Tavern on the Meta

starball

00:12

@Laurel the new identicon feature also seem to refresh every time you click the "Change picture" button. Also insteresting: hinduism.meta.stackexchange.com/q/2420/29115

Laurel

@starball How bizarre. I guess there is no more correlation between the image and anything else

I wonder if the Hinduism situation is something as simple as bad RNG

starball

@Laurel possibly. who knows? It could be that some user-specific info is still used, but a randomly generated / time-specific part is integrated into the calculation each time.

@Laurel or possibly a bad hashing algorithm?

Meta Andrew T.

@starball hmm, I thought it only refreshed when the user hasn't set an Identicon for the first time

Laurel

@starball If you do insecure RNG in a loop, duplicate random numbers can happen. (Discovered that here lol)

It also looks like this initiative removed profile images from SEDE, since they're all null there now

starball

00:40

@Laurel well, most (all?) PRNGs do that

@Laurel huh! interesting.

@MetaAndrewT. oh. maybe mine just hasn't been regenerated yet? I haven't switched to identicon after then change happened. I guess that's why?

Sonic the Anonymous Hedgehog

01:11

What year was the last controversy-free year of SE?

If I recall it started in 2018 with the IPS HNQ removal, then next year it was the Monica affair, then in 2020 it was the sudden termination of two long-time CMs, then some stuff happened in 2021 and 2022...

Journeyman Geek

02:51

@M.A.R. CEO Appreciation Stack Exchange?

Kevin B

03:10

@SonictheAnonymousHedgehog define controversy

Sonic the Anonymous Hedgehog

05:33

@JourneymanGeek Cross-site spammer (on sites you moderate): stackexchange.com/users/28716918/jai-hanuaman-grenite-marble

Shadow The Spring Wizard

06:40

@SonictheAnonymousHedgehog so 2017?

When Jeff left, it was obvious SO/SE won't ever be the same again.

Just matter of time.

@starball yeah, they just generate random hash now.

@Tinkeringbell so you're doing sport now, nice! ;)

@Andreasdetestscensorship this will just make them more dirty. :D

Tinkeringbell

@ShadowTheSpringWizard NEVER :P

Shadow The Spring Wizard

07:17

@Tinkeringbell but... but... you were jogging! :D

Found you doing it!

Well there were always things that annoyed users, e.g. killing the envelope, moving things around.... but not serious things like the recent events.

Sonic the Anonymous Hedgehog

07:42

@ShadowTheSpringWizard There was the top bar change that was received controversially, so I guess not?

@ShadowTheSpringWizard Also, IIRC there was at least one controversy during the Jeff era, but I wasn't around then so I could be misremembering

Also, belated welcome back to the Tavern!

SPArcheon

forgive me for being blunt, but when I see the word "misinterpreting" I think about not understanding some information that was shared with you.
Since the company shared no information other than empty useless buzwords that can mostly be summarized as "Everyone is doing AI. So cool. Much wow. We want to make money with AI too", user aren't left "misinterpreting" what the CEO is doing.

They are left GUESSING what they are trying to do.

Sonic the Anonymous Hedgehog

@ShadowTheSpringWizard In that case, what a bad time to rejoin the network when I did...

Oh, and I almost forgot about the November 2017 layoffs

SPArcheon

@PM2Ring Some will be optimistic like you, some will be quite angry because the company can't even share what their plan has (probably because they know that the "vocal minority" would not like it and that would cause "stress and nightmares"? I mean, if you really think you userbase would be happy... you would want to hype the great things you are doing, not keeping them hidden as long as you can to ***DELAY ANGER***), and some will utterly hate them because they are already angry from all the [redacted] you done in the past and this is only the new drop in an already overflowing cup.

Quite the opposite. Many currently assume ill intent by default. Is that something deserved? I ... don't really care, sorry. It doesn't matter if it is right or wrong, that is just the scenario they got themselves into and the sooner they accept it the better.

Sonic the Anonymous Hedgehog

In 2019, the then-Director of Public Q&A essentially instructed their staff to not listen to meta since everyone there is supposedly out to gang up on staff and the users there are supposedly a tiny minority not representative of the "actual" SE base. SE eventually got data to disprove these claims and hired someone whose purpose, among other things, was to remind them of this.

SPArcheon

07:57

(cont) That said, to be fair, they seem to already know that. The complete withdrawal from any form of exchange with the community is a clear sign that they are at least aware that trying to actually interact with the userbase would require work, more than usual due to the horrible relationship they have with those users right now.

Sonic the Anonymous Hedgehog

It's possible that the current management got this mentality again somehow, and with the one whose purpose was to not allow the company to slip back into it no longer working for SE (voluntarily resigned shortly before the first major blowup of this year). there was nothing stopping them.

I'm a big proponent of at-will employment, but in my view it was her working for SE that stopped other potential blowups before they could happen and her leaving was a bad decision.

SPArcheon

Problem is that instead of trying to douse the fires, with all the work that would imply they instead chose the "opt out" solution.
They choose to go blind and deaf to the users and are currently trying to man the ship alone.

And trust me, this does not work. I saw this before.

the arrogant idea that you don't need the community to man a community will soon or later blow in your face

Sonic the Anonymous Hedgehog

@SPArcheon To be entirely fair, they could be discussing in internal chat rooms how to respond. But that's still not an excuse for remaining silent in the public record, in my view.

Oct 16, 2019 at 18:28, by Sonic the Anonymous Hedgehog

The community is like the structure of a building. You can keep making profits renting out spaces in the building, but if you don't take care of the structure, it will eventually all fall apart.

SPArcheon

I used to help on a online game server. After some changes the latest admins got that splendid idea that "we are the ones running the place, so if you don't like what we do the exit is right there".
Oh, how great it worked! No need for public relationship, no need to discuss, no need to tell others what you were doing...
Problem is, the server shut down about 6 months after. A server that had less than 10 players playing on peak hours can as well as die.

@SonictheAnonymousHedgehog I would be realistic. There was no real reply to many other issues in the past and not even the more recent ones got them to actually say something useful.

I mean, recently there was that other "scandal" about SE "selling training data".

Sonic the Anonymous Hedgehog

@SPArcheon My position on the topic has been, discuss internally how much ever you want, but make it clear that you've at least seen the post and are discussing it. A simple comment saying "We're aware of this post and are discussing how to respond to it" goes a much longer way than complete radio silence.

SPArcheon

08:11

Someone mentioned here that an user even got as far as asking on Laws.SE if they could be legally persecuted for that.

11

Q: Is it illegal for a firm to train an AI model on a CC BY-SA 4.0 corpus and make a commercial use of it without distributing the model under CC BY-SA?

https://meta.stackexchange.com/q/388551/178179 mentions that SE will force some firms to pay to be allowed to train an AI model on the SE data dump (CC BY-SA licensed) and make a commercial use of it without distributing the model under CC BY-SA. This makes me wonder: Is it illegal for a firm to ...

I would be very incredulous if any SE employee told me that they didn't saw that.

I think it gives a clear message.

Sonic the Anonymous Hedgehog

Or even, "We've discussed this feedback and have decided not to incorporate it" eliminates half the backlash.

SPArcheon

> "Your userbase think that what you are doing is an act of force and not even legal"

You would think that anyone with a head on their shoulder would be ALARMED by this.

They would run to clarify thing, to show user that they are not planing anything evil.

Instead... nothing.

Sonic the Anonymous Hedgehog

@SPArcheon That reminds me of a formerly open-source software that went commercial proprietary. The new owners made use of a loophole in the GPL that allows developers to charge for compiled binaries as long as the code is released free of charge. Initially they had no problem with users compiling the code themselves, but eventually they rebuilt the tool from the ground up and stopped releasing it under the GPL.

SPArcheon

Silence. Hay balls rolling on the wind in the deserts.

But wait. Silence would actually be BETTER than what they did.

Almost at the same time someone asked if "Is it illegal for a firm to train an AI model on a CC BY-SA 4.0 corpus and make a commercial use of it without distributing the model under CC BY-SA?", clearly alluding at the company, someone posted a blog post with this title:

> "When AI meets IP: Can artists sue AI imitators? (Ep. 566)"

You can tell me that this is just an unfortunate coincidence. I won't believe you. And anyway even if it was it only proves that you should do more homework before showering us with market oriented buzword bloated blog posts

.... someone wondering if you are doing them wrong by reselling the content THEY wrote and you post a blog titled "When AI meets IP: Can artists sue AI imitators? (Ep. 566)"....

This is worse than silence.

this is L.O.V.E.

And yep, you are reading it correct. I didn't write "Love". I wrote L.O.V.E.

Shadow The Spring Wizard

@SonictheAnonymousHedgehog that's nothing compared to Monica, and now the AI scandals.

SPArcheon

08:20

aka.... this statue.

L.O.V.E. (sculpture)

L.O.V.E., commonly known as Il Dito (Italian for 'the finger') is a sculpture by Italian artist Maurizio Cattelan consisting of a hand with all the fingers severed with the exception of the middle finger. The sculpture is located in Piazza degli Affari in (Milan), where the Italian stock exchange is located. The name L.O.V.E. is the acronym of "Libertà, Odio, Vendetta, Eternità" ("Freedom, Hatred, Revenge, Eternity"). The sculpture, built in 2010, was originally exhibited on the occasion of Cattelan's retrospective at The Royal Palace of Milan. After the exhibition closed, the city Councillor for...

Shadow The Spring Wizard

@SonictheAnonymousHedgehog probably envelope, the early version of inbox. It had features that were just lost, to this day.

@SonictheAnonymousHedgehog thanks!

@PM2Ring no, not at all. He's clearly talking only about Generative AI, that actually create something.

SPArcheon

@ShadowTheSpringWizard Personally, I think that the product they are aiming to is quite clear.

Tinkeringbell

@Andreasdetestscensorship Yeah, that'd be ideal. The stupid part is that users really do use the LLM to create opinions for them...

Shadow The Spring Wizard

@SonictheAnonymousHedgehog well from what I remember, that was "valid" reason of actually not having enough money to keep them all.

SPArcheon

@ShadowTheSpringWizard A "coding support" bot that you can interrogate and will provide code snippets based on SO content

Shadow The Spring Wizard

08:25

@SPArcheon ChatGPT can already do it, nobody will use a buggy bot like that.

@SonictheAnonymousHedgehog and that someone left, without anyone to replace her.

Sonic the Anonymous Hedgehog

@ShadowTheSpringWizard Yes, if you read the two messages I posted after that you'll see my whole point

SPArcheon

@ShadowTheSpringWizard everyone mentions ChatGPT here because they only know that name.

ChatGPT is only a pretty generic model for chat.

and like stable diffusion, the model matters a lot.

Shadow The Spring Wizard

@SonictheAnonymousHedgehog not a bad decision. At this point I'm sure she simply saw what's going to happen, and this time it was bigger than her and she knew she wouldn't be able to stop it, so she chose to leave in order to not be part of it. Clever choice.

@SPArcheon doesn't matter. They're all bad.

SPArcheon

@ShadowTheSpringWizard humbug.

If I try to generate a portrait of "Fluttershy" using the StableDiffusion 2.0 model I get this eldritch abomination

Shadow The Spring Wizard

All those fancy picture generators aren't relevant, they can't be used to answer questions, they can't cause any real harm.

Those are really just a nice thing to have, nothing less, nothing more.

SPArcheon

08:35

If I use a model that was trained specifically on MLP pictures I get this.

I think that unless you too want to be deaf you can get the point.

Train a code generation model on "Alice in Wonderland", you get a mess.

Train the model on a well written and commented code and it may work.

It won't be perfect, it won't be great, but it will be enough for homeworks.

Shadow The Spring Wizard

@SPArcheon that's the whole problem. It may work.

SPArcheon

@ShadowTheSpringWizard Homeworks, the untapped potential userbase that those dastardly vocal users don't want to help, I may add

And who cares if it "doesn't work that great". If the ones trying to get their homework done are happy, "we" won.

And by "we" I mean the company.

SPArcheon

08:54

@JourneymanGeek I was writing an answer to this

30

Q: The new Code of Conduct didn't address significant feedback given. Why ask if you won't incorporate feedback?

Some of the content is unnecessarily verbose. I had pointed some examples out during the review of the draft (both when it was shared with moderators and again in the public review). The "Unacceptable Behavior" section is full of this verbosity. This level of verbosity hurts the readability - and...

discussion review code-of-conduct feedback

do you know if this is referring to the community visible "announcement" post or some separate question made to mods on your private channels?

I don't want to post an answer about the meta post if this is about something happening in the Teacher Lounge

probably @Tinkeringbell may know too

Journeyman Geek

@SPArcheon A few mods, myself included did do a review of it on a seperate SO team

SPArcheon

09:18

Oh, btw... I actually just proposed a change to the mod-agreement-policy tag

since apparently the users never got what the actual intended meaning of the tag is , and the original version never made it clear that answers to such post will be deleted... I made that explicit

Feel free to approve it if you have the power to

Tinkeringbell

@SPArcheon All mods got a preview on the SOfT for all moderators, to give feedback on, and then similar feedback was again given on MSE, and in some cases nothing was done with it both times. There was a select group of mods (that Geek referenced) that got to review it already before it even reached the SOfT for all moderators.

I am quite sure Thomas is referencing the first two, and not the one Geek mentioned.

SPArcheon

09:36

I am thinking to post a question asking if new announcement are poorly written by purpose

See the new policy.

Journeyman Geek

@SPArcheon naw, its just a different way of writing

Its very important we focus on policies and the company direction rather than personal attacks of any kind

and asking if someone wrote it poorly on purpose is personal.

SPArcheon

@JourneymanGeek Yep but I was thinking of something. Most disagree with the policy, but some are like "maybe they actually meant that they just wish mods to be carefully when moderating."

Journeyman Geek

ah

there's a lot of .. uh... conversation going on amongst mods and other community folks going on

SPArcheon

If you post the thing so that it is clear what you mean, you can't have the incurable optimistic users go like "maybe they meant that"

Journeyman Geek

....

points at very long COC

Part of the aim there is to make it hard for people to say 'its not in the rules'

On the other hand, we're not law enforcement, and to some extent, trusting moderators includes giving them some freedom to work out what to do, and what's not right.

SPArcheon

09:41

@JourneymanGeek yeah, the "oh so great Code of Chaos"

Rule 1: be nice, don't be rude, don't be condescending, don't namecall users as if they are wimp chickens that can't live in the real world.

seconds later:

> **Get over it. It's normal. In fact, it's healthy and appropriate.**

Community standards do not maintain themselves: They're maintained by people actively applying them, visibly, in public. Don't whine that all criticism should have been conveyed via private e-mail: That's not how it works. Nor is it useful to insist you've been personally insulted when someone comments that one of your claims was wrong, or that his views differ. **Those are loser attitudes.**

Journeyman Geek

they get called a limp turkey, and someone complains its not specifically banned

@SPArcheon Ah ... actually

I think that's a big disconnect here

SPArcheon

> Get over it kiddo! Grow a [redacted] spine? What you are? A real man or a wimpy crybaby girl? They offended you? So what? Tim lost his keys again, that is the reason. Now go to a gym and do some body building so that you can be an alpha.

I always despised that "Tim lost his keys" meme.

Yep, you are the one wrong for felling hurt. "Be a real man and grow a spine"

Tinkeringbell

Well I've never seen a human without a spine survive, so in that sense, all man need to grow one :P

SPArcheon

@Tinkeringbell if you really want to nitpick that also means that anyone you can tell to "grow a spine" already has one, so you are asking them to grow a second one (or maybe get taller, I don't know if that counts too)

Tinkeringbell

10:00

Heh, maybe!

Shadow The Spring Wizard

10:43

@Tinkeringbell well, better phrase would be "make your spine stronger", so that you can stand tall and withstand insults/pain/whatever better.

This might be legit thing to say to your kids as part of education, but when said to adults you don't know, it's arrogant, implying you're better than them.

I'm also sometimes annoyed by new users being insulted and offended from downvotes or having their posts deleted, but will never tell them such things.

Tinkeringbell

@ShadowTheSpringWizard "grow thicker skin" ;)

SPArcheon

@ShadowTheSpringWizard the issue that I have is the generalization and oversimplification of the issue.

it is an objective truth that sometime new users are alienated by the system in place and it is also true that sometime the new users are perfectly right in doing so because other users are condescending, rude or plain toxic against them.

Yet we generalize the issue to the point that by default the reason is that the user is whining and we even made a meme out of it.

One serious question and a less serious one

First: where exactly the new CoC discuss the AI generated content policy - if that is discussed at all.

and second: Does the new policy forbid the deletion of content because it is AI generated or does it forbid the deletion of AI generated content. Those are two quite different things.

Journeyman Geek

11:24

@SPArcheon we can't determine content is AI generated unless they confess

Shadow The Spring Wizard

@Tinkeringbell but... no spine... ;)

@SPArcheon most likely not at all.

@SPArcheon got a comment from a mod that they can delete only if low quality, They are forbidden from checking if it's AI generated, so deletion upon it isn't relevant.

So this means that "high quality" (i.e. appears as high quality) AI generated content will stay.

@SPArcheon sure, but when people try to explain nicely, it never works either.

SPArcheon

11:46

@JourneymanGeek I was more thinking about the exact wording. I wouldn't be surprised if it is formulated so that technically it is forbidden to delete an AI generated racist/inappropriate/spam message

or at least so that using a tool to identify a racist post is a violation if the post is also AI generated

Glorfindel

12:02

So much for consistency (the program is Visual Studio for Mac)

Tinkeringbell

12:41

Lovely!

SPArcheon

Those who know what this account is about can understand the reasoning.

PM 2Ring

12:58

@SPArcheon Oh, I'm not very optimistic, but I am hopeful that the CEO does have some actual intelligence, despite sounding like a marketing buzzword generator.

"if you really think you userbase would be happy... you would want to hype the great things you are doing, not keeping them hidden [...]" Fair point, but maybe they're still exploring what they can realistically do with AI, so they can't give us solid details for all that stuff yet. But they have told us about one of these new AI tools: the thing that helps with question title creation.

But I fully agree that it is disappointing that they make vague / ambiguous statements. Surely they know by now that that just leads to pessimistic speculation on MSE.

In the absence of solid information, ChatGPT hallucinates. In the absence of solid information, MSE hallucinates. ;) — PM 2Ring Apr 19 at 0:37

SPArcheon

@PM2Ring Yep, and to continue with the analogy we hallucinate based on a training model that has Monica events in it.

PM 2Ring

@ShadowTheSpringWizard Ok, but being enthusiastic about Generative AI doesn't necessarily imply that they want to permit answers that are simple copy & paste from ChatGPT output. OTOH, LLM could be used to clean up typos & grammar of existing and new posts, and it could be used to make a conversational interface to allow intelligent searching of the existing content.

@SPArcheon Agreed. And although there have been some improvements since those days there have been numerous negative events that keep many of us sceptical.

I can understand that SE is concerned that there's been a reduction in traffic on the network, especially on SO, since ChatGPT was released. However, I expect that a lot of the people getting help from ChatGPT instead of SO would have been posting low-quality questions on SO. That sounds like a good outcome to me. If you're a cargo-cult coder, ChatGPT is probably a useful tool. ;)

Feeds

14:04

Building a safer community: Announcing our new Code of Conduct

Philippe Beaudette on June 01, 2023

Since we last updated our Code of Conduct in 2018, the world has shifted dramatically. Hear from our VP of Community as we dive into our newest updates to the Code of Conduct.

Shadow The Spring Wizard

14:20

LOL

@PM2Ring but they now do permit copy and paste from ChatGPT. As for AI-powered search, that something I'm fine with, many people find search to be terrible. But not in the cost of more layoffs.

PM 2Ring

@ShadowTheSpringWizard Not quite. From meta.stackoverflow.com/a/424911/4014959

> The policy is not clear. It has not yet been effectively communicated to moderators. Worse, the statements being made in public are rather different (in details, but critical details, specifically the ones we find most objectionable) from the statements that have been made to us in private.

And of course, no matter what AI policy SE foist on us, we can always downvote raw unverified ChatGPT output because it is "Not useful".

SPArcheon

14:38

> A collaborative effort
A Code of Conduct is a handshake agreement between users and the company and is a collaborative effort that involves the invaluable insights of the community. We have actively engaged with moderators and the community, seeking their perspectives and expertise. Their contributions have been instrumental in shaping this update, ensuring that it reflects the diverse needs and voices of the community.

Collaborative?

PM 2Ring

14:48

FWIW, I learned about SO from Amber Yust, who also has a Fluttershy avatar. Amber was a prominent contributor on the old XKCD forums. She is currently the Director of Privacy Engineering at Google.

Kevin B

i mean, i think it's fair to say there was some collaboration

SPArcheon

@PM2Ring don't give @ShadowTheSpringWizard weird ideas.

Kevin B

it was just completely one sided

SPArcheon

It is already enough when I try to make him believe I am actually Lauren Faust.

PM 2Ring

:)

SPArcheon

15:03

That said, I once mentioned learning Tea Ceremony so... who knows? Maybe I am the Fluttershy Japanese VA

Enough with trolling Shadow for now, those cupcakes won't eat themself.

Nord The Star Wizard

@PM2Ring I feel like that's a bit optimistic (but I would love for it to be true). I can already imagine the bad feedback loop, where users who glorify it come to people (by asking questions) on SO/SE and ask why it said XYZ, and then complain why their question was closed. It's a good way to give headache to the community

@PM2Ring there a reduction in traffic, but I assure you, people have been indeed spamming SO/SE with generated answer, sometimes even admitting they used it

sometimes even using it in the comments themselves

Shadow The Spring Wizard

@PM2Ring not directly, sure. But if they ban all detectors, it has exact same result.

Nord The Star Wizard

@PM2Ring downvoting is fine, but this won't help show that the answer is truly bad. A lot of people accept and upvote answer without actually trying them out (especially on SO)

once it's done, then the answerer usually won't answer back, since they're already have their upvote and accepted rep. (especially if they generated their answer 1:1)

Shadow The Spring Wizard

@SPArcheon wait, you're not??

:D

Nord The Star Wizard

@PM2Ring I used some of their answers at the time. Already knew they were working at Google, but didn't know about the involvement with XKCD :o

Shadow The Spring Wizard

15:10

@NordTheStarWizard safe to say "she" in that case. :)

(not due to avatar on SO, see e.g. the avatar on her blog)

Nord The Star Wizard

true, I just do that out of habit

SPArcheon

@ShadowTheSpringWizard I am clearly an Half-Genie Dev, since I share the same treatment of my honorary sister Shantae. Customer causes chaos, I get the blame :P

PS: now, would the fact that Half-Genies are only females mean anything?

Shadow The Spring Wizard

@NordTheStarWizard which is good as default, sure. ;)

SPArcheon

Shadow The Spring Wizard

@SPArcheon hmm

SPArcheon

15:16

> how to keep the wizard busy for hours, looking for a clue
how to keep the wizard very very busy for hours, looking for a clue

PM 2Ring

@ShadowTheSpringWizard Yes, Amber does prefer "she" pronouns.

Nord The Star Wizard

@PM2Ring didn't know, make sense

@SPArcheon so you have a guide on how to make Shadow run in circle?

PM 2Ring

@ShadowTheSpringWizard Maybe. I've never used those detectors. I've only flagged a few answers (on Physics & Astronomy), saying "Possible ChatGPT answer" in my flag comment.

Nord The Star Wizard

you don't need the obviously biased detector. You can literally look for obvious tell-tale sign

Kevin B

unfortunately mods can't tell which flags are based on NI and which ones are based on a detector browser extension

PM 2Ring

15:25

Agreed. None of my flags were declined. I've also got a perfect track record on plagiarism flags.

@KevinB Sure, but mods claim they aren't mindlessly approving GPT flags, they do their own investigation, which may admittedly include use of detectors.

SPArcheon

@NordTheStarWizard just don't tell them your gender and keep spreading herrings that can go both side.

Like, tell them that you dressed as someone from Castlevania Rondo of Blood at an anime con and as soon as they assume Richter remind them that Maria exists... things like that.

M.A.R.

@SonictheAnonymousHedgehog I think 2022 was relatively uneventful. People even started hoping that maybe alongside the 'superfluous if not actively harmful' changes a couple of decade-old requests would be worked on. Which is also part of why this thing stings quite a bit for some people, and why @Journey is writing theses on meta.SE again

Okay, serious question, do I need to check the new CoC or do I just dismiss the banner

Nord The Star Wizard

@SPArcheon take notes Interesting

M.A.R.

I'm still waiting to know what "Community is the future of AI" is supposed to mean. meta.stackexchange.com/questions/388401/… — PM 2Ring 15 hours ago

PM 2Ring

@NordTheStarWizard Unfortunately, one of those tell-tale signs is when someone with a history of poor English posts an answer with great grammar, so the technique is biased at detecting raw ChatGPT posts from ESL people.

21 hours ago, by PM 2Ring

Unfortunately, one of the simple "tells" of bad content is poor English, so when we see good English coming from a member with a track record of bad English we're immediately suspicious that they're posting GPT content, or plagiarising human-generated content. And that can look like we're picking on people because of their race / country of origin...

Nord The Star Wizard

15:36

@PM2Ring it isn't :) I have been helping in flagging some of those post and giving suggestion. I assure you, poor grammar and biased detector wasn't the only way to find them

Kevin B

@PM2Ring im aware, that wasn't my point

it's that claiming i, or anyone else, don't even use detectors doesn't really... matter, because how the flag is handled by the mod is all that matters and they're obviously doing more than the bare minimum

M.A.R.

@PM2Ring well, we are also biased against newer questions because we can't manage to close all the older questions that violate newer site rules. Why does there need to be a rule to enforce?

2

Shadow The Spring Wizard

@M.A.R. nothing to see there, so yeah you can dismiss it.

M.A.R.

Moderation is sometimes easy, sometimes difficult, but always only feasible with community input, which has never been fully consistent

PM 2Ring

15:52

in Meta Stack Exchange Comment Archive, yesterday, by Boson - StandWithUkraine

I wonder if any current AI detector can determine that the words of the CEO aren't generated by a marketing bot... — PM 2Ring 5 secs ago

Nord The Star Wizard

I wouldn't be surprised if it was mostly generated either by them or someone who work for them

Shadow The Spring Wizard

16:15

@SPArcheon well, that would work, yeah. But pretty sure @Nord already burned himself by approving his gender at some point. ;)

@PM2Ring rofl

dang it appear as link in the starwall. :/

SPArcheon

@ShadowTheSpringWizard and what if they lied? Can you bear that doubt?

Shadow The Spring Wizard

@NordTheStarWizard our CEO is male, that's certain. And no, I don't think he wrote it himself, there are employees who work in marketing who (should) know how to write in a way that attract readers.

Fact they have no clue what the company actually is can explain the content.

Nord The Star Wizard

@ShadowTheSpringWizard I know, I meant this generally (as in, whoever take care of the blog part of SO/SE)

Shadow The Spring Wizard

@SPArcheon nah, I have a good detector. :D

Nord The Star Wizard

@ShadowTheSpringWizard I agree

Shadow The Spring Wizard

16:19

@NordTheStarWizard :)

Big Joe

@Glorfindel do you think you will ever get to 300 or 400 rep ?

Shadow The Spring Wizard

@NordTheStarWizard oh, there are likely quite a few females there, so you were right to use the general case pronoun then. ;)

@BigJoe you mean 300k or 400k on MSE? Or 300 or 400 on some other site he's still not active in?

But yeah, if he'll keep his activity same as these days, he should reach 300k in 2-3 years, if Stack Exchange, or just Meta Stack Exchange, will still exist by then.

I have a feeling that if we'll keep ranting loud enough, they'll simply shut down Meta and be done with it.

They considered it for real not so long ago, so it's not out of the blue. The plan is there, waiting in a drawer.

But maybe it's for the best. It will surely save lots of time to lots of people.

But the side effect of shutting down Meta Stack Exchange (and likely MSO as well) is mass leaving of the vast majority of the old time users, leaving SO and the other sites a pile of low quality content, i.e. end of SE.

</future-prediction>

SPArcheon

16:39

@ShadowTheSpringWizard chat rooms could still be useful to post AI gen images (after all they seem to like AI so it is only due to fill their imgur with those, right?).

Especially since the alternative is some site like Discord.

Shadow The Spring Wizard

@SPArcheon well yes, they might even migrate chat to its own domain to keep it alive.

(I mean main domain, as the sites themselves would be shut down.)

@SPArcheon Discord isn't a site. I think? Never even checked. lol

Spevacus

It has a web version.

It's... Okay. The app is leagues better.

SPArcheon

@Spevacus yet from what I heard the site is less of a spyware than the app is.

Spevacus

The app, likely, collects more info about your device, yeah.

Shadow The Spring Wizard

@Spevacus I don't like web versions of apps originally made for mobile devices.

Spevacus

16:50

While I cannot confirm, I believe the app was made for desktop before it was made for mobile.

Shadow The Spring Wizard

If it's meant for a mobile device, it should be used on a mobile device. ;)

@Spevacus Discord, perhaps. But pretty sure WhatsApp was first for mobiles.

Luuklag

17:03

O/

Big Joe

Yea, @ShadowTheSpringWizard I meant 400k or 300k on MSE

Tinkeringbell

18:05

@Luuklag \o. You've got some timing for dropping by ;)

These are dark times

Luuklag

18:30

Yeah found a few minutes in between things

@Tinkeringbell how come?

I saw new vote buttons, which don't look to pleasing.

Spevacus

@Luuklag That, plus a new decree that prohibits the direct deletion/suspension of AI-generated content posters, plus a couple other things that are not yet publicly announced yet that aren't... that great.

AI content policy post: meta.stackexchange.com/q/389582/622284

JMG also did a thing: meta.stackexchange.com/questions/389558/…

M.A.R.

@Luuklag it's night outside

Tinkeringbell

18:58

@M.A.R. it's 9PM here and outside is showing no signs of night, don't worry :P

@Luuklag Pretty much what Spevacus said

Luuklag

Mmm that Ai policy was bound to be troublesome, I noticed discussions when chatgpt first came around.

Tinkeringbell

It's definitely going to be trouble now

Luuklag

19:14

Thats sounds like just another day at SE

Shadow The Spring Wizard

19:43

@Tinkeringbell sure it's the actual moon and not me faking it? :P

@Luuklag nah, it's worse this time. Looks like management decided to cut any communication with the community (actual community, not the fake term CEO is using) and they don't even try to fix things anymore. Neither technical (no non-minor bug was fixed for... years?) nor the loss of trust, as they keep telling outright and obvious lies, and paint their actions in shiny colors, as if it's great.

PM 2Ring

9

Q: Physics.SE remains a site by humans, for humans

We, the moderators of physics.SE, are deeply concerned about the recently announced "network policy" regarding computer-generated content on the SE network, which effectively attempts to forbid moderators from issuing suspensions or deleting posts for the undisclosed usage of computer-generated c...

discussion featured questions answers policy

Shadow The Spring Wizard

20:00

kudos

M.A.R.

20:50

I remain a human run by a human brain and three human kidneys

Tinkeringbell

Congrats!

Kevin B

🚽

Starship is go for launch

21:09

The new policies are getting insane, I think it is time to take action. I urge everyone to strike for one day as a warning to SE. They think they dont need us. Lets show them they do

Kevin B

i mean

Andreas detests censorship

@Starshipisgoforlaunch You're welcome to join in on the shortly upcoming strike. It will last for longer than a day; that's for sure. Await the Meta post.

Starship is go for launch

when is it and where can I see it @Andreasdetestscensorship

ill absouletly be in

how long will it be for

Andreas detests censorship

in The Meta Room on Stack Overflow Chat, 3 hours ago, by Zoe stands with Ukraine

Anyway, strike announcement draft discussions are ongoing on Discord if y'all want to contribute to that. Or leave a comment directly on GH

@Starshipisgoforlaunch Until the issues have been resolved.

Starship is go for launch

when does it start

Andreas detests censorship

21:22

Don't know.

Starship is go for launch

kk but that a bit important

and why does it still let posting

posts give views views give money after all

se clearly only wants money

Andreas detests censorship

The community needs time to prepare.

Tinkeringbell

@Starshipisgoforlaunch Nope. As much as I hate what's happening, I'm not turning the sites I mod into a free for all for a day.

Starship is go for launch

doesnt answer by last 3 sentences

Kevin B

wdym

we can't stop general users from posting

Starship is go for launch

21:29

SE only cares about money. They will only listen if they get less money @Tinkeringbell

We can't stop them, we can ask them not to

Tinkeringbell

The problem with planned and announced strikes is that they are just as easily abused as the new AI policy

@Starshipisgoforlaunch They won't lose money if I strike for a day, all that happens will be that the strike will be abused to post more crap than usual

Starship is go for launch

Okay then we should make it longer

Tinkeringbell

And don't announce it

Starship is go for launch

dont announce it? then how will everyone know to strike on that day

Tinkeringbell

I dunno, but announcing it is basically telling everyone that that day (and any time after) is basically free to abuse the site.

Surely you can organize somewhere secret.

Starship is go for launch

21:38

if everyone is free to abuse the site that shows SE that it needs us, not that i want spam but if ppl will still post then that it what it shows

also am i now allowed to mass post gpt answers? do i even need attribution anymore?

Kevin B

if you're following the guidelines, you'd quote the GPT content and cite it. but... it's unclear how moderators should handle that given the new policy.

and, if you don't cite it, there's literlaly nothing mods can do about it, given the new policy. so it's probably pretty obvious which route is less likely to result in action, until the policy changes.

Andreas detests censorship

Organizing it in secret is not effective, and does not convey the reason for the strike. Yes, it's in an invitation for abuse. That's the result. And that's the damage SE will have to accept for their harm against the community.

M.A.R.

21:59

Starship is go for launch

22:09

Completly agree @Andreasdetestscensorship

Andreas detests censorship

@Tinkeringbell That's stupid. It must be announced.

Starship is go for launch

Thank you smokre detector that is in fact spam

Transcript for