On 2026-08-22, Stefan Ram wrote:

| The Underlying Impact By forcing a login wall, Reddit attaches a
| distinct user ID to every single page view. This eliminates anonymous
| reading ("lurking") on the classic platform and further restricts the
| public's ability to view user-generated content without participating
| in their data-tracking ecosystem. Alongside the announcement, repre-
| sentatives warn that they "cannot promise" the legacy view will remain
| online indefinitely.

Now that's stupid, one would think they'd try to entice users "ok, now
you have to create an account and log in but we promise old.reddit.com
will always remain operational".

(I've only started to see the impact of this on recent days, I
think. old.reddit.com now indeed doesn't load the content. It doesn't
even load a login form, though...)

--
Nuno Silva

--- PyGate Linux v1.5.19
* Origin: Dragon's Lair, PyGate NNTP<>Fido Gate (3:633/10)

Stefan Ram <ram@zedat.fu-berlin.de> wrote:

| A Timeline of Reddit
|
| 2005 Reddit is founded by Steve Huffman and Alexis Ohanian as a
| simple link-aggregation site.
|
| 2005 Reddit rewrote its entire platform from Common Lisp to Python
| in December 2005, just six months after the site initially launched.
|
| 2008 Reddit goes open-source, allowing the public to see its code.
| The platform positions itself as an unmoderated haven for free speech.
|
| 2011/2012 The platform experiences massive growth. Subreddits are
| entirely managed by volunteer moderators with almost no corporate
| oversight.
|
| 2015 Steve Huffman returns as CEO. Reddit makes the controversial
| decision to shut down its open-source code repository, shifting toward
| proprietary software.
|
| Content Bans Reddit begins actively banning subreddits that violate
| hate-speech rules or generate negative press, marking the first major
| shift toward making the platform attractive to mainstream corporate
| advertisers.

I gave up here. You (or your chatbot) have just made the return of
"Steve Huffman" appear like a major milestone by implying that he
implemented "content bans" immediately afterwards. But your (AI's)
timeline hasn't even mentioned when he'd left before that!

Lines marked with "| " come from my editing, where I start by writing
prompts for the chatbot and then edit the generated texts and format
them for USENET. Chatbot might make mistakes and misrepresent or hal-
lucinate facts.

Yeah so why bother? It's a big company, it'll have this info on a
Wikipedia page with cites and at least some minimal form of peer
review. Typically I do a web search for a topic an AI has babbled
about and find lots of similar text on Wikipedia pages in the
search results - even when those pages are incorrect results
irrelevent to the subject (and the AI's response was complete BS).

In this case of course there _is_ a big Wikipedia page about
Reddit, and even a separate "Timeline of Reddit" page, which
although containing some similar entries, has many more overall,
and to me paints a quite different picture of the organisation.
Their first "Major Event" for example is:

"2005 Reddit's creators help seed Reddit with numerous fake
accounts to make Reddit seem more populated than it
actually was.[1]
[1] https://www.vice.com/en/article/how-reddit-got-huge-tons-of-fake-accounts--2/
"
That'd be enough to turn me off the platform if I'd ever
considered using it (which I haven't), but not worth a mention in
your (AI's) timeline. Of course Wikipedia also bothers to mention
when "Steve Huffman" left, before he returned:

"2009 October Alexis Ohanian and Steve Huffman leave reddit.
Steve Huffman helps form Hipmunk, and Alexis helps
form Breadpig.[19][20]

[19] http://www.redditblog.com/2009/10/fare-thee-well-reddit.html
[20] https://techcrunch.com/2010/09/01/reddit-cofounder-alexis-ohanian-to-join-y-combinator/
"

Oh and it says "Steve Huffman" returned _after_ the "content bans",
following a backlash, which completely reverses the clear
implication in your (AI's) timeline that "Steve Huffman" was
responsible for them:

"2015 June 10 Reddit bans five subreddits, citing an
anti-harassment policy.[63][64] The largest of the banned
subreddits, "fatpeoplehate," had an estimated 151,000 subscribers
at the times of its banning.[63] The other four subreddits are
"hamplanethatred," "transfags," "neofag," and "shitniggerssay."[63]

2015 June-July Reddit bans multiple subreddits and fires Victoria
Taylor, the site's director of talent, who has served on the Reddit
team since 2013. Taylor served as a liaison between the moderators
of specific subreddits (such as IAmA) and Reddit itself, helping
organize and verify interviewees for Reddit's user-led "AmA"
sessions. As a result of this and other frustrations with
Reddit--such as its moderation tools and its new conduct under
Pao--numerous subreddits (such as IAmA, todayilearned, pics and
science) temporarily shut themselves down in protest.[65]
Subsequently, to these and other recent events a petition asking
Pao to step down as CEO reaches over 160,000 signatures.[66] On
July 10, 2015, Pao resigns and is replaced by cofounder Steve
Huffman as CEO.[67]

[63] https://www.theverge.com/2015/6/10/8761763/reddit-harassment-ban-fat-people-hate-subreddit
[64] https://www.nytimes.com/aponline/2015/06/10/business/ap-us-reddit-subreddit-bans.html?_r=0
[65] https://www.wired.com/2015/07/reddit-amageddon/
[66] https://time.com/3946626/ellen-pao-reddit-petition/
[67] https://www.bloomberg.com/news/articles/2015-07-10/ellen-pao-resigns-as-reddit-interim-ceo-after-user-revolt
"
Now I don't have great faith in Wikipedia pages either, I prefer a
good old fashioned article from a single author, but your AI has
clearly left out so much important information from its timeline
that I have to consider it useless. Worse in fact, since any
conclusions one may draw from it will be based on a deeply
misleading narrative, and once formed it's difficult to forget
those conclusions even after the surrounding facts are better
revealed. Especially about something like Reddit where I'm going in
with barely any existing knowledge of the topic.


Well that's my AI-poster rant for the month. I was trying to ignore
these and not waste time arguing with Usenet's AI-fanbois anymore,
but I guess if you're just trolling then you got me. This AI BS is
always full of holes, but it takes 50x as long to find and prove
those holes than it takes to have the AI pump them out, and just so
the AI user can reply to me saying something like "Huh, well
everyone's using it, so who cares anyway?". Which depresses me
because it's probably true.

I will say if anything is going to push me off Usenet, it will be
if too much of this AI rubbish takes over here. I don't want to be
tricked into joining the fantasy land where you AI users seem happy
to live. In fact I'm increasingly worried that I'll have to avoid
everyone else in the world in order to achieve that...

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

--- PyGate Linux v1.5.19
* Origin: Dragon's Lair, PyGate NNTP<>Fido Gate (3:633/10)

On 8/22/26 18:15, Computer Nerd Kev wrote:

Stefan Ram <ram@zedat.fu-berlin.de> wrote:

| A Timeline of Reddit
|
| 2005 Reddit is founded by Steve Huffman and Alexis Ohanian as a
| simple link-aggregation site.
|
| 2005 Reddit rewrote its entire platform from Common Lisp to Python
| in December 2005, just six months after the site initially launched.
|
| 2008 Reddit goes open-source, allowing the public to see its code.
| The platform positions itself as an unmoderated haven for free speech. This was a great article. I 'discovered' Reddit around 2008 or so - it

was on the homepage of my favorite browser at the time, Opera, and I
quite enjoyed it for many years. It is the textbook posterchild of enshittification though. Well waddaya expect, Conde Nast bought it and intends to make a profit. There it is, the fundamental flaw of the
modern internet: looks like people creating things to share, but it's
really corporates mining engagement for profit in an advert economy.

Reddit pushed me so hard to install its app I stopped visiting the site.
Tried out Bluesky, which is a whole different set of flaws.

Usenet on the other hand ...

--- PyGate Linux v1.5.19
* Origin: Dragon's Lair, PyGate NNTP<>Fido Gate (3:633/10)

Retrograde <fungus@amongus.com.invalid> writes:

On 8/22/26 18:15, Computer Nerd Kev wrote:

Stefan Ram <ram@zedat.fu-berlin.de> wrote:

Usenet on the other hand ...

Have the same trajectory and now I'm back enjoin Usenet after decades.

I do not know what is it but most of languages and products follow the
same trajectory of eshitfication, only Common Lisp and Usenet did not.


--- PyGate Linux v1.5.19
* Origin: Dragon's Lair, PyGate NNTP<>Fido Gate (3:633/10)

bixbox <noreply@example.invalid> wrote:

Retrograde <fungus@amongus.com.invalid> writes:

On 8/22/26 18:15, Computer Nerd Kev wrote:

Stefan Ram <ram@zedat.fu-berlin.de> wrote:

Usenet on the other hand ...

Have the same trajectory and now I'm back enjoin Usenet after decades.

I do not know what is it but most of languages and products follow the
same trajectory of eshitfication, only Common Lisp and Usenet did not.

So far as I'm concerned Usenet is in the process of being ruined in
this very thread by people posting BS generated by AI chatbots, and
someone replying to me about how they enjoyed it even though I've
just pointed out how it's misrepresented key facts.

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

--- PyGate Linux v1.5.19
* Origin: Dragon's Lair, PyGate NNTP<>Fido Gate (3:633/10)

not@telling.you.invalid (Computer Nerd Kev) writes:

bixbox <noreply@example.invalid> wrote:

Retrograde <fungus@amongus.com.invalid> writes:

On 8/22/26 18:15, Computer Nerd Kev wrote:

Stefan Ram <ram@zedat.fu-berlin.de> wrote:

Usenet on the other hand ...

So far as I'm concerned Usenet is in the process of being ruined in
this very thread by people posting BS generated by AI chatbots, and
someone replying to me about how they enjoyed it even though I've
just pointed out how it's misrepresented key facts.

there is not much that we can do, maybe because it is a nice there is
not much that can be mint by those AI.

at some point it will be difficult do distinguish between generate stuff
and true human.
i recall in "the player of games" where fake video are indistinguishable
from real one and people start ignoring them.

and maybe we can start signing content with gpg to proove the
authenticity and build a personal list of signsture that we trust.

bix

--- PyGate Linux v1.5.19
* Origin: Dragon's Lair, PyGate NNTP<>Fido Gate (3:633/10)

bixbox <noreply@example.invalid> wrote:

not@telling.you.invalid (Computer Nerd Kev) writes:

bixbox <noreply@example.invalid> wrote:

Retrograde <fungus@amongus.com.invalid> writes:

On 8/22/26 18:15, Computer Nerd Kev wrote:

Stefan Ram <ram@zedat.fu-berlin.de> wrote:

Usenet on the other hand ...

So far as I'm concerned Usenet is in the process of being ruined in
this very thread by people posting BS generated by AI chatbots, and
someone replying to me about how they enjoyed it even though I've
just pointed out how it's misrepresented key facts.

there is not much that we can do, maybe because it is a nice there is
not much that can be mint by those AI.

at some point it will be difficult do distinguish between generate stuff
and true human.

Yes, this is likely an eventual outcome someday.

i recall in "the player of games" where fake video are indistinguishable
from real one and people start ignoring them.

and maybe we can start signing content with gpg to proove the
authenticity and build a personal list of signsture that we trust.

That only works if the signing human is trusted to not sign AI slop as
their own creation. If the signing human wraps AI slop in their
signature, then the authenticity angle of the signature falls apart.
Of course, if a 'trusted signature' starts signing obvious AI slop,
we'd have yet another angle with which to ignore the output of that
poster going forward.


--- PyGate Linux v1.5.19
* Origin: Dragon's Lair, PyGate NNTP<>Fido Gate (3:633/10)

Rich <rich@example.invalid> writes:

bixbox <noreply@example.invalid> wrote:

not@telling.you.invalid (Computer Nerd Kev) writes:

bixbox <noreply@example.invalid> wrote:

Retrograde <fungus@amongus.com.invalid> writes:

On 8/22/26 18:15, Computer Nerd Kev wrote:

Stefan Ram <ram@zedat.fu-berlin.de> wrote:

Usenet on the other hand ...

That only works if the signing human is trusted to not sign AI slop as
their own creation. If the signing human wraps AI slop in their
signature, then the authenticity angle of the signature falls apart.
Of course, if a 'trusted signature' starts signing obvious AI slop,
we'd have yet another angle with which to ignore the output of that
poster going forward.

Yes, it's about a circle of trust, as well as I trust that if someone
produce some genAI content it's clearly disclosed and I have no problem
with that.

bix




--- PyGate Linux v1.5.19
* Origin: Dragon's Lair, PyGate NNTP<>Fido Gate (3:633/10)

bixbox <noreply@example.invalid> wrote:

Retrograde <fungus@amongus.com.invalid> writes:

On 8/22/26 18:15, Computer Nerd Kev wrote:

Stefan Ram <ram@zedat.fu-berlin.de> wrote:

Usenet on the other hand ...

Have the same trajectory and now I'm back enjoin Usenet after decades.

I do not know what is it but most of languages and products follow the
same trajectory of eshitfication, only Common Lisp and Usenet did not.

It?s just because they?re old and obscure, and very few care or even
remember about them.

--- PyGate Linux v1.5.19
* Origin: Dragon's Lair, PyGate NNTP<>Fido Gate (3:633/10)