Tavern on the Meta

12:03 AM

@Andreasdetestscensorship hey, some of us have been in this situation before, and some of us will go through this again!

Kevin B

^

starball

@SonictheAnonymousHedgehog what are your thoughts on the points on biases?

Sonic the Anonymous Hedgehog

@starball Haven't thought about those yet much.

starball

@SonictheAnonymousHedgehog cannot view :( blocked

cocomac

@starball Does this version work? web.archive.org/web/20220809120225/https://daily49er.com/news/… (Wayback Machine save of that page)

starball

12:12 AM

@cocomac yep! I'm busy now but I'll read later.

Kevin B

been there done that, haha

fortunately i was able to resolve it before anything permanent occured

Andreas detests censorship

@MetaAndrewT. I was gone for 2 years because of the previous huge shitshow...

Journeyman Geek

@starball I will review. I wrote this on my little tablet on my cousin's kitchen table since I am travelling and it probably badly needs a few copyediting passes

Andreas detests censorship

@SonictheAnonymousHedgehog That is not about SO.

cocomac

12:28 AM

If an answer I post that is over +2/-0 is deleted, do I lose the reputation...?

Anton Menshov

@cocomac most likely yes.

> Deleting and undeleting posts may reverse reputation effects as well, if these posts have votes. Actions previously taken on deleted posts cease to affect reputation within five minutes (source), unless the post meets both the following criteria (in which case the reputation effects will be permanent) (source):

The post had a score of at least +3
The post was visible (i.e. not deleted) on the site at a moment at least 60 days after it was originally created
This includes reputation earned from approved suggested edits on deleted posts.

see faq on reputation

so, if it is a recent answer and/or it does not receive one more upvote, the reputation should be recalculated.

starball

1:32 AM

@cocomac "the codes" *shrivels up

cocomac

"Then four coding experts examine" uh... how about "professional software developers"? Also, "In his letter, professor [...] also emphasized the accuracy of the A.I. system and said there was only a 5% chance of error." except in a college/university class of hundreds, 5% is way too high IMO (FWIW, I'm against the SE ban, but still)

starball

@SonictheAnonymousHedgehog also, this situation seems to have much more to do with collaboration policies and academic honesty than it has to do with AI-generated-content...

starball

2:06 AM

is it acceptable on MSE to flag ChatGPT content that doesn't use blockquotes and doesn't state that it's generated by ChatGPT?

somebody has been posting such content as an ironic joke.

Anton Menshov

@starball so, why flag it then?

Kevin B

@starball i mean, yes, according to the new policy.

Journeyman Geek

@starball tptb says we can't validate such flags

cocomac

2:41 AM

@starball For ChatGPT, maybe not. If they're intentionally trolling and you have evidence beyond ChatGPT-usage, then that is different

starball

@JourneymanGeek who/what’s tptb?

@AntonMenshov why flag what? I haven’t flagged anything

cocomac

@starball FYI: Discussion: Network policy regarding AI Generated content by Ryan M

Never mind, you left a comment on that post that I missed

Chris - Regenerate Response

@starball Nothing would make me happier than for it to be flagged and removed, for what it's worth.

cocomac

Side note: can individual users vote to delete answers if they believe they are ChatGPT-generated as an alternative to mod-flagging? I don't have vote-to-delete privs (outside of the queues), but I wonder if that is permitted. I'll also likely raise that question on MSE

Chris - Regenerate Response

@cocomac It's not (yet) explicitly forbidden. I suppose red flags are good too.

Andreas detests censorship

2:55 AM

This is not a realistic counter-measure. We're not that many curators. 20 people posting 5 AI answers per hour amounts to 100 answers per hour. Half a day, that's 1200 answers. We can't handle that by voting to delete every answer of theirs. They need to be suspended.

Kevin B

the number of such users with the ability to do so are also quite limited and shrinking.

but yes, we can still use the tools we have, though if the post isn't against any existing policies, you're opening yourself up to being actioned against for abusing your privileges.

Chris - Regenerate Response

@Andreasdetestscensorship Red flags, downvotes, and deletions will all contribute towards an automatic question/answer ban.

Which effectively amounts to a 6-month suspension.

Andreas detests censorship

Or, we can all just show SE our opinion of them, and stop moderating until they stop abusing us.

Chris - Regenerate Response

(Nobody point out to SE that the automatic question/answer ban is probably biased a lot worse than ChatGPT detection).

@Andreasdetestscensorship Or just ignore this policy and keep moderating how we like.

Andreas detests censorship

They take us for granted, and think it's fine to dump a pile of shit on us.

2

Chris - Regenerate Response

3:06 AM

Since they don't appear to have learned after the mass resignations a few years ago.

Andreas detests censorship

They learned that we're lazy employees that don't wanna do what we're supposed to do

Eh, bad joke.

Chris - Regenerate Response

@Andreasdetestscensorship I'm cool with them thinking I'm a lazy employee. As long as I get my back pay :p And preferably a hat lol

Andreas detests censorship

All I'm asking for, is a dollar for each helpful flag I made...

Journeyman Geek

3:37 AM

@starball the powers that be

@cocomac I mean, if they didn't day so...

do it until they say not to

@Andreasdetestscensorship which in turn poisons any LLM that scrapes us... win?

cocomac

I am refraining from flagging as such, but I find violating established guidelines for releasing policies and blatantly violating Meta consensus, and then not posting a separate question for discussion until mods do it themselves to be rather Rude and "inappropriate for respectful discourse"

Andreas detests censorship

@JourneymanGeek Nobody wins by our community getting torn apart, and the repository we built, destroyed.

@cocomac I would certainly accept such a flag, but I'm not a moderator, only a curator now slammed with poop that fell from somewhere higher up.

cocomac

True. Even if the mods Disputed/marked Helpful such a flag, staff would just un-delete the answer, so it isn't really productive to do, even if the thought of red-flagging that answer is somewhat amusing. It also risks suspensions for anyone involved

Journeyman Geek

4:15 AM

@Andreasdetestscensorship I was being sarcastic

tripleee

@SonictheAnonymousHedgehog there is a big difference between "don't suspend violators (lightly)" and "basically don't handle suspected AI-generated content at all". The "hold back on suspensions" part I can understand

Journeyman Geek

@tripleee Its something that could break a site, or cause potential harm (on one of my sites)

cocomac

Is there a practical difference? Other than suspensions, what options do mods have to stop users that post large quantities of ChatGPT-generated answers quickly? Deleting them isn't sustainable if large numbers of users are posting them and flags will be declined. I guess account nukes, but I'd assume if suspensions are banned, so are those most likely

(Not trying to argue, but I'm a bit confused as to the practical difference there)

Kevin B

i mean

given "suspected as being gpt created" isn't a valid reason to take action,

Journeyman Geek

@cocomac delete them as spam or nonsense

literally the 'best' option is 'worse'

tripleee

4:23 AM

@JourneymanGeek say more; you have a site where only suspending users on repeated violations is not sustainable?

cocomac

@KevinB No, of course not. But "detector says 99.68% fake and it contains certain phrases and they're suddenly posting tons of detailed answers quickly out of the blue" is a more convincing reason

Kevin B

that's still not meeting the new high bar

> This standard would exclude most suspensions issued to date.

cocomac

That's a significant part of why I'm against that policy...

Kevin B

which ones wouldn't it exclude?

"I copy pasted this from ChatGPT please let me know if it works"

Journeyman Geek

@tripleee Well - most suspensions start with someone pointing out a post looks chatgpt generated and adding the checker as 'confirmation'

Kevin B

4:26 AM

i never use the checker

never have

they're just so obvious

Journeyman Geek

so if people custom flagged as nonsense, or redflagged them...

starball

@cocomac see spev's answer and those like it.

cocomac

@KevinB I guess a ChatGPT answer with a spam link? And possibly cases where the user admitted to it themself (i.e., "As an AI chatbot, I cannot [...]")

Andreas detests censorship

@JourneymanGeek Doesn’t the new CoC ban sarcasm?

Kevin B

not that broadly

sarcasm isn't even in the new document

Andreas detests censorship

4:35 AM

Then I must’ve interpreted an answer asking for the new CoC to ban sarcasm, as having been approved.

Kevin B

that's my interpretation too

but "sarcasm" shows 0 results, so, maybe they're using some creative wording somewhere.

Andreas detests censorship

That’s the kind of stuff that just makes me steer away from the CoC, and simply just live by «be nice». It makes life easier.

starball

sarcasm isn't banned in the current CoC- just discouraged.

Andreas detests censorship

I identify as sarcasm. Am I discouraged?

Kevin B

no

starball

4:37 AM

and last I checked, the new draft doesn't have specific wording on sarcasm (meta.stackexchange.com/a/388878/997587)

cocomac

Notably, that CoC draft bans "Inauthentic usage - We do not allow any use of the system that ... causes unnecessary and unwanted disruption and/or harm to users and/or the network." Specifically, "Plagiarizing or copying content from websites ... or other online ... tools without proper attribution in a manner that violates our referencing standards."

Kevin B

generally the coc can be ignored if you're interacting with the site and people who are here with respect

as generally that'll lead to you not breaking the coc

starball

@cocomac noice

Andreas detests censorship

I don’t think I’ve read the CoC in a few years. I had a short look at it long ago around the time the pronouns debate was raging.

I have been accused once or twice for breaking the CoC. I think the last time, was in SOCVR, when somebody called me out for using «he», «she», «him» and «her» to refer to users with a clear gender profile. Of course, I wasn’t breaking the CoC.

Kevin B

"clear"

Andreas detests censorship

4:41 AM

@KevinB A male name and picture of a masculine man with a beard, qualifies as «clearly man» for me.

Kevin B

that's not my point

Andreas detests censorship

Then I missed it.

Kevin B

there was no mandate that we use neutral language, it's just preferred. therefore if there is no otherwise clear indication of what is preferred... you haven't messed up yet. it's when it is made clear, via a message or pronouns indicated in the users name, and then ignored that you've messed up. but you most certainly shouldn't rule lawyer around it, causing a disruption would most certainly lower the threshold of whether or not you've crossed a line

cocomac

@starball Done (commented on): meta.stackexchange.com/a/389624

Kevin B

also... wasn't discussing this portion of the coc also made against the coc, at one point

starball

4:48 AM

sadly,
> Feedback cutoff date
> We will be processing feedback given here by May 24th

Andreas detests censorship

@KevinB Not sure if you were around that time, but it filled up the transcript. I stood my ground, and will continue to do so; I decide how to use my own language, and if I’m not allowed to, I decide to leave. But yes, I prefer to show respect towards everyone. But I’m not going to avoid gendered pronouns when I am strongly sure of the gender. The other person just required me to always use «they» and «them». I already use that as a default.

Kevin B

yup

Andreas detests censorship

@KevinB I think so. That’s something I can’t respect.

Kevin B

at the end of the day, a rule that isn't enforced isn't a rule

Andreas detests censorship

That’s not correct.

Kevin B

4:50 AM

(it's a trap)

cocomac

I joined SE well after 2019 - are you all referring to this and the mods that resigned then? (Not giving a viewpoint there, just wondering for context)

Kevin B

yes

starball

@cocomac random question- is there a meaning / story behind your username?

Andreas detests censorship

@cocomac Yes, but I was also talking about a specific event in SOCVR recently. It’s not really something worth bringing up, and is best left buried, but I thought it made sense to mention it now.

Kevin B

i mean, there's also a way to respond to such events that can avoid conflict, without giving up any ground. I tend to prefer that route

cocomac

4:54 AM

@starball Mine? It's somewhat silly, but since you asked... it's short for coconutmacaroon, which is a food that, when I originally came up with the full form of that username, I enjoyed (I've had the full username as inspiration for others since many years prior to joining SE)

starball

@cocomac cool! mine is "star" + "tarball" I don't even know anything about the technical details of archiving / compression, but I like the ring it has :)

cocomac

@starball Thanks, yours sounds nice!

starball

5:12 AM

!!/tea nicael

SmokeDetector

@starball brews a cup of Darjeeling tea for @nicael

nicael

I actually love these new voting buttons, they remind me of ones SE had in around 2014-2015

You can see on the screenshot above what voting buttons used to look on SE and meta (however on Stack Overflow they should probably use older design)

starball

I like the more obvious hitbox visualization. easier to downvote non-answers and questions making improper use of images (when I don't use keyboard)

nicael

hitbox XD

tripleee

5:26 AM

@cocomac there are two distinct strands in those events from 2019; the new CoC was what triggered it all, but their handling of a "violation" by Monica was what caused things to really go ballistic

rene

6:04 AM

→ 18 messages moved to Chimney

Journeyman Geek

7:07 AM

@Andreasdetestscensorship so flag me

😁

@tripleee essentially unpopular changes go better when you have trust and don't alienate big chunks of the active community

2

And frankly we're a lot more invested than staff are in some aspects

Example person

7:42 AM

@Starball, Can it really answer your 'complex' questions?

starball

@Exampleperson not that I know of. I'm just talking to the void :P

/ creating useless comments on the metasmoke entry.

Meta Andrew T.

8:11 AM

@starball the question was bumped by this answer. Every bump will re-trigger a scan to the question and all of its answers

starball

@MetaAndrewT. TIL. thanks!

Shadow The Spring Wizard

12:15 PM

sniff

@JourneymanGeek here ya go, I've flagged you!

SPArcheon

12:34 PM

@JourneymanGeek I asked for coherence, so if Smokey gets shot, it is my fault

Andreas detests censorship

12:50 PM

@JourneymanGeek Will do!

Journeyman Geek

@ShadowTheSpringWizard Other than that being a corgi... looks like you did

Andreas detests censorship

I actually did flag one of Shog’s answers for R/A just to make a point, a few years ago… Flag was accepted, too…

Journeyman Geek

More seriously

Andreas detests censorship

But don’t worry. Not gonna flag you. ;)

Journeyman Geek

if I'm bring an idiot and being hurtful with my sarcasm, as opposed to actually being funny

tell me. Or flagging is actually fine too

I'm entirely aware that I can occationally get carried away and cross the line

Andreas detests censorship

12:55 PM

It’s better to just tell you. I rarely flag my interactions with others.

Journeyman Geek

whatever you're comfortable with

I do take feedback

Andreas detests censorship

So, that time I was accused of breaking the CoC in SOCVR, that guy which was annoyed at me for using gendered pronouns, also repeatedly accused me for thinking with my dick. That is quite rude, but why flag it? Why increase the conflict? It doesn’t help anybody.

SPArcheon

mumble... you know what?

Andreas detests censorship

But you weren’t crossing the line. I was making a sarcastic reply. ;( :P I love sarcasm! @JourneymanGeek

I am sarcasm!

@SPArcheon Hi, pony. :)

SPArcheon

I think this blunt line can be useful now... maybe someone will rethink twice before making my "prophecy" true. You know, it would be simply "too" evident.

Journeyman Geek

1:00 PM

@Andreasdetestscensorship cause a lot of the time direct conflict isn't useful

SPArcheon

> I seriously hope we don't get a new "Monica" for someone mod opposing this nice policy.

Journeyman Geek

That said every so often, it is, but needs a lot of thinking and direction

Andreas detests censorship

@SPArcheon Somebody mentioned yesterday that the SD team is already onboard about taking it down after these events. SE needs SD, so they’re not gonna want it gone.

SPArcheon

@Andreasdetestscensorship read the linked post, you should get why that request is actually 50% sarcasm, 55% mocking them.

Andreas detests censorship

@JourneymanGeek Some things are better handled by cooling off, and just accept that some mean things were said in the heat, and get over with it.

@SPArcheon Oh, I did. Already read it all. Gave it an upvote too.

Journeyman Geek

1:02 PM

@Andreasdetestscensorship Sure, but that's not always what people do

Andreas detests censorship

@JourneymanGeek Then you can flag at that point. Give people a chance to fix it themselves first.

Journeyman Geek

@Andreasdetestscensorship and we encourage people to flag over direct confrontation

always have

SPArcheon

After all like I said Smokey "is heuristic detection of bad content so that actual humans can avoid looking for themselves and have a pre-processed list of posts to inspect" so... it fells kinda similar to something we just banned if my understanding of the events is right.

Andreas detests censorship

Yeah, your answer gets the point across quite clearly.

@JourneymanGeek It seems like we mostly agree here.

Journeyman Geek

......

SPArcheon

1:05 PM

@Andreasdetestscensorship oh, motosubatsu also gets the same point, I just made a far more blunt and satiric version.

Journeyman Geek

Yanno, I just realised part of the disconnect is...

as an old school moderator, I've never considered most suspensions punitative

they're either supposed to, to borrow a Taylor Swift Lyric "Sit in the corner and think what you did" or protect the community from someone who is disruptive

SPArcheon

@JourneymanGeek you clearly never helped as a mod on an online game server.

Journeyman Geek

TPTB are expecting a certain higher standard cause they see it as punitative

SPArcheon

I don't really agree.

the main point Imho is exactly what motosubatsu said

> I strongly feel this mischaracterizes both what the mods were doing in handling GPT generated content by implying that we were going round suspending at the drop of a hat based on shallow guesses or blindly taking detectors as gospel and simultaneously implying that mods were somehow moderating this content without SE's awareness of what was being done and how it was being done. Neither of these are true

They imply that moderators are using tools to fully automate actions and do not use their brain when reviewing the detectors results.

IF that is true then the problem is the minority using the tools that way.

It is not different from the old "robo-reviewer" and should be handled the same way.

But the policy gives once again the idea that they see some bad apples (assuming they really saw them in the first place) and then call the whole tree a big bag of trash

Andreas detests censorship

@SPArcheon They are not going to listen. They ignore everything anybody says. They're screaming while covering their ears and eyes.

SPArcheon

1:23 PM

@Andreasdetestscensorship oh, I don't expect anything to change from anything that will be posted here or in the multiple posts on meta.

Shadow The Spring Wizard

@JourneymanGeek :D

SPArcheon

But somehow I still fell some sort of pleasure from pointing out the evident inconsistencies in their claims.

It is not a secret that I quite liked Socrates Jones and would like a full game with that premise.

Andreas detests censorship

@SPArcheon It'll lead to one thing: burnouts for all of us.

SPArcheon

@Andreasdetestscensorship ? I beg to dissent.

I am on burnout from SE since Monica at lest, probably a little earlier than that.

Andreas detests censorship

@SPArcheon Please seek medical assistance.

SPArcheon

1:28 PM

@Andreasdetestscensorship the fact I am still here does not mean that I "recovered", "forgive", "forgot" what happened or think this is an happy friendly place.

I am here to chat with a few selected users and to post an answer every now and then

and since the system is very broken too and the "gamification" Jeff planned doesn't really work outside SO I don't even pursue technical, actual work to post answers any more, just random things about anime or games that take no time and I already know.

Andreas detests censorship

I actually did contact a psychologist...

Not that it worked, but that's another story.

Just everything came down at the same time; corona, leaving SO in anger, life changes, etc, etc.

There's this post on MSO, posted by a former staff member, that says the SE employees can't sleep at night, and are given nightmares and depression from visiting Meta. Well, I can ensure you that the way SE treats the community, hurts the community members too!

SPArcheon

@Andreasdetestscensorship you can join my request to give Meta a yellow background with a neon light title bar and a constant buzzing sound

Mithical

@Andreasdetestscensorship I literally didn't sleep well last night because my phone kept buzzing with notifications and waking me up :P

Tinkeringbell

@Andreasdetestscensorship I hope to know my own boundaries well enough to avoid that, and be gone long before I get medical problems due to the internet.

Because that'd be a really, really lousy thing of me to do to me.

SPArcheon

1:44 PM

@Tinkeringbell you forgot "that it was in my power to prevent". Internet is fully able to give you problems without you needing to do anything at all.

Tinkeringbell

@SPArcheon Anything is 'in my power to prevent' though. As the devices with internet can be shut down at any time.

Mithical

Yeah, not when they start SWATting you or whatnot.

SPArcheon

@Tinkeringbell I was more thinking of "idiot you know uses YOUR PI to register on social media and impersonates you" etc.

Or "bank saved your credentials in plain text, only disclosed the security breach months after it happens, in the meantime YOU got the pleasure and waste of time to work with Amazon to handle some frauds someone tried to made against you because of this"

Tinkeringbell

@Mithical That's also preventable by being conscious of what you post on the internet yourself.

Same as the first thing @SPA mentions.

Journeyman Geek

→ 19 messages moved to Chimney

Andreas detests censorship

1:51 PM

@Tinkeringbell "The internet". It doesn't matter that this is on the internet. You've invested lots of time and effort in the project. Consider it similar to working every single day at the local library/archive, and expanding it, and pouring your soul, etc, into it...

SPArcheon

@Tinkeringbell not really when for example your government tries to make obligatory to use "traceable electronic money" for transactions if you hope to deduce them from taxes.

Tinkeringbell

@SPArcheon Okay, you win. There are serious edge cases out there where other idiots than me can cause problems due to the internet.

@Andreasdetestscensorship Same thing then, but for real life too. I hope to know my own boundaries well enough to not let me do stupid things to me.

SPArcheon

@Tinkeringbell Bingo. All it takes is some clueless idiot that has the power to force internet on you.

.... at least no one though of requiring a facebook account for voting. YET

Tinkeringbell

@SPArcheon Still I doubt either of those two examples you just mentioned would on their own trigger a medical condition like burnout.

SPArcheon

@Tinkeringbell yep, I was just using Andreas words there.

Andreas detests censorship

1:55 PM

I'm being used!

Tinkeringbell

Get used to it.

SPArcheon

also,mostly thinking of a "online game burnout" thing, as in "I am bored with this game but I also don't want to delete my account since I spent so much time on it"

Andreas detests censorship

@Tinkeringbell You sound like SE.

Journeyman Geek

@Andreasdetestscensorship but are abused?

Tinkeringbell

@Andreasdetestscensorship It rubs off after a while. ;)

Journeyman Geek

1:56 PM

or maybe confused?

Tinkeringbell

Crap I need to run. See ya!

Andreas detests censorship

@Tinkeringbell Don't run the crap!

SPArcheon

2:19 PM

@JourneymanGeek Would I be evil for asking to NOT blur official staff post that get too much downvote love?

something like "since apparently this is going to be quite standard from now on, can we avoid blurring staff posts that gets too much downvotes? They are hard to read after the users express their disagreement, but since that is ignored it is important that the post stays readable"

Shadow The Spring Wizard

@Andreasdetestscensorship sock?

@Tinkeringbell run due to the crap? Been there done that. Good luck! :P

SPArcheon

2:37 PM

Shadow The Spring Wizard

yay! That was quick. :D

Good to be back.

SPArcheon

@ShadowTheSpringWizard going for the Alf layla wa layla pings now?

or am I too late?

Shadow The Spring Wizard

?

@SPArcheon you mean this?

One Thousand and One Nights

One Thousand and One Nights (Arabic: أَلْفُ لَيْلَةٍ وَلَيْلَةٌ, ʾAlf Laylah wa-Laylah) is a collection of Middle Eastern folk tales compiled in Arabic during the Islamic Golden Age. It is often known in English as the Arabian Nights, from the first English-language edition (c. 1706–1721), which rendered the title as The Arabian Nights' Entertainment.The work was collected over many centuries by various authors, translators, and scholars across West Asia, Central Asia, South Asia, and North Africa. Some tales trace their roots back to ancient and medieval Arabic, Sanskrit, Persian, and Mesopotamian...

SPArcheon

yep, I meant "one thousand and one pings"

Shadow The Spring Wizard

What you wrote is the Hebrew version, with tons of grammar errors.

Should be "Elef Layla Ve Layla", or just "אלף לילה ולילה". :D

Alf is a cute alien.

Alf ^

Elef is thousand. ;)

SPArcheon

2:42 PM

some claim that is the persian translation

Shadow The Spring Wizard

@SPArcheon nah, I get more pings in Den, already celebrated 1000, yesterday. :P

(not overall, those would be reaching 100k soon)

SPArcheon

for example

Tausendundeine Nacht

Tausendundeine Nacht (persisch هزار و يک شب, DMG hazār-u yak šab, arabisch ألف ليلة وليلة, DMG alf laila wa-laila) ist eine Sammlung morgenländischer Erzählungen und zugleich ein Klassiker der Weltliteratur. Typologisch handelt es sich um eine Rahmenerzählung mit Schachtelgeschichten. == Geschichte == === Vermutete indische Ursprünge === Aus Sicht der frühesten arabischen Leser hatte das Werk den Reiz der Exotik, es stammt für sie aus einem mythischen „Orient“. Das Strukturprinzip der Rahmengeschichte sowie einige der enthaltenen Tierfabeln weisen auf einen indischen Ursprung hin und st...

here it mentions the "alf laila wa-laila" version too.

Alas, doesn't really matter, you got what I meant

Shadow The Spring Wizard

@SPArcheon ohh.... that's in Arab. I know little bit, but mostly Hebrew.

Should have guessed from the "wa", it's the Arab word for "and".

(in Hebrew it's "ve")

Didn't know about Alf, can count only to ten in Arabic. :D

Andreas detests censorship

3:28 PM

@ShadowTheSpringWizard Yes, please wash my socks.

Tinkeringbell

@ShadowTheSpringWizard Nah.. It's crap that I had to run, because I'm an unfit fatass :P

SPArcheon

3:43 PM

> In addition, the day that the announcement was posted for moderator “head’s up” was a holiday in the United States, Canada, the UK, and other European countries. Given that it was announced on a holiday and was then posted a day later to the public, it is highly likely that not all moderators were able to see and react to the policy edict before it was made public

^ can someone confirm if this line is true?

Was the "head up" really given on an holiday day?

Tinkeringbell

Yep

The Monday after Pentecost is a holiday off in most European countries and the US had Memorial day

balpha

Monday was a holiday in a lot of places, so if we're talking about Monday (two days ago) then yes

SPArcheon

.... wasn't the communication of Monica removal given to her on a holiday too?

Tinkeringbell

@SPArcheon No, day before/hours before

balpha

@Tinkeringbell The UK just calls it "Spring bank holiday". Lazy naming.

SPArcheon

3:46 PM

@Tinkeringbell so, kinda same. DOH.

There's that old saying "don't update the servers on Friday afternoon"

Andreas detests censorship

@Tinkeringbell Haha. I went to the grocery store on Monday, and asked why only the small part of the shop was open, and not the big part. I was told it’s a holiday. Woopsie.

Tinkeringbell

@balpha Dutch are like hobbits. We call it second Pentecost day :P

Andreas detests censorship

Pinse, here.

Tinkeringbell

We also have second Easter day and second Christmas day :)

Andreas detests censorship

Same names here.

Anton Menshov

3:51 PM

@SPArcheon unless you want to cause mayhem.

Kevin B

or get in some overtime

Anton Menshov

so there are many strategic reasons to update servers specifically on Friday afternoon.

Kevin B

also varies by industry

Feeds

4:16 PM

CEO Update: Paving the road forward with AI and community at the center

Prashanth Chandrasekar on May 31, 2023

With all the significant changes in the industry, one thing has remained the same: companies are committed to driving productivity and efficiency throughout their organizations, and we continue to help our customers and community deliver both.

Journeyman Geek

I'm busy tommorrow but that seriously needs questioning

> we continue to help our customers and community deliver both.

cocomac

Notable quotes include: "I remain grateful for our community" and "When I think about Stack Overflow’s future [...] that’s largely based around the work that we’re doing to incorporate GenAI into our products. " and "We believe GenAI can be a similar competitive advantage for Stack Overflow. We have the [...] community trust to responsibly deploy generative AI to improve our existing suite of products"

Journeyman Geek

.... I am doubttful

cocomac

"We have the [..] community trust to responsibly deploy [...] AI" - Citation Needed

"We want to be able to continue to invest back into our community [...] LLMs are trained off Stack Overflow data, which our [...] community has contributed to for nearly 15 years. We should be compensated for that data so we can continue to invest back in our community." - Yes, yes. You should be compensated for our content so you can invest into our community by having "Approximately 10% of our company [...] working on features [...] leveraging GenAI" /s

SPArcheon

@cocomac nope, you meant "so you can invest into firing the people we liked"

See: shog9 for example who we lost in the past. Maybe he could have found the easy way to put this policy in a way that actually makes sense since the core is something he actually discusses years ago.

Remember when Jeff said that the day we get spammers to post actually useful well written content to escape detection we won?

Well, same for AI generated stuff.

Problem is that they didn't got there yet

So, if the policy was something like "Please don't ban people because the content is generated, please ban them if the content is bad or wrong" I think no one would have been upset.

Andreas detests censorship

4:30 PM

@cocomac What a bunch of stupid shit, and outright, direct lies. They do not have our trust!

cocomac

Yes. I'm writing a meta post on that now...

Also noteworthy that they didn't post that Blog to MSE

SPArcheon

@cocomac coincidences, just unfortunate coincidences.

48 mins ago, by SPArcheon

> In addition, the day that the announcement was posted for moderator “head’s up” was a holiday in the United States, Canada, the UK, and other European countries. Given that it was announced on a holiday and was then posted a day later to the public, it is highly likely that not all moderators were able to see and react to the policy edict before it was made public

Journeyman Geek

@cocomac Yet

If they don't someone will

SPArcheon

@JourneymanGeek well, if they post the thing after someone already mentioned it.... "the worms are out of the can, you may as well pretend you already mean to tell us", right?

Journeyman Geek

Or someone else can :D

cocomac

4:35 PM

I got tired of waiting: meta.stackexchange.com/questions/389653/… :)

Andreas detests censorship

My response on the blog post (which obviously will get declined):

> You’re a liar. You do not have our (the community’s) trust!

cocomac

@JourneymanGeek If staff does post one, my question can be merged into the official one as I posted my feedback an a self-answer

I think I hit the daily rep cap

SPArcheon

@cocomac as I said in another room....

> on the plus side, this may finally give me "Trusted user" privileges on Meta.
That said, it is kinda weird to become a trusted user thanks to Winter Bash during December and SE doing horrible things the rest of the year

cocomac

Good luck on getting to 20k! I'm currently hoping to make it to 10k, but that's still pretty far for me (at ~3.6k) now

SPArcheon

BTW, I smile every time someone talks about "model be biased" without actually knowing what that means or happens.

Andreas detests censorship

5:16 PM

@cocomac Just write a few angry posts full of emotions, targeted at SE...

SPArcheon

> ꓲ јսѕt rеаꓲіzеd ꓲ соսꓲd ѕtаrt роѕtіոց mу mеѕѕаցеѕ ꓲіkе tһіѕ аոd ոо оոе ԝоսꓲd ոоtісе ԝһаt іѕ ԝrоոց. ꓮոd ꓲ bеt tһаt tһіѕ ԝоսꓲd rеаꓲꓲу mеѕѕ սр аոу mоdеꓲ "ѕоmеоոе" trіеѕ tо trаіո սѕіոց tһе соոtеոt ꓲ рrоνіdеd.

2

Andreas detests censorship

That is painful.

SPArcheon

just add some zero width spaces in the mix for added "fun"

@Andreasdetestscensorship you notice weird characters?

Here on firefox the only noticeable difference is the "I"

Andreas detests censorship

@SPArcheon Yes. Safari. Everything goes up and down.

SPArcheon

You said "Safari". I stopped there :P

Andreas detests censorship

5:26 PM

u, n and g are shorter than the rest.

@SPArcheon If you're Brave, you may join me on my Safari. I shall put on some Vivaldi for us to enjoy while we spectate the Foxes as they catch Fire.

Yeah, that was bad. Those posted in SOCVR some weeks ago, were better.

SPArcheon

𝔖𝔥𝔬𝔲𝔩𝔡 𝔱𝔥𝔯𝔬𝔴 𝔦𝔫 𝔰𝔬𝔪𝔢 𝔱𝔥𝔢𝔵𝔱 𝔲𝔰𝔦𝔫𝔤 𝔍𝔬𝔲𝔯𝔫𝔢𝔶𝔪𝔞𝔫 𝔣𝔞𝔳𝔬𝔯𝔦𝔱𝔢 𝔣𝔬𝔫𝔱 𝔱𝔬𝔬. 𝔚𝔥𝔬 𝔨𝔫𝔬𝔴𝔰, 𝔪𝔞𝔶𝔟𝔢 𝕮𝔥𝔞𝔱𝔊𝔓𝔗 𝔴𝔦𝔩𝔩 𝔱𝔥𝔦𝔫𝔨 𝔱𝔥𝔦𝔰 𝔦𝔰 𝔠𝔬𝔬𝔩 𝔞𝔫𝔡 𝔰𝔱𝔞𝔯𝔱 𝔭𝔬𝔰𝔱𝔦𝔫𝔤 𝔤𝔢𝔫𝔢𝔯𝔞𝔱𝔢𝔡 𝔠𝔬𝔫𝔱𝔢𝔫𝔱 𝔴𝔦𝔱𝔥 𝔯𝔞𝔫𝔡𝔬𝔪 𝔉𝔯𝔞𝔨𝔱𝔲𝔯 𝔴𝔬𝔯𝔡𝔰

^ @JourneymanGeek

Andreas detests censorship

No....

Stop the torture.

SPArcheon

@Andreasdetestscensorship Foxes?

Firefox symbol is a pony, not a fox.

Andreas detests censorship

@SPArcheon Do you see ponies everywhere?

SPArcheon

no, that is true. Firefox almost became FirePONY officially.

They actually did the logo and all, then stopped due to copyright.

this is not a joke

bugzilla.mozilla.org/show_bug.cgi?id=776053

Andreas detests censorship

5:32 PM

@SPArcheon Stop scaring me. My computer froze when I clicked that.

SPArcheon

scared of cute stuff?

Andreas detests censorship

Ueh

SPArcheon

> “When I was ten, I read fairy tales in secret and would have been ashamed if I had been found doing so. Now that I am fifty, I read them openly. When I became a man I put away childish things, including the fear of childishness and the desire to be very grown up.”
― C.S. Lewis, On Stories: And Other Essays on Literature

Andreas detests censorship

Well, let's all celebrate today, for the rise of MSE; the only site were reputation will continue to be gained only from posting human-written answers, while on the other sides, reputation gained from AI-generated content will eventually outperform the reputation gain from human content.

But clearly, if AI takes over on the other sites, better just shut down MSE; no need for it anymore, and negative opinions about the company don't look good.

Tinkeringbell

5:51 PM

What makes you think MSE was spared and will continue to be spared from the AIpocalypse?

meta.stackexchange.com/a/385712/369802 there's been more since

SPArcheon

@Tinkeringbell well, you know, that got me wondering too.

I mean, who is so stupid to provide their "competitor" with resources that should enable them to do their work better?

Have you ever seen an artist selling their art to stable diffusion models so that they can make art in their place?

Have you see a musician helping someone to make a model to generate music without them?

And yet...

> Is SE [going to be] selling our content for AI model training? And what exactly does "reinvest back into our communities" mean?

Feeds

We were able to replicate and confirm prior authors' detection of a moon orbiting the Earth with high confidence.

SPArcheon

So, I AM actually starting to wonder if this is just the last desperate move to cash in on our content before shutting down the site

After all, reducing workforce, focusing on AI, don't giving a [redacted] about alienated users etc...

all things that go very well with "in a few month we will shut down, let's just try to cash as much as we can while it last"

Thomas Markov

6:15 PM

@SPArcheon A few days ago I would have said that was tinfoil hat level speculation, but now I’d put it in the realm of “feasible”.

SPArcheon

@ThomasMarkov oh, same too, that is just a random idea that came from the recent events.

I wouldn't call it highly probable but I wouldn't exclude that scenario either.

PM 2Ring

@SPArcheon "please ban them if the content is bad or wrong" But one of the main reasons for the SO blanket ban on ChatGPT content is that it usually looks superficially good, which makes it much harder to detect when it's bad unless you're a SME. So it takes more time, effort, and expertise to deal with than a lot of human-generated bad content.

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...

Shadow The Spring Wizard

@Andreasdetestscensorship pony pony pony

@Andreasdetestscensorship g is hard to find, so yeah, make sense it's being shorter. ;)

PM 2Ring

6:33 PM

@SPArcheon Pretty unlikely. But they are desperately trying to figure out how to retain the value of the site content in the face of the changes that AI are making to the landscape.

IMHO, people are misinterpreting what the CEO is trying to do. I think he wants to use AI to improve the quality of the content, eg by make it easier to search. He knows that AI isn't yet good enough to generate content, but it can certainly help to refine what we've got: fix typos, identify obsolete info, etc.

Unfortunately, he writes like a marketing bot, which makes it hard to be sure what he's saying, and it makes many of us distrustful of his statements, especially when he says stuff that appears to be blatant BS. ;)

Spevacus

@ShadowTheSpringWizard :)

cocomac

7:27 PM

Can private rooms be frozen by mods without triggering deletion_?

Laurel

@cocomac yes: "A room is considered worth retaining if it has more than 15 messages by at least 2 users." I have personally seen multiple private rooms freeze (albeit automatically) and the rooms were not deleted.

Andreas detests censorship

@PM2Ring What matters isn’t actually what the CEO writes in a blog post not even aimed at us. What matters, is what they do, and if you combine their actions, the content of all their recent interactions with us, and this latest blog post, you get a picture of their intentions, their values, and their goals.

@PM2Ring Then that’s what he gets. He doesn’t interact with us. He doesn’t participate in the community. He doesn’t try to clarify. He doesn’t try to meet us anywhere. He just sits on the side, and serves us garbage, and direct lies (we do not support the company in this, and we do not trust them).

PM 2Ring

7:56 PM

@Andreasdetestscensorship I get a picture, but it seems to have been painted in a parallel universe, not the one I live in.

Speaking of the dubious benefits of AI:

in Python on Stack Overflow Chat, 3 hours ago, by roganjosh

That moment when you find out that a lot of the code issues you've been fighting against come from GPT4. I just want to find a corner and darkness. This is utterly depressing

Andreas detests censorship

@PM2Ring I heard about a new project using ML to extract images from the brain; the results weren’t that accurate, but they were still near what the humans were seeing. That’s use of ML which I find really awesome. I often imagine things in my head (not seeing them with my eyes),

and fail to recreate them as a drawing or 3D model. If I could extract some image (doesn’t have to be too correct, or with too many details), I could easily get more peace with myself, and something concrete to work off from!

Andreas detests censorship

9:50 PM

@Tinkeringbell We're primarily seeking opinions on MSE (hopefully you get what I mean); so, the actual people using MSE would in no way want AI content, and it would serve little purpose here. Take SO, for instance; AI can actually generate answers there, that seem fine enough to get upvotes.