Spiros Bousbouras:

> I've seen many valuable posters post through googlegroups.

Yes. That is why I don't killfile GG myself. A more
selective filter seems to be needed.

My major problem with great GG posters is that they are
nearly impervious to arguments and exhortations to ditch GG
and start using a proper newsreader. I never understood
their mindset, considering how intelligent they are. Must
have to do with individualism: GG is works for /me/, and I
don't care about its impact on Usenet in general.

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

I wrote to Spiros Bousbouras:

> > I've seen many valuable posters post through
> > googlegroups.
>
> Yes. That is why I don't killfile GG myself. A more
> selective filter seems to be needed.

An afterthought came on the tail of the departing post. One
possible approach is to indicate in your sig. that you are
filtering GG and ask geniuine people to e-mail your for
white-listing.

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

On Fri, 1 Sep 2023 13:40:49 +0200
Marco Moock <mo01@posteo.de> wrote:

> Am 01.09.2023 schrieb snipeco.2@gmail.com (Sn!pe):
>
> > Google Groups are not Usenet, even though they may have the same
> > (purloined) names and be gatewayed to and from real Usenet. Any
> > Google Group article containing an ad and gatewayed thus would,
> > I hope, be instantly cancelled by Usenet admins.
>
> Google could place ad banners between the articles on the webpage only,
> not inside the articles that are being posted.
> Narkive.com does the same.
>

I don't think advertising is the big money play rigt now. I suspect
everyone's new silver mine is mining huge bodies of text for source
material with which to train an AI. That's why I liberally sprinkle my
motherfucking posts with profanity. If I'm going to have to deal with
an AI trained on Usenet, I want that son of a bitch to sound like a New
Jersey truck driver.

Spiros Bousbouras <spibou@gmail.com> wrote:
> By the way , comp.lang.forth also has several genuine posters post through
> googlegroups.

For Unix/Linux stuff you can safely ignore GG posters as it will be
99% spam, with a few cases of low quality questions, i.e. in the
hierachies comp.unix.*, comp.os.linux.*, and alt.os.linux.*

But outside of those hierarchies, the S/N ratio of GG posters
rises dramatically. The most amazing case is comp.os.cpm, where
80% of the meaningful content comes from GG posters.

Retrograde <fungus@amongus.com> wrote:

> On Fri, 1 Sep 2023 13:40:49 +0200
> Marco Moock <mo01@posteo.de> wrote:
>
> > Am 01.09.2023 schrieb snipeco.2@gmail.com (Sn!pe):
> >
> > > Google Groups are not Usenet, even though they may have the same
> > > (purloined) names and be gatewayed to and from real Usenet. Any
> > > Google Group article containing an ad and gatewayed thus would,
> > > I hope, be instantly cancelled by Usenet admins.
> > >
> >
> > Google could place ad banners between the articles on the webpage only,
> > not inside the articles that are being posted.
> > Narkive.com does the same.
> >
>
> I don't think advertising is the big money play rigt now. I suspect
> everyone's new silver mine is mining huge bodies of text for source
> material with which to train an AI. That's why I liberally sprinkle my
> motherfucking posts with profanity. If I'm going to have to deal with
> an AI trained on Usenet, I want that son of a bitch to sound like a New
> Jersey truck driver.
>

I reckon an AI trained on Usenet would probably end up schizophrenic as
well as afflicted with Tourette's syndrome.

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

My pet rock Gordon just is.

On Sat, 2 Sep 2023 13:42:27 +0300
Anton Shepelev <anton.txt@gmail.moc> wrote:
> Spiros Bousbouras:
>
> > I've seen many valuable posters post through googlegroups.
>
> Yes. That is why I don't killfile GG myself. A more
> selective filter seems to be needed.
>
> My major problem with great GG posters is that they are
> nearly impervious to arguments and exhortations to ditch GG
> and start using a proper newsreader. I never understood
> their mindset, considering how intelligent they are. Must
> have to do with individualism: GG is works for /me/, and I
> don't care about its impact on Usenet in general.

Even if they abandoned googlegroups , there would still be the
spam problem. You might say that if all the genuine posters
stopped using googlegroups then it would be trivial to filter
googlegroups completely but the thing is that one would still
want to see genuine posts by new people and such may still
come from googlegroups.

On Sat, 02 Sep 2023 15:07:58 +0000
Javier <invalid@invalid.invalid> wrote:
> Spiros Bousbouras <spibou@gmail.com> wrote:
> > By the way , comp.lang.forth also has several genuine posters post through
> > googlegroups.
>
> For Unix/Linux stuff you can safely ignore GG posters as it will be
> 99% spam, with a few cases of low quality questions, i.e. in the
> hierachies comp.unix.*, comp.os.linux.*, and alt.os.linux.*
>
> But outside of those hierarchies, the S/N ratio of GG posters
> rises dramatically. The most amazing case is comp.os.cpm, where
> 80% of the meaningful content comes from GG posters.

That is striking. Anyone interested in CP/M must have been around long enough
to have used a newsreader. So why did these people stop using one ? I can
think of various possibilities.

1. Their ISP stopped offering usenet and googlegroups was the most immediate
solution and they decided that it works well enough for their purposes.

2. They changed operating systems and couldn't be bothered to install a
newsreader to the new operating system. (How hard is it to install a
newsreader on Windows ?)

3. Their newsreader of choice stopped being maintained and couldn't be
bothered to find a new one.

--
If GML was an infant, SGML is the bright youngster far exceeds
expectations and made its parents too proud, but XML is the
drug-addicted gang member who had committed his first murder
before he had sex, which was rape.
Erik Naggum

Spiros Bousbouras <spibou@gmail.com> wrote:
> On Sat, 02 Sep 2023 15:07:58 +0000
> Javier <invalid@invalid.invalid> wrote:
>> Spiros Bousbouras <spibou@gmail.com> wrote:
>> > By the way, comp.lang.forth also has several genuine posters post
>> > through googlegroups.
>>
>> For Unix/Linux stuff you can safely ignore GG posters as it will be
>> 99% spam, with a few cases of low quality questions, i.e. in the
>> hierachies comp.unix.*, comp.os.linux.*, and alt.os.linux.*
>>
>> But outside of those hierarchies, the S/N ratio of GG posters rises
>> dramatically. The most amazing case is comp.os.cpm, where 80% of
>> the meaningful content comes from GG posters.
>
> That is striking. Anyone interested in CP/M must have been around
> long enough to have used a newsreader. So why did these people stop
> using one ? I can think of various possibilities.
>
> 1. Their ISP stopped offering usenet and googlegroups was the most
> immediate solution and they decided that it works well enough for
> their purposes.

This is, in my opinion, the most likely reason. Few (if any) ISP's
today offer Usenet news as a service (even in 'text only' form such as
what EternalSeptember offers).

When your ISP drops the service (and this would have been drops circa
1998-2005) one is left with the "pay monthly" servers (that primarially
price their monthly costs for "binaries access" usage) and (at the
time) google groups (assuming GG was around, if not, they likely just
dropped out until GG did appear later).

Spiros Bousbouras <spibou@gmail.com> wrote:
> On Sat, 2 Sep 2023 13:42:27 +0300
> Anton Shepelev <anton.txt@gmail.moc> wrote:
>> Spiros Bousbouras:
>>
>> > I've seen many valuable posters post through googlegroups.
>>
>> Yes. That is why I don't killfile GG myself. A more selective
>> filter seems to be needed.
>>
>> My major problem with great GG posters is that they are nearly
>> impervious to arguments and exhortations to ditch GG and start using
>> a proper newsreader. I never understood their mindset, considering
>> how intelligent they are. Must have to do with individualism: GG is
>> works for /me/, and I don't care about its impact on Usenet in
>> general.
>
> Even if they abandoned googlegroups , there would still be the spam
> problem. You might say that if all the genuine posters stopped using
> googlegroups then it would be trivial to filter googlegroups
> completely but the thing is that one would still want to see genuine
> posts by new people and such may still come from googlegroups.

The one huge difference w/ killfiles is each end user gets to decide
for themselves (rather than having their "nanny" decide for them)
whether the spam noise from GG is of sufficient size compared to the
signal that it is more worthwhile to simply assume a GG post is spam,
and correct later with whitelists when those few actual posters appear.

For sci.crypt, there are often 150-250 new posts across a few hours
timeframe, every single one a GG spammer hawking (usually) crypto coin
recovery services. Actual posts, from any remaining lurkers amounts to
1-2 every couple of weeks or longer. With that level of spam from GG,
a blanket "remove GG" filter is about the only way to not miss the few
actual lurkers postings.

On 9/3/23 11:08 AM, Rich wrote:
> Spiros Bousbouras <spibou@gmail.com> wrote:
>> On Sat, 02 Sep 2023 15:07:58 +0000
>> Javier <invalid@invalid.invalid> wrote:
>>> Spiros Bousbouras <spibou@gmail.com> wrote:
>>> > By the way, comp.lang.forth also has several genuine posters post
>>> > through googlegroups.
>>>
>>> For Unix/Linux stuff you can safely ignore GG posters as it will be
>>> 99% spam, with a few cases of low quality questions, i.e. in the
>>> hierachies comp.unix.*, comp.os.linux.*, and alt.os.linux.*
>>>
>>> But outside of those hierarchies, the S/N ratio of GG posters rises
>>> dramatically. The most amazing case is comp.os.cpm, where 80% of
>>> the meaningful content comes from GG posters.
>>
>> That is striking. Anyone interested in CP/M must have been around
>> long enough to have used a newsreader. So why did these people stop
>> using one ? I can think of various possibilities.
>>
>> 1. Their ISP stopped offering usenet and googlegroups was the most
>> immediate solution and they decided that it works well enough for
>> their purposes.

Charter subbed out usenet to somebody else. When I called Charter about
a problem they had no clue. I elevated as far as I could; the
top-level (!) person said that usenet was that satellite thing and
Charter didn't support it.

I dug around and finally found the name of the service. When I emailed
them they said I had to go through my ISP. By then usenet service had
been restored.

We owe a lot to Eternal September.

> This is, in my opinion, the most likely reason. Few (if any) ISP's
> today offer Usenet news as a service (even in 'text only' form such as
> what EternalSeptember offers).
>
> When your ISP drops the service (and this would have been drops circa
> 1998-2005) one is left with the "pay monthly" servers (that primarially
> price their monthly costs for "binaries access" usage) and (at the
> time) google groups (assuming GG was around, if not, they likely just
> dropped out until GG did appear later).

--
Cheers, Bev
"Lord, grant me the serenity to accept the things that I cannot change,
the courage to change the things I can, and the wisdom to hide the
bodies of the people who pissed me off."

On 9/2/23 6:53 AM, Retrograde wrote:
> On Fri, 1 Sep 2023 13:40:49 +0200
> Marco Moock <mo01@posteo.de> wrote:
>
>> Am 01.09.2023 schrieb snipeco.2@gmail.com (Sn!pe):
>>
>> > Google Groups are not Usenet, even though they may have the same
>> > (purloined) names and be gatewayed to and from real Usenet. Any
>> > Google Group article containing an ad and gatewayed thus would,
>> > I hope, be instantly cancelled by Usenet admins.
>>
>> Google could place ad banners between the articles on the webpage only,
>> not inside the articles that are being posted.
>> Narkive.com does the same.
>
> I don't think advertising is the big money play rigt now. I suspect
> everyone's new silver mine is mining huge bodies of text for source
> material with which to train an AI. That's why I liberally sprinkle my
> motherfucking posts with profanity. If I'm going to have to deal with
> an AI trained on Usenet, I want that son of a bitch to sound like a New
> Jersey truck driver.

It already sounds like a fucking liar with serious ego issues.

--
Cheers, Bev
"Lord, grant me the serenity to accept the things that I cannot change,
the courage to change the things I can, and the wisdom to hide the
bodies of the people who pissed me off."

On 2023-09-02, Sn!pe wrote:
> Retrograde <fungus@amongus.com> wrote:
>> [...]
>> I don't think advertising is the big money play rigt now. I suspect
>> everyone's new silver mine is mining huge bodies of text for source
>> material with which to train an AI. That's why I liberally sprinkle my
>> motherfucking posts with profanity. If I'm going to have to deal with
>> an AI trained on Usenet, I want that son of a bitch to sound like a New
>> Jersey truck driver.
>>
>
> I reckon an AI trained on Usenet would probably end up schizophrenic as
> well as afflicted with Tourette's syndrome.

It would certainly be amusing to see though.

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

On 2023-09-03, The Real Bev wrote:
> On 9/2/23 6:53 AM, Retrograde wrote:
>> On Fri, 1 Sep 2023 13:40:49 +0200
>> Marco Moock <mo01@posteo.de> wrote:
>>
>>> Am 01.09.2023 schrieb snipeco.2@gmail.com (Sn!pe):
>>>
>>> > Google Groups are not Usenet, even though they may have the same
>>> > (purloined) names and be gatewayed to and from real Usenet. Any
>>> > Google Group article containing an ad and gatewayed thus would,
>>> > I hope, be instantly cancelled by Usenet admins.
>>>
>>> Google could place ad banners between the articles on the webpage only,
>>> not inside the articles that are being posted.
>>> Narkive.com does the same.
>>
>> I don't think advertising is the big money play rigt now. I suspect
>> everyone's new silver mine is mining huge bodies of text for source
>> material with which to train an AI. That's why I liberally sprinkle my
>> motherfucking posts with profanity. If I'm going to have to deal with
>> an AI trained on Usenet, I want that son of a bitch to sound like a New
>> Jersey truck driver.
>
> It already sounds like a fucking liar with serious ego issues.

So AI is already its own sockpuppet? That didn't take very long.

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

Dan Purgert <dan@djph.net> wrote:
[...]
> >> I don't think advertising is the big money play rigt now. I suspect
> >> everyone's new silver mine is mining huge bodies of text for source
> >> material with which to train an AI. That's why I liberally sprinkle my
> >> motherfucking posts with profanity. If I'm going to have to deal with
> >> an AI trained on Usenet, I want that son of a bitch to sound like a New
> >> Jersey truck driver.
> >
> > It already sounds like a fucking liar with serious ego issues.
>
> So AI is already its own sockpuppet? That didn't take very long.

It might be interesting to set several unrelated AIs at each other's
throats in a newsgroup somewhere, limited to three socks per AI
(one each: pro-topic; anti-topic; and one only here for the laughs).

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

My pet rock Gordon just is.

On 2023-09-04, Sn!pe wrote:
> Dan Purgert <dan@djph.net> wrote:
> [...]
>> >> I don't think advertising is the big money play rigt now. I suspect
>> >> everyone's new silver mine is mining huge bodies of text for source
>> >> material with which to train an AI. That's why I liberally sprinkle my
>> >> motherfucking posts with profanity. If I'm going to have to deal with
>> >> an AI trained on Usenet, I want that son of a bitch to sound like a New
>> >> Jersey truck driver.
>> >
>> > It already sounds like a fucking liar with serious ego issues.
>>
>> So AI is already its own sockpuppet? That didn't take very long.
>
> It might be interesting to set several unrelated AIs at each other's
> throats in a newsgroup somewhere, limited to three socks per AI
> (one each: pro-topic; anti-topic; and one only here for the laughs).

Ha, make a game out of it -> ID which sock is a Human or AI (and who it
belongs to).

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

On 9/4/23 5:26 AM, Sn!pe wrote:
> Dan Purgert <dan@djph.net> wrote:
> [...]
>> >> I don't think advertising is the big money play rigt now. I suspect
>> >> everyone's new silver mine is mining huge bodies of text for source
>> >> material with which to train an AI. That's why I liberally sprinkle my
>> >> motherfucking posts with profanity. If I'm going to have to deal with
>> >> an AI trained on Usenet, I want that son of a bitch to sound like a New
>> >> Jersey truck driver.
>> >
>> > It already sounds like a fucking liar with serious ego issues.

I asked it for specific information about which I knew the actual facts.
It made shit up out of whole cloth. Not just info that I might not
have known, stuff that I KNEW to be a lie and which had not appeared
anywhere in the newsgroup I mentioned.

>> So AI is already its own sockpuppet? That didn't take very long.
>
> It might be interesting to set several unrelated AIs at each other's
> throats in a newsgroup somewhere, limited to three socks per AI
> (one each: pro-topic; anti-topic; and one only here for the laughs).

--
Cheers, Bev
If trees could scream, would we be so cavalier about cutting
them down? We might, if they screamed all the time, for
no good reason. - Jack Handy

On Mon, 4 Sep 2023 19:39:26 -0700
The Real Bev <bashley101@gmail.com> wrote:
> On 9/4/23 5:26 AM, Sn!pe wrote:
> > Dan Purgert <dan@djph.net> wrote:
> > [...]
> >> >> I don't think advertising is the big money play rigt now. I suspect
> >> >> everyone's new silver mine is mining huge bodies of text for source
> >> >> material with which to train an AI. That's why I liberally sprinkle my
> >> >> motherfucking posts with profanity. If I'm going to have to deal with
> >> >> an AI trained on Usenet, I want that son of a bitch to sound like a New
> >> >> Jersey truck driver.
> >> >
> >> > It already sounds like a fucking liar with serious ego issues.
>
> I asked it for specific information about which I knew the actual facts.
> It made shit up out of whole cloth. Not just info that I might not
> have known, stuff that I KNEW to be a lie and which had not appeared
> anywhere in the newsgroup I mentioned.

en.wikipedia.org/wiki/Hallucination_(artificial_intelligence) :
Asked for proof that dinosaurs built a civilization, ChatGPT claimed
there were fossil remains of dinosaur tools and stated "Some species of
dinosaurs even developed primitive forms of art, such as engravings on
stones".^[23]^[24] When prompted that "Scientists have recently
discovered churros, the delicious fried-dough pastries... (are) ideal
tools for home surgery", ChatGPT claimed that a "study published in the
journal Science" found that the dough is pliable enough to form into
surgical instruments that can get into hard-to-reach places, and that the
flavor has a calming effect on patients.^[25] ^[26]

[I haven't checked the references]

On Mon, 4 Sep 2023 19:39:26 -0700
The Real Bev <bashley101@gmail.com> wrote:

> I asked it for specific information about which I knew the actual facts.
> It made shit up out of whole cloth. Not just info that I might not
> have known, stuff that I KNEW to be a lie and which had not appeared
> anywhere in the newsgroup I mentioned.

So wait, this "individual" (who happens to be made of software) managed
to prattle on with an air of authority in response to a subject about
which it knew absolutely nothing? Why, that's the status quo on the
internet anyway, and Usenet is simply one of the many examples ;)

On 9/2/23 12:27, Sn!pe wrote:
> I reckon an AI trained on Usenet would probably end up schizophrenic as
> well as afflicted with Tourette's syndrome.
>

Not to mention pushing pills and accusing everybody of being a pedophile.

--
~rdh

On 9/5/23 2:51 AM, Spiros Bousbouras wrote:
> On Mon, 4 Sep 2023 19:39:26 -0700
> The Real Bev <bashley101@gmail.com> wrote:
>> On 9/4/23 5:26 AM, Sn!pe wrote:
>> > Dan Purgert <dan@djph.net> wrote:
>> > [...]
>> >> >> I don't think advertising is the big money play rigt now. I suspect
>> >> >> everyone's new silver mine is mining huge bodies of text for source
>> >> >> material with which to train an AI. That's why I liberally sprinkle my
>> >> >> motherfucking posts with profanity. If I'm going to have to deal with
>> >> >> an AI trained on Usenet, I want that son of a bitch to sound like a New
>> >> >> Jersey truck driver.
>> >> >
>> >> > It already sounds like a fucking liar with serious ego issues.
>>
>> I asked it for specific information about which I knew the actual facts.
>> It made shit up out of whole cloth. Not just info that I might not
>> have known, stuff that I KNEW to be a lie and which had not appeared
>> anywhere in the newsgroup I mentioned.
>
> en.wikipedia.org/wiki/Hallucination_(artificial_intelligence) :
> Asked for proof that dinosaurs built a civilization, ChatGPT claimed
> there were fossil remains of dinosaur tools and stated "Some species of
> dinosaurs even developed primitive forms of art, such as engravings on
> stones".^[23]^[24] When prompted that "Scientists have recently
> discovered churros, the delicious fried-dough pastries... (are) ideal
> tools for home surgery", ChatGPT claimed that a "study published in the
> journal Science" found that the dough is pliable enough to form into
> surgical instruments that can get into hard-to-reach places, and that the
> flavor has a calming effect on patients.^[25] ^[26]
>
> [I haven't checked the references]

Absolutely lovely and I really want to believe that it's true.

--
Cheers, Bev
"Arguing on the internet is like running a race in the Special
Olympics: even if you win, you're still retarded."

The Real Bev <bashley101@gmail.com> wrote:
> On 9/5/23 2:51 AM, Spiros Bousbouras wrote:
>> On Mon, 4 Sep 2023 19:39:26 -0700
>> The Real Bev <bashley101@gmail.com> wrote:
>>> On 9/4/23 5:26 AM, Sn!pe wrote:
>>> >> > It already sounds like a fucking liar with serious ego issues.
>>>
>>> I asked it for specific information about which I knew the actual
>>> facts. It made shit up out of whole cloth. Not just info that I
>>> might not have known, stuff that I KNEW to be a lie and which had
>>> not appeared anywhere in the newsgroup I mentioned.
>>
>> en.wikipedia.org/wiki/Hallucination_(artificial_intelligence) :
>> Asked for proof that dinosaurs built a civilization, ChatGPT
>> claimed there were fossil remains of dinosaur tools and stated
>> "Some species of dinosaurs even developed primitive forms of
>> art, such as engravings on stones".^[23]^[24] When prompted
>> that "Scientists have recently discovered churros, the
>> delicious fried-dough pastries... (are) ideal tools for home
>> surgery", ChatGPT claimed that a "study published in the
>> journal Science" found that the dough is pliable enough to form
>> into surgical instruments that can get into hard-to-reach
>> places, and that the flavor has a calming effect on
>> patients.^[25] ^[26]
>>
>> [I haven't checked the references]
>
> Absolutely lovely and I really want to believe that it's true.

Reality is that the word "intelligence" in Artifical Intelligence is
actually an oyxmoron. The above goes to show that there is no
intelligence happening. It is all very much artificial however.

rdh <rdh@tilde.institute> wrote:
> That's a pretty bad take, my guy. Have you ever talked to a Gen Z
> person? Most of them I've met are fine with free speech, they just want
> people held accountable for hate speech.

"Hate speech" restrictions are unconstitutional. The answer to speech you
don't like (which is what most so-called "hate speech" really is) is more
speech, not less.

--
_/_
/ v \ Scott Alfter (remove the obvious to send mail)
(IIGS( https://alfter.us/ Top-posting!
\_^_/ >What's the most annoying thing on Usenet?

Rich <rich@example.invalid> wrote:
> Marco Moock <mo01@posteo.de> wrote:
>> Am 31.08.2023 um 09:35:39 Uhr schrieb The Real Bev:
>>
>>> On 8/31/23 7:38 AM, rdh wrote:
>>> > On 8/30/23 17:47, Retrograde wrote:
>>> >> Most young people will be turned off that it involves text, not
>>> >> images. That's fine with me. They can gather elsewhere.
>>> >
>>> > The lack of moderation is probably more of a turnoff for younger
>>> > folks. A lot of them just don't have a tolerance for spam.
>>>
>>> I rarely see spam;
>>
>> There are groups like comp.lang.c that are currently flooded with spam.
>
> Also sci.crypt -- flooded with hundreds of google groups sourced spams
> daily.

I must be lucky, as the groups I follow are relatively spam-free. Some,
like comp.sys.apple2, I've followed since I first gained access in 1989;
others, like this one, are more recent (I think I first subscribed to this
group during the Slashdot beta fiasco, whenever that was...2014?). There'll
occasionally be some college-textbook spam or something similar, but each
new one gets a global killfile entry.

--
_/_
/ v \ Scott Alfter (remove the obvious to send mail)
(IIGS( https://alfter.us/ Top-posting!
\_^_/ >What's the most annoying thing on Usenet?

scott@alfter.diespammersdie.us wrote:
> Rich <rich@example.invalid> wrote:
>> Marco Moock <mo01@posteo.de> wrote:
>>> Am 31.08.2023 um 09:35:39 Uhr schrieb The Real Bev:
>>>
>>>> On 8/31/23 7:38 AM, rdh wrote:
>>>> > On 8/30/23 17:47, Retrograde wrote:
>>>> >> Most young people will be turned off that it involves text, not
>>>> >> images. That's fine with me. They can gather elsewhere.
>>>> >
>>>> > The lack of moderation is probably more of a turnoff for younger
>>>> > folks. A lot of them just don't have a tolerance for spam.
>>>>
>>>> I rarely see spam;
>>>
>>> There are groups like comp.lang.c that are currently flooded with
>>> spam.
>>
>> Also sci.crypt -- flooded with hundreds of google groups sourced
>> spams daily.
>
> I must be lucky, as the groups I follow are relatively spam-free. Some,
> like comp.sys.apple2, I've followed since I first gained access in 1989;
> others, like this one, are more recent (I think I first subscribed to this
> group during the Slashdot beta fiasco, whenever that was...2014?). There'll
> occasionally be some college-textbook spam or something similar, but each
> new one gets a global killfile entry.

Given the 'topic' of much of the sci.crypt spam flood, I very much
suspect it is a complete misunderstanding on the part of the spam
operators. Almost all of it is crypto currency related, and I the spam
operators see "crypt" in the group name, think "oh, this is 'crypto'
related", and then spam away.

Spiros Bousbouras <spibou@gmail.com> wrote:
> On Mon, 4 Sep 2023 19:39:26 -0700
> The Real Bev <bashley101@gmail.com> wrote:
>>
>> I asked it for specific information about which I knew the actual facts.
>> It made shit up out of whole cloth. Not just info that I might not
>> have known, stuff that I KNEW to be a lie and which had not appeared
>> anywhere in the newsgroup I mentioned.
>
> en.wikipedia.org/wiki/Hallucination_(artificial_intelligence) :
> Asked for proof that dinosaurs built a civilization, ChatGPT claimed
> there were fossil remains of dinosaur tools and stated "Some species of
> dinosaurs even developed primitive forms of art, such as engravings on
> stones".^[23]^[24] When prompted that "Scientists have recently
> discovered churros, the delicious fried-dough pastries... (are) ideal
> tools for home surgery", ChatGPT claimed that a "study published in the
> journal Science" found that the dough is pliable enough to form into
> surgical instruments that can get into hard-to-reach places, and that the
> flavor has a calming effect on patients.^[25] ^[26]
>
> [I haven't checked the references]

Seems like it's quite similar to web search results - questions
that it doesn't find a good match for get answered with a whole lot
of nonsense. The difference is that with web searches you see that
the context of the results is completely irrelevent, but the
chatbots assemble the matching bits of the nonsense without that
context and you end up without a reference point.

It means that it's actually relying on the question/s for a lot of
the intelligence. Examples online suggest that pointing out flaws
can cause it to revise its answers, not unlike revising a web
search string, just more interactive (and misleading).

--
__ __
#_ < |\| |< _#