > in the message headers that indicate the origin of the post, not the
> > email address.  So, I wonder, given that I can extract the posting-accounts
> > of all the spams, how to match the posting-account to then determine
> > whether it's a sockpuppet-farm or what, and basically about sending them up.
> 
> Let me see your little script. Post it here.

Here is a list I currently have:

salz.txt
usenet.death.penalty.gz
purify.txt
NewsAgent110-MS.exe
HipCrime's NewsAgent (v1_11).htm
NewsAgent111-BE.zip
SuperCede.exe
NewsAgent023.exe
NewsAgent025.exe
ActiveAgent.java
HipCrime's NewsAgent (v1_02)_files
NewsCancel.java (source code)

(plus updated python versions)



(Maybe your script is inthere somewhere?)



Show me what you got. walk the walk.

-- 
The Starmaker -- To question the unquestionable, ask the unaskable,
to think the unthinkable, mention the unmentionable, say the unsayable, 
and challenge the unchallengeable.

[toc] | [prev] | [next] | [standalone]

#637427

On 02/03/2024 02:46 PM, The Starmaker wrote:
> The Starmaker wrote:
>>
>> Ross Finlayson wrote:
>>>
>>> On 01/30/2024 12:54 PM, Ross Finlayson wrote:
>>>> On Monday, January 29, 2024 at 5:02:05 PM UTC-8, palsing wrote:
>>>>> Tom Roberts wrote:
>>>>>
>>>>>> I use Thunderbird to read Usenet. Recently sci.physics.relativity has
>>>>>> been getting hundreds of spam posts each day, completely overwhelming
>>>>>> legitimate content. These spam posts share the property that they are
>>>>>> written in a non-latin script.
>>>>>
>>>>>> Thunderbird implements message filters that can mark a message Read. So
>>>>>> I created a filter to run on sci.physics.relativity that marks messages
>>>>>> Read. Then when reading the newsgroups, I simply display only unread
>>>>>> messages. The key to making this work is to craft the filter so it marks
>>>>>> messages in which the Subject matches any of a dozen characters picked
>>>>>> from some spam messages.
>>>>>
>>>>>> This doesn't completely eliminate the spam, but it is now only a few
>>>>>> messages per day.
>>>>>
>>>>>> Tom Roberts
>>>>> I would like to do the same thing, so I installed Thunderbird... but setting it up to read newsgroups is beyond my paltry computer skills and is not at all intuitive. If anyone can point to an idiot-proof tutorial for doing this It would be much appreciated.
>>>>>
>>>>> \Paul Alsing
>>>>
>>>> Yeah, it's pretty bad, or, worse anybody's ever seen it.
>>>>
>>>> I as well sort of mow the lawn a bit or mark the spam.
>>>>
>>>> It seems alright if it'll be a sort of clean break:  on Feb 22 according to Google,
>>>> Google will break its compeerage to Usenet, and furthermore make read-only
>>>> the archives, what it has, what until then, will be as it was.
>>>>
>>>> Over on sci.math I've had the idea for a while of making some brief and
>>>> special purpose Usenet compeers, for only some few groups, or, you
>>>> know, the _belles lettres_ of the text hierarchy.
>>>>
>>>> "Meta:  a usenet server just for sci.math"
>>>>     -- https://groups.google.com/g/sci.math/c/zggff_pVEks
>>>>
>>>> So, there you can read the outlook of this kind of thing, then while sort
>>>> of simple as the protocol is simple and its implementations widespread,
>>>> how to deal with the "signal and noise" of "exposed messaging destinations
>>>> on the Internet", well on that thread I'm theorizing a sort of, "NOOBNB protocol",
>>>> figuring to make an otherwise just standard Usenet compeer, and also for
>>>> email or messaging destinations, sort of designed with the expectation that
>>>> there will be spam, and spam and ham are hand in hand, to exclude it in simple terms.
>>>>
>>>> NOOBNB:  New Old Off Bot Non Bad, Curated/Purgatory/Raw triple-feed
>>>>
>>>> (That and a firmer sort of "Load Shed" or "Load Hold" at the transport layer.)
>>>>
>>>> Also it would be real great if at least there was surfaced to the Internet a
>>>> read-only view of any message by its message ID, a "URL", or as for a "URI",
>>>> a "URN", a reliable perma-link in the IETF "news" protocol, namespace.
>>>>
>>>> https://groups.google.com/g/sci.math/c/zggff_pVEks
>>>>
>>>> I wonder that there's a reliable sort of long-term project that surfaces
>>>> "news" protocol message-IDs, ....  It's a stable, standards-based protocol.
>>>>
>>>>
>>>> Thunderbird, "SLRN", ....  Thanks for caring.  We care.
>>>>
>>>>
>>>> https://groups.google.com/g/sci.physics.relativity/c/ToBo6XOymUw
>>>>
>>>
>>> One fellow reached me via e-mail and he said, hey, the Googler spam is
>>> outrageous, can we do anything about it?  Would you write a script to
>>> funnel all their message-ID's into the abuse reporting?  And I was like,
>>> you know, about 2008 I did just that, there was a big spam flood,
>>> and I wrote a little script to find them and extract their posting-account,
>>> and the message-ID, and a little script to post to the posting-host,
>>> each one of the wicked spams.
>>>
>>> At the time that seemed to help, they sort of dried up, here there's
>>> that basically they're not following the charter, but, it's the
>>> posting-account
>>> in the message headers that indicate the origin of the post, not the
>>> email address.  So, I wonder, given that I can extract the posting-accounts
>>> of all the spams, how to match the posting-account to then determine
>>> whether it's a sockpuppet-farm or what, and basically about sending them up.
>>
>> Let me see your little script. Post it here.
>
> Here is a list I currently have:
>
> salz.txt
> usenet.death.penalty.gz
> purify.txt
> NewsAgent110-MS.exe
> HipCrime's NewsAgent (v1_11).htm
> NewsAgent111-BE.zip
> SuperCede.exe
> NewsAgent023.exe
> NewsAgent025.exe
> ActiveAgent.java
> HipCrime's NewsAgent (v1_02)_files
> NewsCancel.java (source code)
>
> (plus updated python versions)
>
>
>
> (Maybe your script is inthere somewhere?)
>
>
>
> Show me what you got. walk the walk.
>


I try to avoid sketchy things like hiring a criminal botnet,
there's the impression that that's looking at 1000's of counts
of computer intrusion.

With those being something about $50K and 10-25 apiece,
there's a pretty significant deterrence to such activities.

I've never much cared for "OAuth", giving away the
keys-to-the-kingdom and all, here it looks like either
a) a bunch of duped browsers clicked away their identities,
or b) it's really that Google and Facebook are more than
half full of fake identities for the sole purpose of being fake.

(How's your new deal going?
   Great, we got a million users.
Why are my conversions around zero?
   Your ad must not speak to them.
Would it help if I spiced it up?
   Don't backtalk me, I'll put you on a list!)

So, it seems mostly a sort of "spam-walling the Internet",
where it was like "we're going to reinvent the Internet",
"no, you aren't", "all right then we'll ruin this one".

As far as search goes, there's something to be said
for a new sort of approach to search, given that
Google, Bing, Duck, ..., _all make the same results_.  It's
just so highly unlikely that they'd _all make the same
results_, you figure they're just one.

So, the idea, for somebody like me who's mostly interested
in writing on the Internet, is that lots of that is of the sort
of "works" vis-a-vis, the "feuilleton" or what you might
call it, ephemeral junk, that I just learned about in
Herman Hesse's "The Glass Bead Game".

Then, there's an idea, that basically to surface high-quality
works to a search, is that there's what's called metadata,
for content like HTML, with regards to Dublin Core and
RDF and so on, about a sort of making for fungible collections
of works, what results searchable fragments of various
larger bodies of works, according to their robots.txt and
their summaries and with regards to crawling the content
and so on, then to make federated common search corpi,
these kinds of things.

[toc] | [prev] | [next] | [standalone]

#637450 — Meta: Re: How I deal with the enormous amount of spam

On 02/04/2024 09:55 AM, Ross Finlayson wrote:
> On 02/03/2024 02:46 PM, The Starmaker wrote:
>> The Starmaker wrote:
>>>
>>> Ross Finlayson wrote:
>>>>
>>>> On 01/30/2024 12:54 PM, Ross Finlayson wrote:
>>>>> On Monday, January 29, 2024 at 5:02:05 PM UTC-8, palsing wrote:
>>>>>> Tom Roberts wrote:
>>>>>>
>>>>>>> I use Thunderbird to read Usenet. Recently sci.physics.relativity
>>>>>>> has
>>>>>>> been getting hundreds of spam posts each day, completely
>>>>>>> overwhelming
>>>>>>> legitimate content. These spam posts share the property that they
>>>>>>> are
>>>>>>> written in a non-latin script.
>>>>>>
>>>>>>> Thunderbird implements message filters that can mark a message
>>>>>>> Read. So
>>>>>>> I created a filter to run on sci.physics.relativity that marks
>>>>>>> messages
>>>>>>> Read. Then when reading the newsgroups, I simply display only unread
>>>>>>> messages. The key to making this work is to craft the filter so
>>>>>>> it marks
>>>>>>> messages in which the Subject matches any of a dozen characters
>>>>>>> picked
>>>>>>> from some spam messages.
>>>>>>
>>>>>>> This doesn't completely eliminate the spam, but it is now only a few
>>>>>>> messages per day.
>>>>>>
>>>>>>> Tom Roberts
>>>>>> I would like to do the same thing, so I installed Thunderbird...
>>>>>> but setting it up to read newsgroups is beyond my paltry computer
>>>>>> skills and is not at all intuitive. If anyone can point to an
>>>>>> idiot-proof tutorial for doing this It would be much appreciated.
>>>>>>
>>>>>> \Paul Alsing
>>>>>
>>>>> Yeah, it's pretty bad, or, worse anybody's ever seen it.
>>>>>
>>>>> I as well sort of mow the lawn a bit or mark the spam.
>>>>>
>>>>> It seems alright if it'll be a sort of clean break:  on Feb 22
>>>>> according to Google,
>>>>> Google will break its compeerage to Usenet, and furthermore make
>>>>> read-only
>>>>> the archives, what it has, what until then, will be as it was.
>>>>>
>>>>> Over on sci.math I've had the idea for a while of making some brief
>>>>> and
>>>>> special purpose Usenet compeers, for only some few groups, or, you
>>>>> know, the _belles lettres_ of the text hierarchy.
>>>>>
>>>>> "Meta:  a usenet server just for sci.math"
>>>>>     -- https://groups.google.com/g/sci.math/c/zggff_pVEks
>>>>>
>>>>> So, there you can read the outlook of this kind of thing, then
>>>>> while sort
>>>>> of simple as the protocol is simple and its implementations
>>>>> widespread,
>>>>> how to deal with the "signal and noise" of "exposed messaging
>>>>> destinations
>>>>> on the Internet", well on that thread I'm theorizing a sort of,
>>>>> "NOOBNB protocol",
>>>>> figuring to make an otherwise just standard Usenet compeer, and
>>>>> also for
>>>>> email or messaging destinations, sort of designed with the
>>>>> expectation that
>>>>> there will be spam, and spam and ham are hand in hand, to exclude
>>>>> it in simple terms.
>>>>>
>>>>> NOOBNB:  New Old Off Bot Non Bad, Curated/Purgatory/Raw triple-feed
>>>>>
>>>>> (That and a firmer sort of "Load Shed" or "Load Hold" at the
>>>>> transport layer.)
>>>>>
>>>>> Also it would be real great if at least there was surfaced to the
>>>>> Internet a
>>>>> read-only view of any message by its message ID, a "URL", or as for
>>>>> a "URI",
>>>>> a "URN", a reliable perma-link in the IETF "news" protocol, namespace.
>>>>>
>>>>> https://groups.google.com/g/sci.math/c/zggff_pVEks
>>>>>
>>>>> I wonder that there's a reliable sort of long-term project that
>>>>> surfaces
>>>>> "news" protocol message-IDs, ....  It's a stable, standards-based
>>>>> protocol.
>>>>>
>>>>>
>>>>> Thunderbird, "SLRN", ....  Thanks for caring.  We care.
>>>>>
>>>>>
>>>>> https://groups.google.com/g/sci.physics.relativity/c/ToBo6XOymUw
>>>>>
>>>>
>>>> One fellow reached me via e-mail and he said, hey, the Googler spam is
>>>> outrageous, can we do anything about it?  Would you write a script to
>>>> funnel all their message-ID's into the abuse reporting?  And I was
>>>> like,
>>>> you know, about 2008 I did just that, there was a big spam flood,
>>>> and I wrote a little script to find them and extract their
>>>> posting-account,
>>>> and the message-ID, and a little script to post to the posting-host,
>>>> each one of the wicked spams.
>>>>
>>>> At the time that seemed to help, they sort of dried up, here there's
>>>> that basically they're not following the charter, but, it's the
>>>> posting-account
>>>> in the message headers that indicate the origin of the post, not the
>>>> email address.  So, I wonder, given that I can extract the
>>>> posting-accounts
>>>> of all the spams, how to match the posting-account to then determine
>>>> whether it's a sockpuppet-farm or what, and basically about sending
>>>> them up.
>>>
>>> Let me see your little script. Post it here.
>>
>> Here is a list I currently have:
>>
>> salz.txt
>> usenet.death.penalty.gz
>> purify.txt
>> NewsAgent110-MS.exe
>> HipCrime's NewsAgent (v1_11).htm
>> NewsAgent111-BE.zip
>> SuperCede.exe
>> NewsAgent023.exe
>> NewsAgent025.exe
>> ActiveAgent.java
>> HipCrime's NewsAgent (v1_02)_files
>> NewsCancel.java (source code)
>>
>> (plus updated python versions)
>>
>>
>>
>> (Maybe your script is inthere somewhere?)
>>
>>
>>
>> Show me what you got. walk the walk.
>>
>
>
> I try to avoid sketchy things like hiring a criminal botnet,
> there's the impression that that's looking at 1000's of counts
> of computer intrusion.
>
> With those being something about $50K and 10-25 apiece,
> there's a pretty significant deterrence to such activities.
>
> I've never much cared for "OAuth", giving away the
> keys-to-the-kingdom and all, here it looks like either
> a) a bunch of duped browsers clicked away their identities,
> or b) it's really that Google and Facebook are more than
> half full of fake identities for the sole purpose of being fake.
>
> (How's your new deal going?
>    Great, we got a million users.
> Why are my conversions around zero?
>    Your ad must not speak to them.
> Would it help if I spiced it up?
>    Don't backtalk me, I'll put you on a list!)
>
> So, it seems mostly a sort of "spam-walling the Internet",
> where it was like "we're going to reinvent the Internet",
> "no, you aren't", "all right then we'll ruin this one".
>
> As far as search goes, there's something to be said
> for a new sort of approach to search, given that
> Google, Bing, Duck, ..., _all make the same results_.  It's
> just so highly unlikely that they'd _all make the same
> results_, you figure they're just one.
>
> So, the idea, for somebody like me who's mostly interested
> in writing on the Internet, is that lots of that is of the sort
> of "works" vis-a-vis, the "feuilleton" or what you might
> call it, ephemeral junk, that I just learned about in
> Herman Hesse's "The Glass Bead Game".
>
> Then, there's an idea, that basically to surface high-quality
> works to a search, is that there's what's called metadata,
> for content like HTML, with regards to Dublin Core and
> RDF and so on, about a sort of making for fungible collections
> of works, what results searchable fragments of various
> larger bodies of works, according to their robots.txt and
> their summaries and with regards to crawling the content
> and so on, then to make federated common search corpi,
> these kinds of things.
>
>
>


It's like "why are they building that new data center",
and it's like "well it's like Artificial Intelligence, inside
that data center is a million virts and each one has a
browser emulator and a phone app sandbox and a
little notecard that prompts its name, basically it's
a million-headed hydra called a sims-bot-farm,
that for pennies on the dollar is an instant audience."

"Wow, great, do they get a cut?"  "Don't be talking about my cut."

Usenet traffic had been up recently, ....

I think they used to call it "astro-turfing".
"Artificial Intelligence?"  "No, 'Fake eyeballs'."

[toc] | [prev] | [next] | [standalone]

#637543 — Re: Meta: Re: How I deal with the enormous amount of spam

Ross Finlayson wrote:
> 
> On 02/04/2024 09:55 AM, Ross Finlayson wrote:
> > On 02/03/2024 02:46 PM, The Starmaker wrote:
> >> The Starmaker wrote:
> >>>
> >>> Ross Finlayson wrote:
> >>>>
> >>>> On 01/30/2024 12:54 PM, Ross Finlayson wrote:
> >>>>> On Monday, January 29, 2024 at 5:02:05 PM UTC-8, palsing wrote:
> >>>>>> Tom Roberts wrote:
> >>>>>>
> >>>>>>> I use Thunderbird to read Usenet. Recently sci.physics.relativity
> >>>>>>> has
> >>>>>>> been getting hundreds of spam posts each day, completely
> >>>>>>> overwhelming
> >>>>>>> legitimate content. These spam posts share the property that they
> >>>>>>> are
> >>>>>>> written in a non-latin script.
> >>>>>>
> >>>>>>> Thunderbird implements message filters that can mark a message
> >>>>>>> Read. So
> >>>>>>> I created a filter to run on sci.physics.relativity that marks
> >>>>>>> messages
> >>>>>>> Read. Then when reading the newsgroups, I simply display only unread
> >>>>>>> messages. The key to making this work is to craft the filter so
> >>>>>>> it marks
> >>>>>>> messages in which the Subject matches any of a dozen characters
> >>>>>>> picked
> >>>>>>> from some spam messages.
> >>>>>>
> >>>>>>> This doesn't completely eliminate the spam, but it is now only a few
> >>>>>>> messages per day.
> >>>>>>
> >>>>>>> Tom Roberts
> >>>>>> I would like to do the same thing, so I installed Thunderbird...
> >>>>>> but setting it up to read newsgroups is beyond my paltry computer
> >>>>>> skills and is not at all intuitive. If anyone can point to an
> >>>>>> idiot-proof tutorial for doing this It would be much appreciated.
> >>>>>>
> >>>>>> \Paul Alsing
> >>>>>
> >>>>> Yeah, it's pretty bad, or, worse anybody's ever seen it.
> >>>>>
> >>>>> I as well sort of mow the lawn a bit or mark the spam.
> >>>>>
> >>>>> It seems alright if it'll be a sort of clean break:  on Feb 22
> >>>>> according to Google,
> >>>>> Google will break its compeerage to Usenet, and furthermore make
> >>>>> read-only
> >>>>> the archives, what it has, what until then, will be as it was.
> >>>>>
> >>>>> Over on sci.math I've had the idea for a while of making some brief
> >>>>> and
> >>>>> special purpose Usenet compeers, for only some few groups, or, you
> >>>>> know, the _belles lettres_ of the text hierarchy.
> >>>>>
> >>>>> "Meta:  a usenet server just for sci.math"
> >>>>>     -- https://groups.google.com/g/sci.math/c/zggff_pVEks
> >>>>>
> >>>>> So, there you can read the outlook of this kind of thing, then
> >>>>> while sort
> >>>>> of simple as the protocol is simple and its implementations
> >>>>> widespread,
> >>>>> how to deal with the "signal and noise" of "exposed messaging
> >>>>> destinations
> >>>>> on the Internet", well on that thread I'm theorizing a sort of,
> >>>>> "NOOBNB protocol",
> >>>>> figuring to make an otherwise just standard Usenet compeer, and
> >>>>> also for
> >>>>> email or messaging destinations, sort of designed with the
> >>>>> expectation that
> >>>>> there will be spam, and spam and ham are hand in hand, to exclude
> >>>>> it in simple terms.
> >>>>>
> >>>>> NOOBNB:  New Old Off Bot Non Bad, Curated/Purgatory/Raw triple-feed
> >>>>>
> >>>>> (That and a firmer sort of "Load Shed" or "Load Hold" at the
> >>>>> transport layer.)
> >>>>>
> >>>>> Also it would be real great if at least there was surfaced to the
> >>>>> Internet a
> >>>>> read-only view of any message by its message ID, a "URL", or as for
> >>>>> a "URI",
> >>>>> a "URN", a reliable perma-link in the IETF "news" protocol, namespace.
> >>>>>
> >>>>> https://groups.google.com/g/sci.math/c/zggff_pVEks
> >>>>>
> >>>>> I wonder that there's a reliable sort of long-term project that
> >>>>> surfaces
> >>>>> "news" protocol message-IDs, ....  It's a stable, standards-based
> >>>>> protocol.
> >>>>>
> >>>>>
> >>>>> Thunderbird, "SLRN", ....  Thanks for caring.  We care.
> >>>>>
> >>>>>
> >>>>> https://groups.google.com/g/sci.physics.relativity/c/ToBo6XOymUw
> >>>>>
> >>>>
> >>>> One fellow reached me via e-mail and he said, hey, the Googler spam is
> >>>> outrageous, can we do anything about it?  Would you write a script to
> >>>> funnel all their message-ID's into the abuse reporting?  And I was
> >>>> like,
> >>>> you know, about 2008 I did just that, there was a big spam flood,
> >>>> and I wrote a little script to find them and extract their
> >>>> posting-account,
> >>>> and the message-ID, and a little script to post to the posting-host,
> >>>> each one of the wicked spams.
> >>>>
> >>>> At the time that seemed to help, they sort of dried up, here there's
> >>>> that basically they're not following the charter, but, it's the
> >>>> posting-account
> >>>> in the message headers that indicate the origin of the post, not the
> >>>> email address.  So, I wonder, given that I can extract the
> >>>> posting-accounts
> >>>> of all the spams, how to match the posting-account to then determine
> >>>> whether it's a sockpuppet-farm or what, and basically about sending
> >>>> them up.
> >>>
> >>> Let me see your little script. Post it here.
> >>
> >> Here is a list I currently have:
> >>
> >> salz.txt
> >> usenet.death.penalty.gz
> >> purify.txt
> >> NewsAgent110-MS.exe
> >> HipCrime's NewsAgent (v1_11).htm
> >> NewsAgent111-BE.zip
> >> SuperCede.exe
> >> NewsAgent023.exe
> >> NewsAgent025.exe
> >> ActiveAgent.java
> >> HipCrime's NewsAgent (v1_02)_files
> >> NewsCancel.java (source code)
> >>
> >> (plus updated python versions)
> >>
> >>
> >>
> >> (Maybe your script is inthere somewhere?)
> >>
> >>
> >>
> >> Show me what you got. walk the walk.
> >>
> >
> >
> > I try to avoid sketchy things like hiring a criminal botnet,
> > there's the impression that that's looking at 1000's of counts
> > of computer intrusion.
> >
> > With those being something about $50K and 10-25 apiece,
> > there's a pretty significant deterrence to such activities.
> >
> > I've never much cared for "OAuth", giving away the
> > keys-to-the-kingdom and all, here it looks like either
> > a) a bunch of duped browsers clicked away their identities,
> > or b) it's really that Google and Facebook are more than
> > half full of fake identities for the sole purpose of being fake.
> >
> > (How's your new deal going?
> >    Great, we got a million users.
> > Why are my conversions around zero?
> >    Your ad must not speak to them.
> > Would it help if I spiced it up?
> >    Don't backtalk me, I'll put you on a list!)
> >
> > So, it seems mostly a sort of "spam-walling the Internet",
> > where it was like "we're going to reinvent the Internet",
> > "no, you aren't", "all right then we'll ruin this one".
> >
> > As far as search goes, there's something to be said
> > for a new sort of approach to search, given that
> > Google, Bing, Duck, ..., _all make the same results_.  It's
> > just so highly unlikely that they'd _all make the same
> > results_, you figure they're just one.
> >
> > So, the idea, for somebody like me who's mostly interested
> > in writing on the Internet, is that lots of that is of the sort
> > of "works" vis-a-vis, the "feuilleton" or what you might
> > call it, ephemeral junk, that I just learned about in
> > Herman Hesse's "The Glass Bead Game".
> >
> > Then, there's an idea, that basically to surface high-quality
> > works to a search, is that there's what's called metadata,
> > for content like HTML, with regards to Dublin Core and
> > RDF and so on, about a sort of making for fungible collections
> > of works, what results searchable fragments of various
> > larger bodies of works, according to their robots.txt and
> > their summaries and with regards to crawling the content
> > and so on, then to make federated common search corpi,
> > these kinds of things.
> >
> >
> >
> 
> It's like "why are they building that new data center",
> and it's like "well it's like Artificial Intelligence, inside
> that data center is a million virts and each one has a
> browser emulator and a phone app sandbox and a
> little notecard that prompts its name, basically it's
> a million-headed hydra called a sims-bot-farm,
> that for pennies on the dollar is an instant audience."
> 
> "Wow, great, do they get a cut?"  "Don't be talking about my cut."
> 
> Usenet traffic had been up recently, ....
> 
> I think they used to call it "astro-turfing".
> "Artificial Intelligence?"  "No, 'Fake eyeballs'."


I have NewsAgent111-MS.exe

I seem to be missing version 2.0

Do you have the 2.0 version?

I'll trade you.

I'll give you my python version with (GUI)!!!! (Tinter)


let's trade!


don't bogart



-- 
The Starmaker -- To question the unquestionable, ask the unaskable,
to think the unthinkable, mention the unmentionable, say the unsayable, 
and challenge the unchallengeable.

[toc] | [prev] | [next] | [standalone]

#637551 — Re: Meta: Re: How I deal with the enormous amount of spam

The Starmaker wrote:
> 
> Ross Finlayson wrote:
> >
> > On 02/04/2024 09:55 AM, Ross Finlayson wrote:
> > > On 02/03/2024 02:46 PM, The Starmaker wrote:
> > >> The Starmaker wrote:
> > >>>
> > >>> Ross Finlayson wrote:
> > >>>>
> > >>>> On 01/30/2024 12:54 PM, Ross Finlayson wrote:
> > >>>>> On Monday, January 29, 2024 at 5:02:05 PM UTC-8, palsing wrote:
> > >>>>>> Tom Roberts wrote:
> > >>>>>>
> > >>>>>>> I use Thunderbird to read Usenet. Recently sci.physics.relativity
> > >>>>>>> has
> > >>>>>>> been getting hundreds of spam posts each day, completely
> > >>>>>>> overwhelming
> > >>>>>>> legitimate content. These spam posts share the property that they
> > >>>>>>> are
> > >>>>>>> written in a non-latin script.
> > >>>>>>
> > >>>>>>> Thunderbird implements message filters that can mark a message
> > >>>>>>> Read. So
> > >>>>>>> I created a filter to run on sci.physics.relativity that marks
> > >>>>>>> messages
> > >>>>>>> Read. Then when reading the newsgroups, I simply display only unread
> > >>>>>>> messages. The key to making this work is to craft the filter so
> > >>>>>>> it marks
> > >>>>>>> messages in which the Subject matches any of a dozen characters
> > >>>>>>> picked
> > >>>>>>> from some spam messages.
> > >>>>>>
> > >>>>>>> This doesn't completely eliminate the spam, but it is now only a few
> > >>>>>>> messages per day.
> > >>>>>>
> > >>>>>>> Tom Roberts
> > >>>>>> I would like to do the same thing, so I installed Thunderbird...
> > >>>>>> but setting it up to read newsgroups is beyond my paltry computer
> > >>>>>> skills and is not at all intuitive. If anyone can point to an
> > >>>>>> idiot-proof tutorial for doing this It would be much appreciated.
> > >>>>>>
> > >>>>>> \Paul Alsing
> > >>>>>
> > >>>>> Yeah, it's pretty bad, or, worse anybody's ever seen it.
> > >>>>>
> > >>>>> I as well sort of mow the lawn a bit or mark the spam.
> > >>>>>
> > >>>>> It seems alright if it'll be a sort of clean break:  on Feb 22
> > >>>>> according to Google,
> > >>>>> Google will break its compeerage to Usenet, and furthermore make
> > >>>>> read-only
> > >>>>> the archives, what it has, what until then, will be as it was.
> > >>>>>
> > >>>>> Over on sci.math I've had the idea for a while of making some brief
> > >>>>> and
> > >>>>> special purpose Usenet compeers, for only some few groups, or, you
> > >>>>> know, the _belles lettres_ of the text hierarchy.
> > >>>>>
> > >>>>> "Meta:  a usenet server just for sci.math"
> > >>>>>     -- https://groups.google.com/g/sci.math/c/zggff_pVEks
> > >>>>>
> > >>>>> So, there you can read the outlook of this kind of thing, then
> > >>>>> while sort
> > >>>>> of simple as the protocol is simple and its implementations
> > >>>>> widespread,
> > >>>>> how to deal with the "signal and noise" of "exposed messaging
> > >>>>> destinations
> > >>>>> on the Internet", well on that thread I'm theorizing a sort of,
> > >>>>> "NOOBNB protocol",
> > >>>>> figuring to make an otherwise just standard Usenet compeer, and
> > >>>>> also for
> > >>>>> email or messaging destinations, sort of designed with the
> > >>>>> expectation that
> > >>>>> there will be spam, and spam and ham are hand in hand, to exclude
> > >>>>> it in simple terms.
> > >>>>>
> > >>>>> NOOBNB:  New Old Off Bot Non Bad, Curated/Purgatory/Raw triple-feed
> > >>>>>
> > >>>>> (That and a firmer sort of "Load Shed" or "Load Hold" at the
> > >>>>> transport layer.)
> > >>>>>
> > >>>>> Also it would be real great if at least there was surfaced to the
> > >>>>> Internet a
> > >>>>> read-only view of any message by its message ID, a "URL", or as for
> > >>>>> a "URI",
> > >>>>> a "URN", a reliable perma-link in the IETF "news" protocol, namespace.
> > >>>>>
> > >>>>> https://groups.google.com/g/sci.math/c/zggff_pVEks
> > >>>>>
> > >>>>> I wonder that there's a reliable sort of long-term project that
> > >>>>> surfaces
> > >>>>> "news" protocol message-IDs, ....  It's a stable, standards-based
> > >>>>> protocol.
> > >>>>>
> > >>>>>
> > >>>>> Thunderbird, "SLRN", ....  Thanks for caring.  We care.
> > >>>>>
> > >>>>>
> > >>>>> https://groups.google.com/g/sci.physics.relativity/c/ToBo6XOymUw
> > >>>>>
> > >>>>
> > >>>> One fellow reached me via e-mail and he said, hey, the Googler spam is
> > >>>> outrageous, can we do anything about it?  Would you write a script to
> > >>>> funnel all their message-ID's into the abuse reporting?  And I was
> > >>>> like,
> > >>>> you know, about 2008 I did just that, there was a big spam flood,
> > >>>> and I wrote a little script to find them and extract their
> > >>>> posting-account,
> > >>>> and the message-ID, and a little script to post to the posting-host,
> > >>>> each one of the wicked spams.
> > >>>>
> > >>>> At the time that seemed to help, they sort of dried up, here there's
> > >>>> that basically they're not following the charter, but, it's the
> > >>>> posting-account
> > >>>> in the message headers that indicate the origin of the post, not the
> > >>>> email address.  So, I wonder, given that I can extract the
> > >>>> posting-accounts
> > >>>> of all the spams, how to match the posting-account to then determine
> > >>>> whether it's a sockpuppet-farm or what, and basically about sending
> > >>>> them up.
> > >>>
> > >>> Let me see your little script. Post it here.
> > >>
> > >> Here is a list I currently have:
> > >>
> > >> salz.txt
> > >> usenet.death.penalty.gz
> > >> purify.txt
> > >> NewsAgent110-MS.exe
> > >> HipCrime's NewsAgent (v1_11).htm
> > >> NewsAgent111-BE.zip
> > >> SuperCede.exe
> > >> NewsAgent023.exe
> > >> NewsAgent025.exe
> > >> ActiveAgent.java
> > >> HipCrime's NewsAgent (v1_02)_files
> > >> NewsCancel.java (source code)
> > >>
> > >> (plus updated python versions)
> > >>
> > >>
> > >>
> > >> (Maybe your script is inthere somewhere?)
> > >>
> > >>
> > >>
> > >> Show me what you got. walk the walk.
> > >>
> > >
> > >
> > > I try to avoid sketchy things like hiring a criminal botnet,
> > > there's the impression that that's looking at 1000's of counts
> > > of computer intrusion.
> > >
> > > With those being something about $50K and 10-25 apiece,
> > > there's a pretty significant deterrence to such activities.
> > >
> > > I've never much cared for "OAuth", giving away the
> > > keys-to-the-kingdom and all, here it looks like either
> > > a) a bunch of duped browsers clicked away their identities,
> > > or b) it's really that Google and Facebook are more than
> > > half full of fake identities for the sole purpose of being fake.
> > >
> > > (How's your new deal going?
> > >    Great, we got a million users.
> > > Why are my conversions around zero?
> > >    Your ad must not speak to them.
> > > Would it help if I spiced it up?
> > >    Don't backtalk me, I'll put you on a list!)
> > >
> > > So, it seems mostly a sort of "spam-walling the Internet",
> > > where it was like "we're going to reinvent the Internet",
> > > "no, you aren't", "all right then we'll ruin this one".
> > >
> > > As far as search goes, there's something to be said
> > > for a new sort of approach to search, given that
> > > Google, Bing, Duck, ..., _all make the same results_.  It's
> > > just so highly unlikely that they'd _all make the same
> > > results_, you figure they're just one.
> > >
> > > So, the idea, for somebody like me who's mostly interested
> > > in writing on the Internet, is that lots of that is of the sort
> > > of "works" vis-a-vis, the "feuilleton" or what you might
> > > call it, ephemeral junk, that I just learned about in
> > > Herman Hesse's "The Glass Bead Game".
> > >
> > > Then, there's an idea, that basically to surface high-quality
> > > works to a search, is that there's what's called metadata,
> > > for content like HTML, with regards to Dublin Core and
> > > RDF and so on, about a sort of making for fungible collections
> > > of works, what results searchable fragments of various
> > > larger bodies of works, according to their robots.txt and
> > > their summaries and with regards to crawling the content
> > > and so on, then to make federated common search corpi,
> > > these kinds of things.
> > >
> > >
> > >
> >
> > It's like "why are they building that new data center",
> > and it's like "well it's like Artificial Intelligence, inside
> > that data center is a million virts and each one has a
> > browser emulator and a phone app sandbox and a
> > little notecard that prompts its name, basically it's
> > a million-headed hydra called a sims-bot-farm,
> > that for pennies on the dollar is an instant audience."
> >
> > "Wow, great, do they get a cut?"  "Don't be talking about my cut."
> >
> > Usenet traffic had been up recently, ....
> >
> > I think they used to call it "astro-turfing".
> > "Artificial Intelligence?"  "No, 'Fake eyeballs'."
> 
> I have NewsAgent111-MS.exe
> 
> I seem to be missing version 2.0
> 
> Do you have the 2.0 version?
> 
> I'll trade you.
> 
> I'll give you my python version with (GUI)!!!! (Tinter)
> 
> let's trade!
> 
> don't bogart

I seem to be missing this version:

https://web.archive.org/web/20051023050609/http://newsagent.p5.org.uk/

Do you have it? you must have! 





-- 
The Starmaker -- To question the unquestionable, ask the unaskable,
to think the unthinkable, mention the unmentionable, say the unsayable, 
and challenge the unchallengeable.

[toc] | [prev] | [next] | [standalone]

#637690 — Re: Meta: Re: How I deal with the enormous amount of spam

On 02/04/2024 12:53 PM, The Starmaker wrote:
> The Starmaker wrote:
>>
>> Ross Finlayson wrote:
>>>
>>> On 02/04/2024 09:55 AM, Ross Finlayson wrote:
>>>> On 02/03/2024 02:46 PM, The Starmaker wrote:
>>>>> The Starmaker wrote:
>>>>>>
>>>>>> Ross Finlayson wrote:
>>>>>>>
>>>>>>> On 01/30/2024 12:54 PM, Ross Finlayson wrote:
>>>>>>>> On Monday, January 29, 2024 at 5:02:05 PM UTC-8, palsing wrote:
>>>>>>>>> Tom Roberts wrote:
>>>>>>>>>
>>>>>>>>>> I use Thunderbird to read Usenet. Recently sci.physics.relativity
>>>>>>>>>> has
>>>>>>>>>> been getting hundreds of spam posts each day, completely
>>>>>>>>>> overwhelming
>>>>>>>>>> legitimate content. These spam posts share the property that they
>>>>>>>>>> are
>>>>>>>>>> written in a non-latin script.
>>>>>>>>>
>>>>>>>>>> Thunderbird implements message filters that can mark a message
>>>>>>>>>> Read. So
>>>>>>>>>> I created a filter to run on sci.physics.relativity that marks
>>>>>>>>>> messages
>>>>>>>>>> Read. Then when reading the newsgroups, I simply display only unread
>>>>>>>>>> messages. The key to making this work is to craft the filter so
>>>>>>>>>> it marks
>>>>>>>>>> messages in which the Subject matches any of a dozen characters
>>>>>>>>>> picked
>>>>>>>>>> from some spam messages.
>>>>>>>>>
>>>>>>>>>> This doesn't completely eliminate the spam, but it is now only a few
>>>>>>>>>> messages per day.
>>>>>>>>>
>>>>>>>>>> Tom Roberts
>>>>>>>>> I would like to do the same thing, so I installed Thunderbird...
>>>>>>>>> but setting it up to read newsgroups is beyond my paltry computer
>>>>>>>>> skills and is not at all intuitive. If anyone can point to an
>>>>>>>>> idiot-proof tutorial for doing this It would be much appreciated.
>>>>>>>>>
>>>>>>>>> \Paul Alsing
>>>>>>>>
>>>>>>>> Yeah, it's pretty bad, or, worse anybody's ever seen it.
>>>>>>>>
>>>>>>>> I as well sort of mow the lawn a bit or mark the spam.
>>>>>>>>
>>>>>>>> It seems alright if it'll be a sort of clean break:  on Feb 22
>>>>>>>> according to Google,
>>>>>>>> Google will break its compeerage to Usenet, and furthermore make
>>>>>>>> read-only
>>>>>>>> the archives, what it has, what until then, will be as it was.
>>>>>>>>
>>>>>>>> Over on sci.math I've had the idea for a while of making some brief
>>>>>>>> and
>>>>>>>> special purpose Usenet compeers, for only some few groups, or, you
>>>>>>>> know, the _belles lettres_ of the text hierarchy.
>>>>>>>>
>>>>>>>> "Meta:  a usenet server just for sci.math"
>>>>>>>>      -- https://groups.google.com/g/sci.math/c/zggff_pVEks
>>>>>>>>
>>>>>>>> So, there you can read the outlook of this kind of thing, then
>>>>>>>> while sort
>>>>>>>> of simple as the protocol is simple and its implementations
>>>>>>>> widespread,
>>>>>>>> how to deal with the "signal and noise" of "exposed messaging
>>>>>>>> destinations
>>>>>>>> on the Internet", well on that thread I'm theorizing a sort of,
>>>>>>>> "NOOBNB protocol",
>>>>>>>> figuring to make an otherwise just standard Usenet compeer, and
>>>>>>>> also for
>>>>>>>> email or messaging destinations, sort of designed with the
>>>>>>>> expectation that
>>>>>>>> there will be spam, and spam and ham are hand in hand, to exclude
>>>>>>>> it in simple terms.
>>>>>>>>
>>>>>>>> NOOBNB:  New Old Off Bot Non Bad, Curated/Purgatory/Raw triple-feed
>>>>>>>>
>>>>>>>> (That and a firmer sort of "Load Shed" or "Load Hold" at the
>>>>>>>> transport layer.)
>>>>>>>>
>>>>>>>> Also it would be real great if at least there was surfaced to the
>>>>>>>> Internet a
>>>>>>>> read-only view of any message by its message ID, a "URL", or as for
>>>>>>>> a "URI",
>>>>>>>> a "URN", a reliable perma-link in the IETF "news" protocol, namespace.
>>>>>>>>
>>>>>>>> https://groups.google.com/g/sci.math/c/zggff_pVEks
>>>>>>>>
>>>>>>>> I wonder that there's a reliable sort of long-term project that
>>>>>>>> surfaces
>>>>>>>> "news" protocol message-IDs, ....  It's a stable, standards-based
>>>>>>>> protocol.
>>>>>>>>
>>>>>>>>
>>>>>>>> Thunderbird, "SLRN", ....  Thanks for caring.  We care.
>>>>>>>>
>>>>>>>>
>>>>>>>> https://groups.google.com/g/sci.physics.relativity/c/ToBo6XOymUw
>>>>>>>>
>>>>>>>
>>>>>>> One fellow reached me via e-mail and he said, hey, the Googler spam is
>>>>>>> outrageous, can we do anything about it?  Would you write a script to
>>>>>>> funnel all their message-ID's into the abuse reporting?  And I was
>>>>>>> like,
>>>>>>> you know, about 2008 I did just that, there was a big spam flood,
>>>>>>> and I wrote a little script to find them and extract their
>>>>>>> posting-account,
>>>>>>> and the message-ID, and a little script to post to the posting-host,
>>>>>>> each one of the wicked spams.
>>>>>>>
>>>>>>> At the time that seemed to help, they sort of dried up, here there's
>>>>>>> that basically they're not following the charter, but, it's the
>>>>>>> posting-account
>>>>>>> in the message headers that indicate the origin of the post, not the
>>>>>>> email address.  So, I wonder, given that I can extract the
>>>>>>> posting-accounts
>>>>>>> of all the spams, how to match the posting-account to then determine
>>>>>>> whether it's a sockpuppet-farm or what, and basically about sending
>>>>>>> them up.
>>>>>>
>>>>>> Let me see your little script. Post it here.
>>>>>
>>>>> Here is a list I currently have:
>>>>>
>>>>> salz.txt
>>>>> usenet.death.penalty.gz
>>>>> purify.txt
>>>>> NewsAgent110-MS.exe
>>>>> HipCrime's NewsAgent (v1_11).htm
>>>>> NewsAgent111-BE.zip
>>>>> SuperCede.exe
>>>>> NewsAgent023.exe
>>>>> NewsAgent025.exe
>>>>> ActiveAgent.java
>>>>> HipCrime's NewsAgent (v1_02)_files
>>>>> NewsCancel.java (source code)
>>>>>
>>>>> (plus updated python versions)
>>>>>
>>>>>
>>>>>
>>>>> (Maybe your script is inthere somewhere?)
>>>>>
>>>>>
>>>>>
>>>>> Show me what you got. walk the walk.
>>>>>
>>>>
>>>>
>>>> I try to avoid sketchy things like hiring a criminal botnet,
>>>> there's the impression that that's looking at 1000's of counts
>>>> of computer intrusion.
>>>>
>>>> With those being something about $50K and 10-25 apiece,
>>>> there's a pretty significant deterrence to such activities.
>>>>
>>>> I've never much cared for "OAuth", giving away the
>>>> keys-to-the-kingdom and all, here it looks like either
>>>> a) a bunch of duped browsers clicked away their identities,
>>>> or b) it's really that Google and Facebook are more than
>>>> half full of fake identities for the sole purpose of being fake.
>>>>
>>>> (How's your new deal going?
>>>>     Great, we got a million users.
>>>> Why are my conversions around zero?
>>>>     Your ad must not speak to them.
>>>> Would it help if I spiced it up?
>>>>     Don't backtalk me, I'll put you on a list!)
>>>>
>>>> So, it seems mostly a sort of "spam-walling the Internet",
>>>> where it was like "we're going to reinvent the Internet",
>>>> "no, you aren't", "all right then we'll ruin this one".
>>>>
>>>> As far as search goes, there's something to be said
>>>> for a new sort of approach to search, given that
>>>> Google, Bing, Duck, ..., _all make the same results_.  It's
>>>> just so highly unlikely that they'd _all make the same
>>>> results_, you figure they're just one.
>>>>
>>>> So, the idea, for somebody like me who's mostly interested
>>>> in writing on the Internet, is that lots of that is of the sort
>>>> of "works" vis-a-vis, the "feuilleton" or what you might
>>>> call it, ephemeral junk, that I just learned about in
>>>> Herman Hesse's "The Glass Bead Game".
>>>>
>>>> Then, there's an idea, that basically to surface high-quality
>>>> works to a search, is that there's what's called metadata,
>>>> for content like HTML, with regards to Dublin Core and
>>>> RDF and so on, about a sort of making for fungible collections
>>>> of works, what results searchable fragments of various
>>>> larger bodies of works, according to their robots.txt and
>>>> their summaries and with regards to crawling the content
>>>> and so on, then to make federated common search corpi,
>>>> these kinds of things.
>>>>
>>>>
>>>>
>>>
>>> It's like "why are they building that new data center",
>>> and it's like "well it's like Artificial Intelligence, inside
>>> that data center is a million virts and each one has a
>>> browser emulator and a phone app sandbox and a
>>> little notecard that prompts its name, basically it's
>>> a million-headed hydra called a sims-bot-farm,
>>> that for pennies on the dollar is an instant audience."
>>>
>>> "Wow, great, do they get a cut?"  "Don't be talking about my cut."
>>>
>>> Usenet traffic had been up recently, ....
>>>
>>> I think they used to call it "astro-turfing".
>>> "Artificial Intelligence?"  "No, 'Fake eyeballs'."
>>
>> I have NewsAgent111-MS.exe
>>
>> I seem to be missing version 2.0
>>
>> Do you have the 2.0 version?
>>
>> I'll trade you.
>>
>> I'll give you my python version with (GUI)!!!! (Tinter)
>>
>> let's trade!
>>
>> don't bogart
>
> I seem to be missing this version:
>
> https://web.archive.org/web/20051023050609/http://newsagent.p5.org.uk/
>
> Do you have it? you must have!
>
>
>
>
>

Nope, I just wrote a little script to connect to NNTP
with a Yes/No button on the subject, tapped through
those, and a little script to send an HTTP request to
the publicly-facing return-to-sender in-box, for each.

Here's all the sources you need:  IETF RFC editor.
Look for "NNTP".  How to advise Google of this is
that each domain on the Internet is supposed to
have an "abuse@domain" email inbox, though there's
probably also a web request interface, as with regards
to publicly facing services, and expected to be
good actors on the network.

Anyways if you read through "Meta: a usenet server
just for sci.math", what I have in mind is a sort
of author's and writer's oriented installation,
basically making for vanity printouts and generating
hypertext collections of contents and authors and
subjects and these kinds of things, basically for
on the order of "find all the postings of Archimedes
Plutonium, and, the threads they are in, and,
make a hypertext page of all that, a linear timeline,
and also thread it out as a linear sequence".

I.e. people who actually post to Usenet are sometimes
having written interesting things, and, thus having
it so that it would be simplified to generate message-ID
listings and their corresponding standard URL's in the
standard IETF "news" URL protocol, and to point that
at a given news server or like XLink, is for treating
Usenet its archives like a living museum of all these
different authors posts and their interactions together.

I.e., here it's "belles lettres" and "fair use",
not just "belles" and "use".


It seemed nice of Google Groups to front this for a long time,
now they're quitting.

I imagine Internet Relay Chat's still insane, though.

Anyways I stay away from any warez and am proud that
since about Y2K at least I've never bootlegged anything,
and never uploaded a bootleg.  Don't want to give old Shylock
excuses, and besides, I wrote software for a living.

[toc] | [prev] | [next] | [standalone]

#637776 — Re: Meta: Re: How I deal with the enormous amount of spam

Ross Finlayson wrote:
> 
> On 02/04/2024 12:53 PM, The Starmaker wrote:
> > The Starmaker wrote:
> >>
> >> Ross Finlayson wrote:
> >>>
> >>> On 02/04/2024 09:55 AM, Ross Finlayson wrote:
> >>>> On 02/03/2024 02:46 PM, The Starmaker wrote:
> >>>>> The Starmaker wrote:
> >>>>>>
> >>>>>> Ross Finlayson wrote:
> >>>>>>>
> >>>>>>> On 01/30/2024 12:54 PM, Ross Finlayson wrote:
> >>>>>>>> On Monday, January 29, 2024 at 5:02:05 PM UTC-8, palsing wrote:
> >>>>>>>>> Tom Roberts wrote:
> >>>>>>>>>
> >>>>>>>>>> I use Thunderbird to read Usenet. Recently sci.physics.relativity
> >>>>>>>>>> has
> >>>>>>>>>> been getting hundreds of spam posts each day, completely
> >>>>>>>>>> overwhelming
> >>>>>>>>>> legitimate content. These spam posts share the property that they
> >>>>>>>>>> are
> >>>>>>>>>> written in a non-latin script.
> >>>>>>>>>
> >>>>>>>>>> Thunderbird implements message filters that can mark a message
> >>>>>>>>>> Read. So
> >>>>>>>>>> I created a filter to run on sci.physics.relativity that marks
> >>>>>>>>>> messages
> >>>>>>>>>> Read. Then when reading the newsgroups, I simply display only unread
> >>>>>>>>>> messages. The key to making this work is to craft the filter so
> >>>>>>>>>> it marks
> >>>>>>>>>> messages in which the Subject matches any of a dozen characters
> >>>>>>>>>> picked
> >>>>>>>>>> from some spam messages.
> >>>>>>>>>
> >>>>>>>>>> This doesn't completely eliminate the spam, but it is now only a few
> >>>>>>>>>> messages per day.
> >>>>>>>>>
> >>>>>>>>>> Tom Roberts
> >>>>>>>>> I would like to do the same thing, so I installed Thunderbird...
> >>>>>>>>> but setting it up to read newsgroups is beyond my paltry computer
> >>>>>>>>> skills and is not at all intuitive. If anyone can point to an
> >>>>>>>>> idiot-proof tutorial for doing this It would be much appreciated.
> >>>>>>>>>
> >>>>>>>>> \Paul Alsing
> >>>>>>>>
> >>>>>>>> Yeah, it's pretty bad, or, worse anybody's ever seen it.
> >>>>>>>>
> >>>>>>>> I as well sort of mow the lawn a bit or mark the spam.
> >>>>>>>>
> >>>>>>>> It seems alright if it'll be a sort of clean break:  on Feb 22
> >>>>>>>> according to Google,
> >>>>>>>> Google will break its compeerage to Usenet, and furthermore make
> >>>>>>>> read-only
> >>>>>>>> the archives, what it has, what until then, will be as it was.
> >>>>>>>>
> >>>>>>>> Over on sci.math I've had the idea for a while of making some brief
> >>>>>>>> and
> >>>>>>>> special purpose Usenet compeers, for only some few groups, or, you
> >>>>>>>> know, the _belles lettres_ of the text hierarchy.
> >>>>>>>>
> >>>>>>>> "Meta:  a usenet server just for sci.math"
> >>>>>>>>      -- https://groups.google.com/g/sci.math/c/zggff_pVEks
> >>>>>>>>
> >>>>>>>> So, there you can read the outlook of this kind of thing, then
> >>>>>>>> while sort
> >>>>>>>> of simple as the protocol is simple and its implementations
> >>>>>>>> widespread,
> >>>>>>>> how to deal with the "signal and noise" of "exposed messaging
> >>>>>>>> destinations
> >>>>>>>> on the Internet", well on that thread I'm theorizing a sort of,
> >>>>>>>> "NOOBNB protocol",
> >>>>>>>> figuring to make an otherwise just standard Usenet compeer, and
> >>>>>>>> also for
> >>>>>>>> email or messaging destinations, sort of designed with the
> >>>>>>>> expectation that
> >>>>>>>> there will be spam, and spam and ham are hand in hand, to exclude
> >>>>>>>> it in simple terms.
> >>>>>>>>
> >>>>>>>> NOOBNB:  New Old Off Bot Non Bad, Curated/Purgatory/Raw triple-feed
> >>>>>>>>
> >>>>>>>> (That and a firmer sort of "Load Shed" or "Load Hold" at the
> >>>>>>>> transport layer.)
> >>>>>>>>
> >>>>>>>> Also it would be real great if at least there was surfaced to the
> >>>>>>>> Internet a
> >>>>>>>> read-only view of any message by its message ID, a "URL", or as for
> >>>>>>>> a "URI",
> >>>>>>>> a "URN", a reliable perma-link in the IETF "news" protocol, namespace.
> >>>>>>>>
> >>>>>>>> https://groups.google.com/g/sci.math/c/zggff_pVEks
> >>>>>>>>
> >>>>>>>> I wonder that there's a reliable sort of long-term project that
> >>>>>>>> surfaces
> >>>>>>>> "news" protocol message-IDs, ....  It's a stable, standards-based
> >>>>>>>> protocol.
> >>>>>>>>
> >>>>>>>>
> >>>>>>>> Thunderbird, "SLRN", ....  Thanks for caring.  We care.
> >>>>>>>>
> >>>>>>>>
> >>>>>>>> https://groups.google.com/g/sci.physics.relativity/c/ToBo6XOymUw
> >>>>>>>>
> >>>>>>>
> >>>>>>> One fellow reached me via e-mail and he said, hey, the Googler spam is
> >>>>>>> outrageous, can we do anything about it?  Would you write a script to
> >>>>>>> funnel all their message-ID's into the abuse reporting?  And I was
> >>>>>>> like,
> >>>>>>> you know, about 2008 I did just that, there was a big spam flood,
> >>>>>>> and I wrote a little script to find them and extract their
> >>>>>>> posting-account,
> >>>>>>> and the message-ID, and a little script to post to the posting-host,
> >>>>>>> each one of the wicked spams.
> >>>>>>>
> >>>>>>> At the time that seemed to help, they sort of dried up, here there's
> >>>>>>> that basically they're not following the charter, but, it's the
> >>>>>>> posting-account
> >>>>>>> in the message headers that indicate the origin of the post, not the
> >>>>>>> email address.  So, I wonder, given that I can extract the
> >>>>>>> posting-accounts
> >>>>>>> of all the spams, how to match the posting-account to then determine
> >>>>>>> whether it's a sockpuppet-farm or what, and basically about sending
> >>>>>>> them up.
> >>>>>>
> >>>>>> Let me see your little script. Post it here.
> >>>>>
> >>>>> Here is a list I currently have:
> >>>>>
> >>>>> salz.txt
> >>>>> usenet.death.penalty.gz
> >>>>> purify.txt
> >>>>> NewsAgent110-MS.exe
> >>>>> HipCrime's NewsAgent (v1_11).htm
> >>>>> NewsAgent111-BE.zip
> >>>>> SuperCede.exe
> >>>>> NewsAgent023.exe
> >>>>> NewsAgent025.exe
> >>>>> ActiveAgent.java
> >>>>> HipCrime's NewsAgent (v1_02)_files
> >>>>> NewsCancel.java (source code)
> >>>>>
> >>>>> (plus updated python versions)
> >>>>>
> >>>>>
> >>>>>
> >>>>> (Maybe your script is inthere somewhere?)
> >>>>>
> >>>>>
> >>>>>
> >>>>> Show me what you got. walk the walk.
> >>>>>
> >>>>
> >>>>
> >>>> I try to avoid sketchy things like hiring a criminal botnet,
> >>>> there's the impression that that's looking at 1000's of counts
> >>>> of computer intrusion.
> >>>>
> >>>> With those being something about $50K and 10-25 apiece,
> >>>> there's a pretty significant deterrence to such activities.
> >>>>
> >>>> I've never much cared for "OAuth", giving away the
> >>>> keys-to-the-kingdom and all, here it looks like either
> >>>> a) a bunch of duped browsers clicked away their identities,
> >>>> or b) it's really that Google and Facebook are more than
> >>>> half full of fake identities for the sole purpose of being fake.
> >>>>
> >>>> (How's your new deal going?
> >>>>     Great, we got a million users.
> >>>> Why are my conversions around zero?
> >>>>     Your ad must not speak to them.
> >>>> Would it help if I spiced it up?
> >>>>     Don't backtalk me, I'll put you on a list!)
> >>>>
> >>>> So, it seems mostly a sort of "spam-walling the Internet",
> >>>> where it was like "we're going to reinvent the Internet",
> >>>> "no, you aren't", "all right then we'll ruin this one".
> >>>>
> >>>> As far as search goes, there's something to be said
> >>>> for a new sort of approach to search, given that
> >>>> Google, Bing, Duck, ..., _all make the same results_.  It's
> >>>> just so highly unlikely that they'd _all make the same
> >>>> results_, you figure they're just one.
> >>>>
> >>>> So, the idea, for somebody like me who's mostly interested
> >>>> in writing on the Internet, is that lots of that is of the sort
> >>>> of "works" vis-a-vis, the "feuilleton" or what you might
> >>>> call it, ephemeral junk, that I just learned about in
> >>>> Herman Hesse's "The Glass Bead Game".
> >>>>
> >>>> Then, there's an idea, that basically to surface high-quality
> >>>> works to a search, is that there's what's called metadata,
> >>>> for content like HTML, with regards to Dublin Core and
> >>>> RDF and so on, about a sort of making for fungible collections
> >>>> of works, what results searchable fragments of various
> >>>> larger bodies of works, according to their robots.txt and
> >>>> their summaries and with regards to crawling the content
> >>>> and so on, then to make federated common search corpi,
> >>>> these kinds of things.
> >>>>
> >>>>
> >>>>
> >>>
> >>> It's like "why are they building that new data center",
> >>> and it's like "well it's like Artificial Intelligence, inside
> >>> that data center is a million virts and each one has a
> >>> browser emulator and a phone app sandbox and a
> >>> little notecard that prompts its name, basically it's
> >>> a million-headed hydra called a sims-bot-farm,
> >>> that for pennies on the dollar is an instant audience."
> >>>
> >>> "Wow, great, do they get a cut?"  "Don't be talking about my cut."
> >>>
> >>> Usenet traffic had been up recently, ....
> >>>
> >>> I think they used to call it "astro-turfing".
> >>> "Artificial Intelligence?"  "No, 'Fake eyeballs'."
> >>
> >> I have NewsAgent111-MS.exe
> >>
> >> I seem to be missing version 2.0
> >>
> >> Do you have the 2.0 version?
> >>
> >> I'll trade you.
> >>
> >> I'll give you my python version with (GUI)!!!! (Tinter)
> >>
> >> let's trade!
> >>
> >> don't bogart
> >
> > I seem to be missing this version:
> >
> > https://web.archive.org/web/20051023050609/http://newsagent.p5.org.uk/
> >
> > Do you have it? you must have!
> >
> >
> >
> >
> >
> 
> Nope, I just wrote a little script to connect to NNTP
> with a Yes/No button on the subject, tapped through
> those, and a little script to send an HTTP request to
> the publicly-facing return-to-sender in-box, for each.
> 
> Here's all the sources you need:  IETF RFC editor.
> Look for "NNTP".  How to advise Google of this is
> that each domain on the Internet is supposed to
> have an "abuse@domain" email inbox, though there's
> probably also a web request interface, as with regards
> to publicly facing services, and expected to be
> good actors on the network.
> 
> Anyways if you read through "Meta: a usenet server
> just for sci.math", what I have in mind is a sort
> of author's and writer's oriented installation,
> basically making for vanity printouts and generating
> hypertext collections of contents and authors and
> subjects and these kinds of things, basically for
> on the order of "find all the postings of Archimedes
> Plutonium, and, the threads they are in, and,
> make a hypertext page of all that, a linear timeline,
> and also thread it out as a linear sequence".
> 
> I.e. people who actually post to Usenet are sometimes
> having written interesting things, and, thus having
> it so that it would be simplified to generate message-ID
> listings and their corresponding standard URL's in the
> standard IETF "news" URL protocol, and to point that
> at a given news server or like XLink, is for treating
> Usenet its archives like a living museum of all these
> different authors posts and their interactions together.
> 
> I.e., here it's "belles lettres" and "fair use",
> not just "belles" and "use".
> 
> It seemed nice of Google Groups to front this for a long time,
> now they're quitting.
> 
> I imagine Internet Relay Chat's still insane, though.
> 
> Anyways I stay away from any warez and am proud that
> since about Y2K at least I've never bootlegged anything,
> and never uploaded a bootleg.  Don't want to give old Shylock
> excuses, and besides, I wrote software for a living.

Anyways, I don't know who was talking about "any warez" or "bootlegs",
since I was refering to programs and scripts that reads:


"FREE, which means you can copy it and redistribute" 
"Similarly, the source is provided as reference and can be redistributed
freely as well. "

HipCrime's NewsAgent (v2.0) is FREE, which means you can copy it and
redistribute it at will, as long as you give credit to the original
author. Similarly, the source is provided as reference and can be
redistributed freely as well. 

https://web.archive.org/web/20051023050609/http://newsagent.p5.org.uk/


You seem to be too much 'in your head', on a high horse...



"FREE, which means you can copy it and redistribute" 
"Similarly, the source is provided as reference and can be redistributed
freely as well. "


So, show me that wicked script you wrote : "funnel all their
message-ID's"
by people you call spammers who 'funnel' their products and services
through Usenet newsgroups.


You are sooooo wicked.



and a nanofossils






-- 
The Starmaker -- To question the unquestionable, ask the unaskable,
to think the unthinkable, mention the unmentionable, say the unsayable, 
and challenge the unchallengeable.

[toc] | [prev] | [next] | [standalone]

#638627 — Re: Meta: Re: How I deal with the enormous amount of spam

The Starmaker wrote:
> 
> Ross Finlayson wrote:
> >
> > On 02/04/2024 12:53 PM, The Starmaker wrote:
> > > The Starmaker wrote:
> > >>
> > >> Ross Finlayson wrote:
> > >>>
> > >>> On 02/04/2024 09:55 AM, Ross Finlayson wrote:
> > >>>> On 02/03/2024 02:46 PM, The Starmaker wrote:
> > >>>>> The Starmaker wrote:
> > >>>>>>
> > >>>>>> Ross Finlayson wrote:
> > >>>>>>>
> > >>>>>>> On 01/30/2024 12:54 PM, Ross Finlayson wrote:
> > >>>>>>>> On Monday, January 29, 2024 at 5:02:05 PM UTC-8, palsing wrote:
> > >>>>>>>>> Tom Roberts wrote:
> > >>>>>>>>>
> > >>>>>>>>>> I use Thunderbird to read Usenet. Recently sci.physics.relativity
> > >>>>>>>>>> has
> > >>>>>>>>>> been getting hundreds of spam posts each day, completely
> > >>>>>>>>>> overwhelming
> > >>>>>>>>>> legitimate content. These spam posts share the property that they
> > >>>>>>>>>> are
> > >>>>>>>>>> written in a non-latin script.
> > >>>>>>>>>
> > >>>>>>>>>> Thunderbird implements message filters that can mark a message
> > >>>>>>>>>> Read. So
> > >>>>>>>>>> I created a filter to run on sci.physics.relativity that marks
> > >>>>>>>>>> messages
> > >>>>>>>>>> Read. Then when reading the newsgroups, I simply display only unread
> > >>>>>>>>>> messages. The key to making this work is to craft the filter so
> > >>>>>>>>>> it marks
> > >>>>>>>>>> messages in which the Subject matches any of a dozen characters
> > >>>>>>>>>> picked
> > >>>>>>>>>> from some spam messages.
> > >>>>>>>>>
> > >>>>>>>>>> This doesn't completely eliminate the spam, but it is now only a few
> > >>>>>>>>>> messages per day.
> > >>>>>>>>>
> > >>>>>>>>>> Tom Roberts
> > >>>>>>>>> I would like to do the same thing, so I installed Thunderbird...
> > >>>>>>>>> but setting it up to read newsgroups is beyond my paltry computer
> > >>>>>>>>> skills and is not at all intuitive. If anyone can point to an
> > >>>>>>>>> idiot-proof tutorial for doing this It would be much appreciated.
> > >>>>>>>>>
> > >>>>>>>>> \Paul Alsing
> > >>>>>>>>
> > >>>>>>>> Yeah, it's pretty bad, or, worse anybody's ever seen it.
> > >>>>>>>>
> > >>>>>>>> I as well sort of mow the lawn a bit or mark the spam.
> > >>>>>>>>
> > >>>>>>>> It seems alright if it'll be a sort of clean break:  on Feb 22
> > >>>>>>>> according to Google,
> > >>>>>>>> Google will break its compeerage to Usenet, and furthermore make
> > >>>>>>>> read-only
> > >>>>>>>> the archives, what it has, what until then, will be as it was.
> > >>>>>>>>
> > >>>>>>>> Over on sci.math I've had the idea for a while of making some brief
> > >>>>>>>> and
> > >>>>>>>> special purpose Usenet compeers, for only some few groups, or, you
> > >>>>>>>> know, the _belles lettres_ of the text hierarchy.
> > >>>>>>>>
> > >>>>>>>> "Meta:  a usenet server just for sci.math"
> > >>>>>>>>      -- https://groups.google.com/g/sci.math/c/zggff_pVEks
> > >>>>>>>>
> > >>>>>>>> So, there you can read the outlook of this kind of thing, then
> > >>>>>>>> while sort
> > >>>>>>>> of simple as the protocol is simple and its implementations
> > >>>>>>>> widespread,
> > >>>>>>>> how to deal with the "signal and noise" of "exposed messaging
> > >>>>>>>> destinations
> > >>>>>>>> on the Internet", well on that thread I'm theorizing a sort of,
> > >>>>>>>> "NOOBNB protocol",
> > >>>>>>>> figuring to make an otherwise just standard Usenet compeer, and
> > >>>>>>>> also for
> > >>>>>>>> email or messaging destinations, sort of designed with the
> > >>>>>>>> expectation that
> > >>>>>>>> there will be spam, and spam and ham are hand in hand, to exclude
> > >>>>>>>> it in simple terms.
> > >>>>>>>>
> > >>>>>>>> NOOBNB:  New Old Off Bot Non Bad, Curated/Purgatory/Raw triple-feed
> > >>>>>>>>
> > >>>>>>>> (That and a firmer sort of "Load Shed" or "Load Hold" at the
> > >>>>>>>> transport layer.)
> > >>>>>>>>
> > >>>>>>>> Also it would be real great if at least there was surfaced to the
> > >>>>>>>> Internet a
> > >>>>>>>> read-only view of any message by its message ID, a "URL", or as for
> > >>>>>>>> a "URI",
> > >>>>>>>> a "URN", a reliable perma-link in the IETF "news" protocol, namespace.
> > >>>>>>>>
> > >>>>>>>> https://groups.google.com/g/sci.math/c/zggff_pVEks
> > >>>>>>>>
> > >>>>>>>> I wonder that there's a reliable sort of long-term project that
> > >>>>>>>> surfaces
> > >>>>>>>> "news" protocol message-IDs, ....  It's a stable, standards-based
> > >>>>>>>> protocol.
> > >>>>>>>>
> > >>>>>>>>
> > >>>>>>>> Thunderbird, "SLRN", ....  Thanks for caring.  We care.
> > >>>>>>>>
> > >>>>>>>>
> > >>>>>>>> https://groups.google.com/g/sci.physics.relativity/c/ToBo6XOymUw
> > >>>>>>>>
> > >>>>>>>
> > >>>>>>> One fellow reached me via e-mail and he said, hey, the Googler spam is
> > >>>>>>> outrageous, can we do anything about it?  Would you write a script to
> > >>>>>>> funnel all their message-ID's into the abuse reporting?  And I was
> > >>>>>>> like,
> > >>>>>>> you know, about 2008 I did just that, there was a big spam flood,
> > >>>>>>> and I wrote a little script to find them and extract their
> > >>>>>>> posting-account,
> > >>>>>>> and the message-ID, and a little script to post to the posting-host,
> > >>>>>>> each one of the wicked spams.
> > >>>>>>>
> > >>>>>>> At the time that seemed to help, they sort of dried up, here there's
> > >>>>>>> that basically they're not following the charter, but, it's the
> > >>>>>>> posting-account
> > >>>>>>> in the message headers that indicate the origin of the post, not the
> > >>>>>>> email address.  So, I wonder, given that I can extract the
> > >>>>>>> posting-accounts
> > >>>>>>> of all the spams, how to match the posting-account to then determine
> > >>>>>>> whether it's a sockpuppet-farm or what, and basically about sending
> > >>>>>>> them up.
> > >>>>>>
> > >>>>>> Let me see your little script. Post it here.
> > >>>>>
> > >>>>> Here is a list I currently have:
> > >>>>>
> > >>>>> salz.txt
> > >>>>> usenet.death.penalty.gz
> > >>>>> purify.txt
> > >>>>> NewsAgent110-MS.exe
> > >>>>> HipCrime's NewsAgent (v1_11).htm
> > >>>>> NewsAgent111-BE.zip
> > >>>>> SuperCede.exe
> > >>>>> NewsAgent023.exe
> > >>>>> NewsAgent025.exe
> > >>>>> ActiveAgent.java
> > >>>>> HipCrime's NewsAgent (v1_02)_files
> > >>>>> NewsCancel.java (source code)
> > >>>>>
> > >>>>> (plus updated python versions)
> > >>>>>
> > >>>>>
> > >>>>>
> > >>>>> (Maybe your script is inthere somewhere?)
> > >>>>>
> > >>>>>
> > >>>>>
> > >>>>> Show me what you got. walk the walk.
> > >>>>>
> > >>>>
> > >>>>
> > >>>> I try to avoid sketchy things like hiring a criminal botnet,
> > >>>> there's the impression that that's looking at 1000's of counts
> > >>>> of computer intrusion.
> > >>>>
> > >>>> With those being something about $50K and 10-25 apiece,
> > >>>> there's a pretty significant deterrence to such activities.
> > >>>>
> > >>>> I've never much cared for "OAuth", giving away the
> > >>>> keys-to-the-kingdom and all, here it looks like either
> > >>>> a) a bunch of duped browsers clicked away their identities,
> > >>>> or b) it's really that Google and Facebook are more than
> > >>>> half full of fake identities for the sole purpose of being fake.
> > >>>>
> > >>>> (How's your new deal going?
> > >>>>     Great, we got a million users.
> > >>>> Why are my conversions around zero?
> > >>>>     Your ad must not speak to them.
> > >>>> Would it help if I spiced it up?
> > >>>>     Don't backtalk me, I'll put you on a list!)
> > >>>>
> > >>>> So, it seems mostly a sort of "spam-walling the Internet",
> > >>>> where it was like "we're going to reinvent the Internet",
> > >>>> "no, you aren't", "all right then we'll ruin this one".
> > >>>>
> > >>>> As far as search goes, there's something to be said
> > >>>> for a new sort of approach to search, given that
> > >>>> Google, Bing, Duck, ..., _all make the same results_.  It's
> > >>>> just so highly unlikely that they'd _all make the same
> > >>>> results_, you figure they're just one.
> > >>>>
> > >>>> So, the idea, for somebody like me who's mostly interested
> > >>>> in writing on the Internet, is that lots of that is of the sort
> > >>>> of "works" vis-a-vis, the "feuilleton" or what you might
> > >>>> call it, ephemeral junk, that I just learned about in
> > >>>> Herman Hesse's "The Glass Bead Game".
> > >>>>
> > >>>> Then, there's an idea, that basically to surface high-quality
> > >>>> works to a search, is that there's what's called metadata,
> > >>>> for content like HTML, with regards to Dublin Core and
> > >>>> RDF and so on, about a sort of making for fungible collections
> > >>>> of works, what results searchable fragments of various
> > >>>> larger bodies of works, according to their robots.txt and
> > >>>> their summaries and with regards to crawling the content
> > >>>> and so on, then to make federated common search corpi,
> > >>>> these kinds of things.
> > >>>>
> > >>>>
> > >>>>
> > >>>
> > >>> It's like "why are they building that new data center",
> > >>> and it's like "well it's like Artificial Intelligence, inside
> > >>> that data center is a million virts and each one has a
> > >>> browser emulator and a phone app sandbox and a
> > >>> little notecard that prompts its name, basically it's
> > >>> a million-headed hydra called a sims-bot-farm,
> > >>> that for pennies on the dollar is an instant audience."
> > >>>
> > >>> "Wow, great, do they get a cut?"  "Don't be talking about my cut."
> > >>>
> > >>> Usenet traffic had been up recently, ....
> > >>>
> > >>> I think they used to call it "astro-turfing".
> > >>> "Artificial Intelligence?"  "No, 'Fake eyeballs'."
> > >>
> > >> I have NewsAgent111-MS.exe
> > >>
> > >> I seem to be missing version 2.0
> > >>
> > >> Do you have the 2.0 version?
> > >>
> > >> I'll trade you.
> > >>
> > >> I'll give you my python version with (GUI)!!!! (Tinter)
> > >>
> > >> let's trade!
> > >>
> > >> don't bogart
> > >
> > > I seem to be missing this version:
> > >
> > > https://web.archive.org/web/20051023050609/http://newsagent.p5.org.uk/
> > >
> > > Do you have it? you must have!
> > >
> > >
> > >
> > >
> > >
> >
> > Nope, I just wrote a little script to connect to NNTP
> > with a Yes/No button on the subject, tapped through
> > those, and a little script to send an HTTP request to
> > the publicly-facing return-to-sender in-box, for each.
> >
> > Here's all the sources you need:  IETF RFC editor.
> > Look for "NNTP".  How to advise Google of this is
> > that each domain on the Internet is supposed to
> > have an "abuse@domain" email inbox, though there's
> > probably also a web request interface, as with regards
> > to publicly facing services, and expected to be
> > good actors on the network.
> >
> > Anyways if you read through "Meta: a usenet server
> > just for sci.math", what I have in mind is a sort
> > of author's and writer's oriented installation,
> > basically making for vanity printouts and generating
> > hypertext collections of contents and authors and
> > subjects and these kinds of things, basically for
> > on the order of "find all the postings of Archimedes
> > Plutonium, and, the threads they are in, and,
> > make a hypertext page of all that, a linear timeline,
> > and also thread it out as a linear sequence".
> >
> > I.e. people who actually post to Usenet are sometimes
> > having written interesting things, and, thus having
> > it so that it would be simplified to generate message-ID
> > listings and their corresponding standard URL's in the
> > standard IETF "news" URL protocol, and to point that
> > at a given news server or like XLink, is for treating
> > Usenet its archives like a living museum of all these
> > different authors posts and their interactions together.
> >
> > I.e., here it's "belles lettres" and "fair use",
> > not just "belles" and "use".
> >
> > It seemed nice of Google Groups to front this for a long time,
> > now they're quitting.
> >
> > I imagine Internet Relay Chat's still insane, though.
> >
> > Anyways I stay away from any warez and am proud that
> > since about Y2K at least I've never bootlegged anything,
> > and never uploaded a bootleg.  Don't want to give old Shylock
> > excuses, and besides, I wrote software for a living.
> 
> Anyways, I don't know who was talking about "any warez" or "bootlegs",
> since I was refering to programs and scripts that reads:
> 
> "FREE, which means you can copy it and redistribute"
> "Similarly, the source is provided as reference and can be redistributed
> freely as well. "
> 
> HipCrime's NewsAgent (v2.0) is FREE, which means you can copy it and
> redistribute it at will, as long as you give credit to the original
> author. Similarly, the source is provided as reference and can be
> redistributed freely as well.
> 
> https://web.archive.org/web/20051023050609/http://newsagent.p5.org.uk/
> 
> You seem to be too much 'in your head', on a high horse...
> 
> "FREE, which means you can copy it and redistribute"
> "Similarly, the source is provided as reference and can be redistributed
> freely as well. "
> 
> So, show me that wicked script you wrote : "funnel all their
> message-ID's"
> by people you call spammers who 'funnel' their products and services
> through Usenet newsgroups.
> 
> You are sooooo wicked.
> 
> and a nanofossils

Anyways, there is only one person that know what 'nanofossils' means,
and that is  Ross Finlayson.


I just realized that  Ross Finlayson doesn't know of NEWSAGENT.


Anyways, ...

"Anyways"???? Who talks like that?


Anyways..


the problem of the 'flooding' is not the spammers, it's the 'scientific
community'. They caused the problem.
They removed the feature that NewsAgent used to get rid of ALL flooding
and spammers. But, but, the
members of the scientific community could not trust their own members to
use it against them. 

If one member of the 'scientific community' disagreed with another
member of the 'scientific community'...they were removed!


Too much power.


I called it...God Mode.











-- 
The Starmaker -- To question the unquestionable, ask the unaskable,
to think the unthinkable, mention the unmentionable, say the unsayable, 
and challenge the unchallengeable.

[toc] | [prev] | [next] | [standalone]

#638637 — Re: Meta: Re: How I deal with the enormous amount of spam

The Starmaker wrote:
> 
> The Starmaker wrote:
> >
> > Ross Finlayson wrote:
> > >
> > > On 02/04/2024 12:53 PM, The Starmaker wrote:
> > > > The Starmaker wrote:
> > > >>
> > > >> Ross Finlayson wrote:
> > > >>>
> > > >>> On 02/04/2024 09:55 AM, Ross Finlayson wrote:
> > > >>>> On 02/03/2024 02:46 PM, The Starmaker wrote:
> > > >>>>> The Starmaker wrote:
> > > >>>>>>
> > > >>>>>> Ross Finlayson wrote:
> > > >>>>>>>
> > > >>>>>>> On 01/30/2024 12:54 PM, Ross Finlayson wrote:
> > > >>>>>>>> On Monday, January 29, 2024 at 5:02:05 PM UTC-8, palsing wrote:
> > > >>>>>>>>> Tom Roberts wrote:
> > > >>>>>>>>>
> > > >>>>>>>>>> I use Thunderbird to read Usenet. Recently sci.physics.relativity
> > > >>>>>>>>>> has
> > > >>>>>>>>>> been getting hundreds of spam posts each day, completely
> > > >>>>>>>>>> overwhelming
> > > >>>>>>>>>> legitimate content. These spam posts share the property that they
> > > >>>>>>>>>> are
> > > >>>>>>>>>> written in a non-latin script.
> > > >>>>>>>>>
> > > >>>>>>>>>> Thunderbird implements message filters that can mark a message
> > > >>>>>>>>>> Read. So
> > > >>>>>>>>>> I created a filter to run on sci.physics.relativity that marks
> > > >>>>>>>>>> messages
> > > >>>>>>>>>> Read. Then when reading the newsgroups, I simply display only unread
> > > >>>>>>>>>> messages. The key to making this work is to craft the filter so
> > > >>>>>>>>>> it marks
> > > >>>>>>>>>> messages in which the Subject matches any of a dozen characters
> > > >>>>>>>>>> picked
> > > >>>>>>>>>> from some spam messages.
> > > >>>>>>>>>
> > > >>>>>>>>>> This doesn't completely eliminate the spam, but it is now only a few
> > > >>>>>>>>>> messages per day.
> > > >>>>>>>>>
> > > >>>>>>>>>> Tom Roberts
> > > >>>>>>>>> I would like to do the same thing, so I installed Thunderbird...
> > > >>>>>>>>> but setting it up to read newsgroups is beyond my paltry computer
> > > >>>>>>>>> skills and is not at all intuitive. If anyone can point to an
> > > >>>>>>>>> idiot-proof tutorial for doing this It would be much appreciated.
> > > >>>>>>>>>
> > > >>>>>>>>> \Paul Alsing
> > > >>>>>>>>
> > > >>>>>>>> Yeah, it's pretty bad, or, worse anybody's ever seen it.
> > > >>>>>>>>
> > > >>>>>>>> I as well sort of mow the lawn a bit or mark the spam.
> > > >>>>>>>>
> > > >>>>>>>> It seems alright if it'll be a sort of clean break:  on Feb 22
> > > >>>>>>>> according to Google,
> > > >>>>>>>> Google will break its compeerage to Usenet, and furthermore make
> > > >>>>>>>> read-only
> > > >>>>>>>> the archives, what it has, what until then, will be as it was.
> > > >>>>>>>>
> > > >>>>>>>> Over on sci.math I've had the idea for a while of making some brief
> > > >>>>>>>> and
> > > >>>>>>>> special purpose Usenet compeers, for only some few groups, or, you
> > > >>>>>>>> know, the _belles lettres_ of the text hierarchy.
> > > >>>>>>>>
> > > >>>>>>>> "Meta:  a usenet server just for sci.math"
> > > >>>>>>>>      -- https://groups.google.com/g/sci.math/c/zggff_pVEks
> > > >>>>>>>>
> > > >>>>>>>> So, there you can read the outlook of this kind of thing, then
> > > >>>>>>>> while sort
> > > >>>>>>>> of simple as the protocol is simple and its implementations
> > > >>>>>>>> widespread,
> > > >>>>>>>> how to deal with the "signal and noise" of "exposed messaging
> > > >>>>>>>> destinations
> > > >>>>>>>> on the Internet", well on that thread I'm theorizing a sort of,
> > > >>>>>>>> "NOOBNB protocol",
> > > >>>>>>>> figuring to make an otherwise just standard Usenet compeer, and
> > > >>>>>>>> also for
> > > >>>>>>>> email or messaging destinations, sort of designed with the
> > > >>>>>>>> expectation that
> > > >>>>>>>> there will be spam, and spam and ham are hand in hand, to exclude
> > > >>>>>>>> it in simple terms.
> > > >>>>>>>>
> > > >>>>>>>> NOOBNB:  New Old Off Bot Non Bad, Curated/Purgatory/Raw triple-feed
> > > >>>>>>>>
> > > >>>>>>>> (That and a firmer sort of "Load Shed" or "Load Hold" at the
> > > >>>>>>>> transport layer.)
> > > >>>>>>>>
> > > >>>>>>>> Also it would be real great if at least there was surfaced to the
> > > >>>>>>>> Internet a
> > > >>>>>>>> read-only view of any message by its message ID, a "URL", or as for
> > > >>>>>>>> a "URI",
> > > >>>>>>>> a "URN", a reliable perma-link in the IETF "news" protocol, namespace.
> > > >>>>>>>>
> > > >>>>>>>> https://groups.google.com/g/sci.math/c/zggff_pVEks
> > > >>>>>>>>
> > > >>>>>>>> I wonder that there's a reliable sort of long-term project that
> > > >>>>>>>> surfaces
> > > >>>>>>>> "news" protocol message-IDs, ....  It's a stable, standards-based
> > > >>>>>>>> protocol.
> > > >>>>>>>>
> > > >>>>>>>>
> > > >>>>>>>> Thunderbird, "SLRN", ....  Thanks for caring.  We care.
> > > >>>>>>>>
> > > >>>>>>>>
> > > >>>>>>>> https://groups.google.com/g/sci.physics.relativity/c/ToBo6XOymUw
> > > >>>>>>>>
> > > >>>>>>>
> > > >>>>>>> One fellow reached me via e-mail and he said, hey, the Googler spam is
> > > >>>>>>> outrageous, can we do anything about it?  Would you write a script to
> > > >>>>>>> funnel all their message-ID's into the abuse reporting?  And I was
> > > >>>>>>> like,
> > > >>>>>>> you know, about 2008 I did just that, there was a big spam flood,
> > > >>>>>>> and I wrote a little script to find them and extract their
> > > >>>>>>> posting-account,
> > > >>>>>>> and the message-ID, and a little script to post to the posting-host,
> > > >>>>>>> each one of the wicked spams.
> > > >>>>>>>
> > > >>>>>>> At the time that seemed to help, they sort of dried up, here there's
> > > >>>>>>> that basically they're not following the charter, but, it's the
> > > >>>>>>> posting-account
> > > >>>>>>> in the message headers that indicate the origin of the post, not the
> > > >>>>>>> email address.  So, I wonder, given that I can extract the
> > > >>>>>>> posting-accounts
> > > >>>>>>> of all the spams, how to match the posting-account to then determine
> > > >>>>>>> whether it's a sockpuppet-farm or what, and basically about sending
> > > >>>>>>> them up.
> > > >>>>>>
> > > >>>>>> Let me see your little script. Post it here.
> > > >>>>>
> > > >>>>> Here is a list I currently have:
> > > >>>>>
> > > >>>>> salz.txt
> > > >>>>> usenet.death.penalty.gz
> > > >>>>> purify.txt
> > > >>>>> NewsAgent110-MS.exe
> > > >>>>> HipCrime's NewsAgent (v1_11).htm
> > > >>>>> NewsAgent111-BE.zip
> > > >>>>> SuperCede.exe
> > > >>>>> NewsAgent023.exe
> > > >>>>> NewsAgent025.exe
> > > >>>>> ActiveAgent.java
> > > >>>>> HipCrime's NewsAgent (v1_02)_files
> > > >>>>> NewsCancel.java (source code)
> > > >>>>>
> > > >>>>> (plus updated python versions)
> > > >>>>>
> > > >>>>>
> > > >>>>>
> > > >>>>> (Maybe your script is inthere somewhere?)
> > > >>>>>
> > > >>>>>
> > > >>>>>
> > > >>>>> Show me what you got. walk the walk.
> > > >>>>>
> > > >>>>
> > > >>>>
> > > >>>> I try to avoid sketchy things like hiring a criminal botnet,
> > > >>>> there's the impression that that's looking at 1000's of counts
> > > >>>> of computer intrusion.
> > > >>>>
> > > >>>> With those being something about $50K and 10-25 apiece,
> > > >>>> there's a pretty significant deterrence to such activities.
> > > >>>>
> > > >>>> I've never much cared for "OAuth", giving away the
> > > >>>> keys-to-the-kingdom and all, here it looks like either
> > > >>>> a) a bunch of duped browsers clicked away their identities,
> > > >>>> or b) it's really that Google and Facebook are more than
> > > >>>> half full of fake identities for the sole purpose of being fake.
> > > >>>>
> > > >>>> (How's your new deal going?
> > > >>>>     Great, we got a million users.
> > > >>>> Why are my conversions around zero?
> > > >>>>     Your ad must not speak to them.
> > > >>>> Would it help if I spiced it up?
> > > >>>>     Don't backtalk me, I'll put you on a list!)
> > > >>>>
> > > >>>> So, it seems mostly a sort of "spam-walling the Internet",
> > > >>>> where it was like "we're going to reinvent the Internet",
> > > >>>> "no, you aren't", "all right then we'll ruin this one".
> > > >>>>
> > > >>>> As far as search goes, there's something to be said
> > > >>>> for a new sort of approach to search, given that
> > > >>>> Google, Bing, Duck, ..., _all make the same results_.  It's
> > > >>>> just so highly unlikely that they'd _all make the same
> > > >>>> results_, you figure they're just one.
> > > >>>>
> > > >>>> So, the idea, for somebody like me who's mostly interested
> > > >>>> in writing on the Internet, is that lots of that is of the sort
> > > >>>> of "works" vis-a-vis, the "feuilleton" or what you might
> > > >>>> call it, ephemeral junk, that I just learned about in
> > > >>>> Herman Hesse's "The Glass Bead Game".
> > > >>>>
> > > >>>> Then, there's an idea, that basically to surface high-quality
> > > >>>> works to a search, is that there's what's called metadata,
> > > >>>> for content like HTML, with regards to Dublin Core and
> > > >>>> RDF and so on, about a sort of making for fungible collections
> > > >>>> of works, what results searchable fragments of various
> > > >>>> larger bodies of works, according to their robots.txt and
> > > >>>> their summaries and with regards to crawling the content
> > > >>>> and so on, then to make federated common search corpi,
> > > >>>> these kinds of things.
> > > >>>>
> > > >>>>
> > > >>>>
> > > >>>
> > > >>> It's like "why are they building that new data center",
> > > >>> and it's like "well it's like Artificial Intelligence, inside
> > > >>> that data center is a million virts and each one has a
> > > >>> browser emulator and a phone app sandbox and a
> > > >>> little notecard that prompts its name, basically it's
> > > >>> a million-headed hydra called a sims-bot-farm,
> > > >>> that for pennies on the dollar is an instant audience."
> > > >>>
> > > >>> "Wow, great, do they get a cut?"  "Don't be talking about my cut."
> > > >>>
> > > >>> Usenet traffic had been up recently, ....
> > > >>>
> > > >>> I think they used to call it "astro-turfing".
> > > >>> "Artificial Intelligence?"  "No, 'Fake eyeballs'."
> > > >>
> > > >> I have NewsAgent111-MS.exe
> > > >>
> > > >> I seem to be missing version 2.0
> > > >>
> > > >> Do you have the 2.0 version?
> > > >>
> > > >> I'll trade you.
> > > >>
> > > >> I'll give you my python version with (GUI)!!!! (Tinter)
> > > >>
> > > >> let's trade!
> > > >>
> > > >> don't bogart
> > > >
> > > > I seem to be missing this version:
> > > >
> > > > https://web.archive.org/web/20051023050609/http://newsagent.p5.org.uk/
> > > >
> > > > Do you have it? you must have!
> > > >
> > > >
> > > >
> > > >
> > > >
> > >
> > > Nope, I just wrote a little script to connect to NNTP
> > > with a Yes/No button on the subject, tapped through
> > > those, and a little script to send an HTTP request to
> > > the publicly-facing return-to-sender in-box, for each.
> > >
> > > Here's all the sources you need:  IETF RFC editor.
> > > Look for "NNTP".  How to advise Google of this is
> > > that each domain on the Internet is supposed to
> > > have an "abuse@domain" email inbox, though there's
> > > probably also a web request interface, as with regards
> > > to publicly facing services, and expected to be
> > > good actors on the network.
> > >
> > > Anyways if you read through "Meta: a usenet server
> > > just for sci.math", what I have in mind is a sort
> > > of author's and writer's oriented installation,
> > > basically making for vanity printouts and generating
> > > hypertext collections of contents and authors and
> > > subjects and these kinds of things, basically for
> > > on the order of "find all the postings of Archimedes
> > > Plutonium, and, the threads they are in, and,
> > > make a hypertext page of all that, a linear timeline,
> > > and also thread it out as a linear sequence".
> > >
> > > I.e. people who actually post to Usenet are sometimes
> > > having written interesting things, and, thus having
> > > it so that it would be simplified to generate message-ID
> > > listings and their corresponding standard URL's in the
> > > standard IETF "news" URL protocol, and to point that
> > > at a given news server or like XLink, is for treating
> > > Usenet its archives like a living museum of all these
> > > different authors posts and their interactions together.
> > >
> > > I.e., here it's "belles lettres" and "fair use",
> > > not just "belles" and "use".
> > >
> > > It seemed nice of Google Groups to front this for a long time,
> > > now they're quitting.
> > >
> > > I imagine Internet Relay Chat's still insane, though.
> > >
> > > Anyways I stay away from any warez and am proud that
> > > since about Y2K at least I've never bootlegged anything,
> > > and never uploaded a bootleg.  Don't want to give old Shylock
> > > excuses, and besides, I wrote software for a living.
> >
> > Anyways, I don't know who was talking about "any warez" or "bootlegs",
> > since I was refering to programs and scripts that reads:
> >
> > "FREE, which means you can copy it and redistribute"
> > "Similarly, the source is provided as reference and can be redistributed
> > freely as well. "
> >
> > HipCrime's NewsAgent (v2.0) is FREE, which means you can copy it and
> > redistribute it at will, as long as you give credit to the original
> > author. Similarly, the source is provided as reference and can be
> > redistributed freely as well.
> >
> > https://web.archive.org/web/20051023050609/http://newsagent.p5.org.uk/
> >
> > You seem to be too much 'in your head', on a high horse...
> >
> > "FREE, which means you can copy it and redistribute"
> > "Similarly, the source is provided as reference and can be redistributed
> > freely as well. "
> >
> > So, show me that wicked script you wrote : "funnel all their
> > message-ID's"
> > by people you call spammers who 'funnel' their products and services
> > through Usenet newsgroups.
> >
> > You are sooooo wicked.
> >
> > and a nanofossils
> 
> Anyways, there is only one person that know what 'nanofossils' means,
> and that is  Ross Finlayson.
> 
> I just realized that  Ross Finlayson doesn't know of NEWSAGENT.
> 
> Anyways, ...
> 
> "Anyways"???? Who talks like that?
> 
> Anyways..
> 
> the problem of the 'flooding' is not the spammers, it's the 'scientific
> community'. They caused the problem.
> They removed the feature that NewsAgent used to get rid of ALL flooding
> and spammers. But, but, the
> members of the scientific community could not trust their own members to
> use it against them.
> 
> If one member of the 'scientific community' disagreed with another
> member of the 'scientific community'...they were removed!
> 
> Too much power.
> 
> I called it...God Mode.

Using NewsAgent in God Mode was great! Ecept...if you didn't know how to use it 
properly you can make a mistake and remove *EVERYONE'S* posts by accident.

Everyone just completely disapeared!


Oops. i made a booboo.


Like that Twilight Zone episode where everyone disapears by a click of a watch.



Where is everybody? MAJOR KILLFILE!

So, which is worse?



-- 
The Starmaker -- To question the unquestionable, ask the unaskable,
to think the unthinkable, mention the unmentionable, say the unsayable, 
and challenge the unchallengeable.

[toc] | [prev] | [next] | [standalone]

#638640 — Re: Meta: Re: How I deal with the enormous amount of spam

The Starmaker wrote:
> 
> The Starmaker wrote:
> >
> > The Starmaker wrote:
> > >
> > > Ross Finlayson wrote:
> > > >
> > > > On 02/04/2024 12:53 PM, The Starmaker wrote:
> > > > > The Starmaker wrote:
> > > > >>
> > > > >> Ross Finlayson wrote:
> > > > >>>
> > > > >>> On 02/04/2024 09:55 AM, Ross Finlayson wrote:
> > > > >>>> On 02/03/2024 02:46 PM, The Starmaker wrote:
> > > > >>>>> The Starmaker wrote:
> > > > >>>>>>
> > > > >>>>>> Ross Finlayson wrote:
> > > > >>>>>>>
> > > > >>>>>>> On 01/30/2024 12:54 PM, Ross Finlayson wrote:
> > > > >>>>>>>> On Monday, January 29, 2024 at 5:02:05 PM UTC-8, palsing wrote:
> > > > >>>>>>>>> Tom Roberts wrote:
> > > > >>>>>>>>>
> > > > >>>>>>>>>> I use Thunderbird to read Usenet. Recently sci.physics.relativity
> > > > >>>>>>>>>> has
> > > > >>>>>>>>>> been getting hundreds of spam posts each day, completely
> > > > >>>>>>>>>> overwhelming
> > > > >>>>>>>>>> legitimate content. These spam posts share the property that they
> > > > >>>>>>>>>> are
> > > > >>>>>>>>>> written in a non-latin script.
> > > > >>>>>>>>>
> > > > >>>>>>>>>> Thunderbird implements message filters that can mark a message
> > > > >>>>>>>>>> Read. So
> > > > >>>>>>>>>> I created a filter to run on sci.physics.relativity that marks
> > > > >>>>>>>>>> messages
> > > > >>>>>>>>>> Read. Then when reading the newsgroups, I simply display only unread
> > > > >>>>>>>>>> messages. The key to making this work is to craft the filter so
> > > > >>>>>>>>>> it marks
> > > > >>>>>>>>>> messages in which the Subject matches any of a dozen characters
> > > > >>>>>>>>>> picked
> > > > >>>>>>>>>> from some spam messages.
> > > > >>>>>>>>>
> > > > >>>>>>>>>> This doesn't completely eliminate the spam, but it is now only a few
> > > > >>>>>>>>>> messages per day.
> > > > >>>>>>>>>
> > > > >>>>>>>>>> Tom Roberts
> > > > >>>>>>>>> I would like to do the same thing, so I installed Thunderbird...
> > > > >>>>>>>>> but setting it up to read newsgroups is beyond my paltry computer
> > > > >>>>>>>>> skills and is not at all intuitive. If anyone can point to an
> > > > >>>>>>>>> idiot-proof tutorial for doing this It would be much appreciated.
> > > > >>>>>>>>>
> > > > >>>>>>>>> \Paul Alsing
> > > > >>>>>>>>
> > > > >>>>>>>> Yeah, it's pretty bad, or, worse anybody's ever seen it.
> > > > >>>>>>>>
> > > > >>>>>>>> I as well sort of mow the lawn a bit or mark the spam.
> > > > >>>>>>>>
> > > > >>>>>>>> It seems alright if it'll be a sort of clean break:  on Feb 22
> > > > >>>>>>>> according to Google,
> > > > >>>>>>>> Google will break its compeerage to Usenet, and furthermore make
> > > > >>>>>>>> read-only
> > > > >>>>>>>> the archives, what it has, what until then, will be as it was.
> > > > >>>>>>>>
> > > > >>>>>>>> Over on sci.math I've had the idea for a while of making some brief
> > > > >>>>>>>> and
> > > > >>>>>>>> special purpose Usenet compeers, for only some few groups, or, you
> > > > >>>>>>>> know, the _belles lettres_ of the text hierarchy.
> > > > >>>>>>>>
> > > > >>>>>>>> "Meta:  a usenet server just for sci.math"
> > > > >>>>>>>>      -- https://groups.google.com/g/sci.math/c/zggff_pVEks
> > > > >>>>>>>>
> > > > >>>>>>>> So, there you can read the outlook of this kind of thing, then
> > > > >>>>>>>> while sort
> > > > >>>>>>>> of simple as the protocol is simple and its implementations
> > > > >>>>>>>> widespread,
> > > > >>>>>>>> how to deal with the "signal and noise" of "exposed messaging
> > > > >>>>>>>> destinations
> > > > >>>>>>>> on the Internet", well on that thread I'm theorizing a sort of,
> > > > >>>>>>>> "NOOBNB protocol",
> > > > >>>>>>>> figuring to make an otherwise just standard Usenet compeer, and
> > > > >>>>>>>> also for
> > > > >>>>>>>> email or messaging destinations, sort of designed with the
> > > > >>>>>>>> expectation that
> > > > >>>>>>>> there will be spam, and spam and ham are hand in hand, to exclude
> > > > >>>>>>>> it in simple terms.
> > > > >>>>>>>>
> > > > >>>>>>>> NOOBNB:  New Old Off Bot Non Bad, Curated/Purgatory/Raw triple-feed
> > > > >>>>>>>>
> > > > >>>>>>>> (That and a firmer sort of "Load Shed" or "Load Hold" at the
> > > > >>>>>>>> transport layer.)
> > > > >>>>>>>>
> > > > >>>>>>>> Also it would be real great if at least there was surfaced to the
> > > > >>>>>>>> Internet a
> > > > >>>>>>>> read-only view of any message by its message ID, a "URL", or as for
> > > > >>>>>>>> a "URI",
> > > > >>>>>>>> a "URN", a reliable perma-link in the IETF "news" protocol, namespace.
> > > > >>>>>>>>
> > > > >>>>>>>> https://groups.google.com/g/sci.math/c/zggff_pVEks
> > > > >>>>>>>>
> > > > >>>>>>>> I wonder that there's a reliable sort of long-term project that
> > > > >>>>>>>> surfaces
> > > > >>>>>>>> "news" protocol message-IDs, ....  It's a stable, standards-based
> > > > >>>>>>>> protocol.
> > > > >>>>>>>>
> > > > >>>>>>>>
> > > > >>>>>>>> Thunderbird, "SLRN", ....  Thanks for caring.  We care.
> > > > >>>>>>>>
> > > > >>>>>>>>
> > > > >>>>>>>> https://groups.google.com/g/sci.physics.relativity/c/ToBo6XOymUw
> > > > >>>>>>>>
> > > > >>>>>>>
> > > > >>>>>>> One fellow reached me via e-mail and he said, hey, the Googler spam is
> > > > >>>>>>> outrageous, can we do anything about it?  Would you write a script to
> > > > >>>>>>> funnel all their message-ID's into the abuse reporting?  And I was
> > > > >>>>>>> like,
> > > > >>>>>>> you know, about 2008 I did just that, there was a big spam flood,
> > > > >>>>>>> and I wrote a little script to find them and extract their
> > > > >>>>>>> posting-account,
> > > > >>>>>>> and the message-ID, and a little script to post to the posting-host,
> > > > >>>>>>> each one of the wicked spams.
> > > > >>>>>>>
> > > > >>>>>>> At the time that seemed to help, they sort of dried up, here there's
> > > > >>>>>>> that basically they're not following the charter, but, it's the
> > > > >>>>>>> posting-account
> > > > >>>>>>> in the message headers that indicate the origin of the post, not the
> > > > >>>>>>> email address.  So, I wonder, given that I can extract the
> > > > >>>>>>> posting-accounts
> > > > >>>>>>> of all the spams, how to match the posting-account to then determine
> > > > >>>>>>> whether it's a sockpuppet-farm or what, and basically about sending
> > > > >>>>>>> them up.
> > > > >>>>>>
> > > > >>>>>> Let me see your little script. Post it here.
> > > > >>>>>
> > > > >>>>> Here is a list I currently have:
> > > > >>>>>
> > > > >>>>> salz.txt
> > > > >>>>> usenet.death.penalty.gz
> > > > >>>>> purify.txt
> > > > >>>>> NewsAgent110-MS.exe
> > > > >>>>> HipCrime's NewsAgent (v1_11).htm
> > > > >>>>> NewsAgent111-BE.zip
> > > > >>>>> SuperCede.exe
> > > > >>>>> NewsAgent023.exe
> > > > >>>>> NewsAgent025.exe
> > > > >>>>> ActiveAgent.java
> > > > >>>>> HipCrime's NewsAgent (v1_02)_files
> > > > >>>>> NewsCancel.java (source code)
> > > > >>>>>
> > > > >>>>> (plus updated python versions)
> > > > >>>>>
> > > > >>>>>
> > > > >>>>>
> > > > >>>>> (Maybe your script is inthere somewhere?)
> > > > >>>>>
> > > > >>>>>
> > > > >>>>>
> > > > >>>>> Show me what you got. walk the walk.
> > > > >>>>>
> > > > >>>>
> > > > >>>>
> > > > >>>> I try to avoid sketchy things like hiring a criminal botnet,
> > > > >>>> there's the impression that that's looking at 1000's of counts
> > > > >>>> of computer intrusion.
> > > > >>>>
> > > > >>>> With those being something about $50K and 10-25 apiece,
> > > > >>>> there's a pretty significant deterrence to such activities.
> > > > >>>>
> > > > >>>> I've never much cared for "OAuth", giving away the
> > > > >>>> keys-to-the-kingdom and all, here it looks like either
> > > > >>>> a) a bunch of duped browsers clicked away their identities,
> > > > >>>> or b) it's really that Google and Facebook are more than
> > > > >>>> half full of fake identities for the sole purpose of being fake.
> > > > >>>>
> > > > >>>> (How's your new deal going?
> > > > >>>>     Great, we got a million users.
> > > > >>>> Why are my conversions around zero?
> > > > >>>>     Your ad must not speak to them.
> > > > >>>> Would it help if I spiced it up?
> > > > >>>>     Don't backtalk me, I'll put you on a list!)
> > > > >>>>
> > > > >>>> So, it seems mostly a sort of "spam-walling the Internet",
> > > > >>>> where it was like "we're going to reinvent the Internet",
> > > > >>>> "no, you aren't", "all right then we'll ruin this one".
> > > > >>>>
> > > > >>>> As far as search goes, there's something to be said
> > > > >>>> for a new sort of approach to search, given that
> > > > >>>> Google, Bing, Duck, ..., _all make the same results_.  It's
> > > > >>>> just so highly unlikely that they'd _all make the same
> > > > >>>> results_, you figure they're just one.
> > > > >>>>
> > > > >>>> So, the idea, for somebody like me who's mostly interested
> > > > >>>> in writing on the Internet, is that lots of that is of the sort
> > > > >>>> of "works" vis-a-vis, the "feuilleton" or what you might
> > > > >>>> call it, ephemeral junk, that I just learned about in
> > > > >>>> Herman Hesse's "The Glass Bead Game".
> > > > >>>>
> > > > >>>> Then, there's an idea, that basically to surface high-quality
> > > > >>>> works to a search, is that there's what's called metadata,
> > > > >>>> for content like HTML, with regards to Dublin Core and
> > > > >>>> RDF and so on, about a sort of making for fungible collections
> > > > >>>> of works, what results searchable fragments of various
> > > > >>>> larger bodies of works, according to their robots.txt and
> > > > >>>> their summaries and with regards to crawling the content
> > > > >>>> and so on, then to make federated common search corpi,
> > > > >>>> these kinds of things.
> > > > >>>>
> > > > >>>>
> > > > >>>>
> > > > >>>
> > > > >>> It's like "why are they building that new data center",
> > > > >>> and it's like "well it's like Artificial Intelligence, inside
> > > > >>> that data center is a million virts and each one has a
> > > > >>> browser emulator and a phone app sandbox and a
> > > > >>> little notecard that prompts its name, basically it's
> > > > >>> a million-headed hydra called a sims-bot-farm,
> > > > >>> that for pennies on the dollar is an instant audience."
> > > > >>>
> > > > >>> "Wow, great, do they get a cut?"  "Don't be talking about my cut."
> > > > >>>
> > > > >>> Usenet traffic had been up recently, ....
> > > > >>>
> > > > >>> I think they used to call it "astro-turfing".
> > > > >>> "Artificial Intelligence?"  "No, 'Fake eyeballs'."
> > > > >>
> > > > >> I have NewsAgent111-MS.exe
> > > > >>
> > > > >> I seem to be missing version 2.0
> > > > >>
> > > > >> Do you have the 2.0 version?
> > > > >>
> > > > >> I'll trade you.
> > > > >>
> > > > >> I'll give you my python version with (GUI)!!!! (Tinter)
> > > > >>
> > > > >> let's trade!
> > > > >>
> > > > >> don't bogart
> > > > >
> > > > > I seem to be missing this version:
> > > > >
> > > > > https://web.archive.org/web/20051023050609/http://newsagent.p5.org.uk/
> > > > >
> > > > > Do you have it? you must have!
> > > > >
> > > > >
> > > > >
> > > > >
> > > > >
> > > >
> > > > Nope, I just wrote a little script to connect to NNTP
> > > > with a Yes/No button on the subject, tapped through
> > > > those, and a little script to send an HTTP request to
> > > > the publicly-facing return-to-sender in-box, for each.
> > > >
> > > > Here's all the sources you need:  IETF RFC editor.
> > > > Look for "NNTP".  How to advise Google of this is
> > > > that each domain on the Internet is supposed to
> > > > have an "abuse@domain" email inbox, though there's
> > > > probably also a web request interface, as with regards
> > > > to publicly facing services, and expected to be
> > > > good actors on the network.
> > > >
> > > > Anyways if you read through "Meta: a usenet server
> > > > just for sci.math", what I have in mind is a sort
> > > > of author's and writer's oriented installation,
> > > > basically making for vanity printouts and generating
> > > > hypertext collections of contents and authors and
> > > > subjects and these kinds of things, basically for
> > > > on the order of "find all the postings of Archimedes
> > > > Plutonium, and, the threads they are in, and,
> > > > make a hypertext page of all that, a linear timeline,
> > > > and also thread it out as a linear sequence".
> > > >
> > > > I.e. people who actually post to Usenet are sometimes
> > > > having written interesting things, and, thus having
> > > > it so that it would be simplified to generate message-ID
> > > > listings and their corresponding standard URL's in the
> > > > standard IETF "news" URL protocol, and to point that
> > > > at a given news server or like XLink, is for treating
> > > > Usenet its archives like a living museum of all these
> > > > different authors posts and their interactions together.
> > > >
> > > > I.e., here it's "belles lettres" and "fair use",
> > > > not just "belles" and "use".
> > > >
> > > > It seemed nice of Google Groups to front this for a long time,
> > > > now they're quitting.
> > > >
> > > > I imagine Internet Relay Chat's still insane, though.
> > > >
> > > > Anyways I stay away from any warez and am proud that
> > > > since about Y2K at least I've never bootlegged anything,
> > > > and never uploaded a bootleg.  Don't want to give old Shylock
> > > > excuses, and besides, I wrote software for a living.
> > >
> > > Anyways, I don't know who was talking about "any warez" or "bootlegs",
> > > since I was refering to programs and scripts that reads:
> > >
> > > "FREE, which means you can copy it and redistribute"
> > > "Similarly, the source is provided as reference and can be redistributed
> > > freely as well. "
> > >
> > > HipCrime's NewsAgent (v2.0) is FREE, which means you can copy it and
> > > redistribute it at will, as long as you give credit to the original
> > > author. Similarly, the source is provided as reference and can be
> > > redistributed freely as well.
> > >
> > > https://web.archive.org/web/20051023050609/http://newsagent.p5.org.uk/
> > >
> > > You seem to be too much 'in your head', on a high horse...
> > >
> > > "FREE, which means you can copy it and redistribute"
> > > "Similarly, the source is provided as reference and can be redistributed
> > > freely as well. "
> > >
> > > So, show me that wicked script you wrote : "funnel all their
> > > message-ID's"
> > > by people you call spammers who 'funnel' their products and services
> > > through Usenet newsgroups.
> > >
> > > You are sooooo wicked.
> > >
> > > and a nanofossils
> >
> > Anyways, there is only one person that know what 'nanofossils' means,
> > and that is  Ross Finlayson.
> >
> > I just realized that  Ross Finlayson doesn't know of NEWSAGENT.
> >
> > Anyways, ...
> >
> > "Anyways"???? Who talks like that?
> >
> > Anyways..
> >
> > the problem of the 'flooding' is not the spammers, it's the 'scientific
> > community'. They caused the problem.
> > They removed the feature that NewsAgent used to get rid of ALL flooding
> > and spammers. But, but, the
> > members of the scientific community could not trust their own members to
> > use it against them.
> >
> > If one member of the 'scientific community' disagreed with another
> > member of the 'scientific community'...they were removed!
> >
> > Too much power.
> >
> > I called it...God Mode.
> 
> Using NewsAgent in God Mode was great! Ecept...if you didn't know how to use it
> properly you can make a mistake and remove *EVERYONE'S* posts by accident.
> 
> Everyone just completely disapeared!
> 
> Oops. i made a booboo.
> 
> Like that Twilight Zone episode where everyone disapears by a click of a watch.
> 
> Where is everybody? MAJOR KILLFILE!
> 
> So, which is worse?

You know what GOD MODE Killfile is? That means you not only killed filed everyone, but you also sort of
turn everyones esles killfile on. Nobody sees nobody.


It's like the Atomic Bomb of Usenet!


(only yous guys make bombs like that)
(then yous get angry when everyone has the atomic bomb)

typical.





-- 
The Starmaker -- To question the unquestionable, ask the unaskable,
to think the unthinkable, mention the unmentionable, say the unsayable, 
and challenge the unchallengeable.

[toc] | [prev] | [next] | [standalone]

#640049 — Re: Meta: Re: How I deal with the enormous amount of spam

On 02/04/2024 06:28 PM, Ross Finlayson wrote:
> On 02/04/2024 12:53 PM, The Starmaker wrote:
>> The Starmaker wrote:
>>>
>>> Ross Finlayson wrote:
>>>>
>>>> On 02/04/2024 09:55 AM, Ross Finlayson wrote:
>>>>> On 02/03/2024 02:46 PM, The Starmaker wrote:
>>>>>> The Starmaker wrote:
>>>>>>>
>>>>>>> Ross Finlayson wrote:
>>>>>>>>
>>>>>>>> On 01/30/2024 12:54 PM, Ross Finlayson wrote:
>>>>>>>>> On Monday, January 29, 2024 at 5:02:05 PM UTC-8, palsing wrote:
>>>>>>>>>> Tom Roberts wrote:
>>>>>>>>>>
>>>>>>>>>>> I use Thunderbird to read Usenet. Recently
>>>>>>>>>>> sci.physics.relativity
>>>>>>>>>>> has
>>>>>>>>>>> been getting hundreds of spam posts each day, completely
>>>>>>>>>>> overwhelming
>>>>>>>>>>> legitimate content. These spam posts share the property that
>>>>>>>>>>> they
>>>>>>>>>>> are
>>>>>>>>>>> written in a non-latin script.
>>>>>>>>>>
>>>>>>>>>>> Thunderbird implements message filters that can mark a message
>>>>>>>>>>> Read. So
>>>>>>>>>>> I created a filter to run on sci.physics.relativity that marks
>>>>>>>>>>> messages
>>>>>>>>>>> Read. Then when reading the newsgroups, I simply display only
>>>>>>>>>>> unread
>>>>>>>>>>> messages. The key to making this work is to craft the filter so
>>>>>>>>>>> it marks
>>>>>>>>>>> messages in which the Subject matches any of a dozen characters
>>>>>>>>>>> picked
>>>>>>>>>>> from some spam messages.
>>>>>>>>>>
>>>>>>>>>>> This doesn't completely eliminate the spam, but it is now
>>>>>>>>>>> only a few
>>>>>>>>>>> messages per day.
>>>>>>>>>>
>>>>>>>>>>> Tom Roberts
>>>>>>>>>> I would like to do the same thing, so I installed Thunderbird...
>>>>>>>>>> but setting it up to read newsgroups is beyond my paltry computer
>>>>>>>>>> skills and is not at all intuitive. If anyone can point to an
>>>>>>>>>> idiot-proof tutorial for doing this It would be much appreciated.
>>>>>>>>>>
>>>>>>>>>> \Paul Alsing
>>>>>>>>>
>>>>>>>>> Yeah, it's pretty bad, or, worse anybody's ever seen it.
>>>>>>>>>
>>>>>>>>> I as well sort of mow the lawn a bit or mark the spam.
>>>>>>>>>
>>>>>>>>> It seems alright if it'll be a sort of clean break:  on Feb 22
>>>>>>>>> according to Google,
>>>>>>>>> Google will break its compeerage to Usenet, and furthermore make
>>>>>>>>> read-only
>>>>>>>>> the archives, what it has, what until then, will be as it was.
>>>>>>>>>
>>>>>>>>> Over on sci.math I've had the idea for a while of making some
>>>>>>>>> brief
>>>>>>>>> and
>>>>>>>>> special purpose Usenet compeers, for only some few groups, or, you
>>>>>>>>> know, the _belles lettres_ of the text hierarchy.
>>>>>>>>>
>>>>>>>>> "Meta:  a usenet server just for sci.math"
>>>>>>>>>      -- https://groups.google.com/g/sci.math/c/zggff_pVEks
>>>>>>>>>
>>>>>>>>> So, there you can read the outlook of this kind of thing, then
>>>>>>>>> while sort
>>>>>>>>> of simple as the protocol is simple and its implementations
>>>>>>>>> widespread,
>>>>>>>>> how to deal with the "signal and noise" of "exposed messaging
>>>>>>>>> destinations
>>>>>>>>> on the Internet", well on that thread I'm theorizing a sort of,
>>>>>>>>> "NOOBNB protocol",
>>>>>>>>> figuring to make an otherwise just standard Usenet compeer, and
>>>>>>>>> also for
>>>>>>>>> email or messaging destinations, sort of designed with the
>>>>>>>>> expectation that
>>>>>>>>> there will be spam, and spam and ham are hand in hand, to exclude
>>>>>>>>> it in simple terms.
>>>>>>>>>
>>>>>>>>> NOOBNB:  New Old Off Bot Non Bad, Curated/Purgatory/Raw
>>>>>>>>> triple-feed
>>>>>>>>>
>>>>>>>>> (That and a firmer sort of "Load Shed" or "Load Hold" at the
>>>>>>>>> transport layer.)
>>>>>>>>>
>>>>>>>>> Also it would be real great if at least there was surfaced to the
>>>>>>>>> Internet a
>>>>>>>>> read-only view of any message by its message ID, a "URL", or as
>>>>>>>>> for
>>>>>>>>> a "URI",
>>>>>>>>> a "URN", a reliable perma-link in the IETF "news" protocol,
>>>>>>>>> namespace.
>>>>>>>>>
>>>>>>>>> https://groups.google.com/g/sci.math/c/zggff_pVEks
>>>>>>>>>
>>>>>>>>> I wonder that there's a reliable sort of long-term project that
>>>>>>>>> surfaces
>>>>>>>>> "news" protocol message-IDs, ....  It's a stable, standards-based
>>>>>>>>> protocol.
>>>>>>>>>
>>>>>>>>>
>>>>>>>>> Thunderbird, "SLRN", ....  Thanks for caring.  We care.
>>>>>>>>>
>>>>>>>>>
>>>>>>>>> https://groups.google.com/g/sci.physics.relativity/c/ToBo6XOymUw
>>>>>>>>>
>>>>>>>>
>>>>>>>> One fellow reached me via e-mail and he said, hey, the Googler
>>>>>>>> spam is
>>>>>>>> outrageous, can we do anything about it?  Would you write a
>>>>>>>> script to
>>>>>>>> funnel all their message-ID's into the abuse reporting?  And I was
>>>>>>>> like,
>>>>>>>> you know, about 2008 I did just that, there was a big spam flood,
>>>>>>>> and I wrote a little script to find them and extract their
>>>>>>>> posting-account,
>>>>>>>> and the message-ID, and a little script to post to the
>>>>>>>> posting-host,
>>>>>>>> each one of the wicked spams.
>>>>>>>>
>>>>>>>> At the time that seemed to help, they sort of dried up, here
>>>>>>>> there's
>>>>>>>> that basically they're not following the charter, but, it's the
>>>>>>>> posting-account
>>>>>>>> in the message headers that indicate the origin of the post, not
>>>>>>>> the
>>>>>>>> email address.  So, I wonder, given that I can extract the
>>>>>>>> posting-accounts
>>>>>>>> of all the spams, how to match the posting-account to then
>>>>>>>> determine
>>>>>>>> whether it's a sockpuppet-farm or what, and basically about sending
>>>>>>>> them up.
>>>>>>>
>>>>>>> Let me see your little script. Post it here.
>>>>>>
>>>>>> Here is a list I currently have:
>>>>>>
>>>>>> salz.txt
>>>>>> usenet.death.penalty.gz
>>>>>> purify.txt
>>>>>> NewsAgent110-MS.exe
>>>>>> HipCrime's NewsAgent (v1_11).htm
>>>>>> NewsAgent111-BE.zip
>>>>>> SuperCede.exe
>>>>>> NewsAgent023.exe
>>>>>> NewsAgent025.exe
>>>>>> ActiveAgent.java
>>>>>> HipCrime's NewsAgent (v1_02)_files
>>>>>> NewsCancel.java (source code)
>>>>>>
>>>>>> (plus updated python versions)
>>>>>>
>>>>>>
>>>>>>
>>>>>> (Maybe your script is inthere somewhere?)
>>>>>>
>>>>>>
>>>>>>
>>>>>> Show me what you got. walk the walk.
>>>>>>
>>>>>
>>>>>
>>>>> I try to avoid sketchy things like hiring a criminal botnet,
>>>>> there's the impression that that's looking at 1000's of counts
>>>>> of computer intrusion.
>>>>>
>>>>> With those being something about $50K and 10-25 apiece,
>>>>> there's a pretty significant deterrence to such activities.
>>>>>
>>>>> I've never much cared for "OAuth", giving away the
>>>>> keys-to-the-kingdom and all, here it looks like either
>>>>> a) a bunch of duped browsers clicked away their identities,
>>>>> or b) it's really that Google and Facebook are more than
>>>>> half full of fake identities for the sole purpose of being fake.
>>>>>
>>>>> (How's your new deal going?
>>>>>     Great, we got a million users.
>>>>> Why are my conversions around zero?
>>>>>     Your ad must not speak to them.
>>>>> Would it help if I spiced it up?
>>>>>     Don't backtalk me, I'll put you on a list!)
>>>>>
>>>>> So, it seems mostly a sort of "spam-walling the Internet",
>>>>> where it was like "we're going to reinvent the Internet",
>>>>> "no, you aren't", "all right then we'll ruin this one".
>>>>>
>>>>> As far as search goes, there's something to be said
>>>>> for a new sort of approach to search, given that
>>>>> Google, Bing, Duck, ..., _all make the same results_.  It's
>>>>> just so highly unlikely that they'd _all make the same
>>>>> results_, you figure they're just one.
>>>>>
>>>>> So, the idea, for somebody like me who's mostly interested
>>>>> in writing on the Internet, is that lots of that is of the sort
>>>>> of "works" vis-a-vis, the "feuilleton" or what you might
>>>>> call it, ephemeral junk, that I just learned about in
>>>>> Herman Hesse's "The Glass Bead Game".
>>>>>
>>>>> Then, there's an idea, that basically to surface high-quality
>>>>> works to a search, is that there's what's called metadata,
>>>>> for content like HTML, with regards to Dublin Core and
>>>>> RDF and so on, about a sort of making for fungible collections
>>>>> of works, what results searchable fragments of various
>>>>> larger bodies of works, according to their robots.txt and
>>>>> their summaries and with regards to crawling the content
>>>>> and so on, then to make federated common search corpi,
>>>>> these kinds of things.
>>>>>
>>>>>
>>>>>
>>>>
>>>> It's like "why are they building that new data center",
>>>> and it's like "well it's like Artificial Intelligence, inside
>>>> that data center is a million virts and each one has a
>>>> browser emulator and a phone app sandbox and a
>>>> little notecard that prompts its name, basically it's
>>>> a million-headed hydra called a sims-bot-farm,
>>>> that for pennies on the dollar is an instant audience."
>>>>
>>>> "Wow, great, do they get a cut?"  "Don't be talking about my cut."
>>>>
>>>> Usenet traffic had been up recently, ....
>>>>
>>>> I think they used to call it "astro-turfing".
>>>> "Artificial Intelligence?"  "No, 'Fake eyeballs'."
>>>
>>> I have NewsAgent111-MS.exe
>>>
>>> I seem to be missing version 2.0
>>>
>>> Do you have the 2.0 version?
>>>
>>> I'll trade you.
>>>
>>> I'll give you my python version with (GUI)!!!! (Tinter)
>>>
>>> let's trade!
>>>
>>> don't bogart
>>
>> I seem to be missing this version:
>>
>> https://web.archive.org/web/20051023050609/http://newsagent.p5.org.uk/
>>
>> Do you have it? you must have!
>>
>>
>>
>>
>>
>
> Nope, I just wrote a little script to connect to NNTP
> with a Yes/No button on the subject, tapped through
> those, and a little script to send an HTTP request to
> the publicly-facing return-to-sender in-box, for each.
>
> Here's all the sources you need:  IETF RFC editor.
> Look for "NNTP".  How to advise Google of this is
> that each domain on the Internet is supposed to
> have an "abuse@domain" email inbox, though there's
> probably also a web request interface, as with regards
> to publicly facing services, and expected to be
> good actors on the network.
>
> Anyways if you read through "Meta: a usenet server
> just for sci.math", what I have in mind is a sort
> of author's and writer's oriented installation,
> basically making for vanity printouts and generating
> hypertext collections of contents and authors and
> subjects and these kinds of things, basically for
> on the order of "find all the postings of Archimedes
> Plutonium, and, the threads they are in, and,
> make a hypertext page of all that, a linear timeline,
> and also thread it out as a linear sequence".
>
> I.e. people who actually post to Usenet are sometimes
> having written interesting things, and, thus having
> it so that it would be simplified to generate message-ID
> listings and their corresponding standard URL's in the
> standard IETF "news" URL protocol, and to point that
> at a given news server or like XLink, is for treating
> Usenet its archives like a living museum of all these
> different authors posts and their interactions together.
>
> I.e., here it's "belles lettres" and "fair use",
> not just "belles" and "use".
>
>
> It seemed nice of Google Groups to front this for a long time,
> now they're quitting.
>
> I imagine Internet Relay Chat's still insane, though.
>
> Anyways I stay away from any warez and am proud that
> since about Y2K at least I've never bootlegged anything,
> and never uploaded a bootleg.  Don't want to give old Shylock
> excuses, and besides, I wrote software for a living.
>
>


How to get a grip on the Gogole spams since six months ago or so
seems to be along the lines of trawling the newsgroups, pulling
down their posts since about that date, then extracting from
each one whether it has "KINDLE EPUB EBOOK" or Thai code,
has that mostly they can all be identified because they make
an X-Content-Transfer-Encoding base64 header, then that the content
is sort of inscrutable block unless its uncoded.

So it looks like it's possible to identify only off the subject
and other headers, then pretty definitively off the format,
which ones are these spams.

Then each one of these has a Google posting account in
one of the Google headers.

Injection-Info: google-groups.googlegroups.com;
posting-host=146.70.11.7; posting-account=hJ31DwoAAADGk9KnJ0tR36KM3U7DAsJC

It's that posting-account that basically is the abuser's,
whether or not it's an innocent dupe's after something like OAuth,
is undetermined.

Another sort of indicating bit is that they start with

X-Forwarded-Encrypted / X-Received / X-Forwarded-Encrypted / X-Received

pointing at some SMTP id's, what with regards to that looks like an SMTP
gateway,
what with respect to what might be alternate forms of message injection,
while the Path of each of the posts indicates as coming from
postnews.google.com .
Not all do, though.

Those though look like usual Google posters' posts, so it seems like
an automation of some Groups API on the Google side.

So anyways the idea is to

get the list of groups on a usenet server, GROUPS
get the count of headers since a few weeks ago
get the overview of headers
find likely spams
make a list of spammed groups
get the spammed count of headers since October
get the spammed overview of headers since October
find likely spams
pull down the headers
extract the posting-account


then part of the challenge is not including any threaded replies,
in the sense that some people replied to these posts in their rejection,
to make sure that an algorithm to mark spams is avoiding
Type I/II errors or the false positive/negative. I.e., such posts
in their replies, in their own content, don't have the same
characteristics.

(Or anything that contains bit ly links or "common exact-links in the spam".
Also all those "Case Analysis and Case Study Solution" spams,
look kind of similar. In fact when the spam started up I thought to
myself "hey I wonder if that's those 'Case Analysis and Case Study
Solution' spammers". )

For example on 10/2/23 ,I replied to a spam, so looking at it,
b0796889-551b-4637-be38-b590b5de2efcn@googlegroups.com
I would want to disambiguate spam reply rejections, from spams.
I'm not sure yet if the spams with same subject headers are
actually threaded replies or just have same subject, I imagine
that they just have the same subject header and aren't threaded replies.

So, there would basically be for cross-checking "likely spam"
(not replies, no references) and "likely not-spam" (replies, references).


Otherwise it does look just like one of the spams, with a
Content-Transfer-Encoding and that, but not the "encrypted SMTP"
bit.

"Note: Meta title tags should typically be around ...", is
one of the blast-fax mail-merge prompts that slips out,
with the idea of that finding that quote in the source
code will probably indicate the origin of the software.

So, the idea is to key off of posting account, then compute counts
for these sorts relations

posting-account -> email-address
posting-account -> targetted-group

then to compute how many spams were sent, by who, to where,
and whether posting-account <-> email-address is 1-1 or fraudulent,
then result a neat list of posting-accounts to batch up in a sort of
report and send it up to Google as a curated sort of spam report.

Then there's an idea that thus results a sort of spam rule,
about making "federated spam rules" type of a thing,
with regards to things like "spam blacklists" and these
kinds of things, and heuristics or rules, vis-a-vis, that
neural net classifiers are inscrutable and instead there
is to be a sort of "open quality rules" for relating messages,
to groups.

So, we can identify the spams, and, sort of to their origins.
Across all the Usenet groups.

[toc] | [prev] | [next] | [standalone]

Page 1 of 3  [1] 2 3  Next page →

Back to top | Article view | sci.physics.relativity

csiph-web