Register a SA Forums Account here!
JOINING THE SA FORUMS WILL REMOVE THIS BIG AD, THE ANNOYING UNDERLINED ADS, AND STUPID INTERSTITIAL ADS!!!

You can: log in, read the tech support FAQ, or request your lost password. This dumb message (and those ads) will appear on every screen until you register! Get rid of this crap by registering your own SA Forums Account and joining roughly 150,000 Goons, for the one-time price of $9.95! We charge money because it costs us money per month for bills, and since we don't believe in showing ads to our users, we try to make the money back through forum registrations.
How many quarters after Q1 2016 till Marissa Mayer is unemployed?
1 or fewer
2
4
Her job is guaranteed; what are you even talking about?
View Results
 
  • Post
  • Reply
Boris Galerkin
Dec 17, 2011

I don't understand why I can't harass people online. Seriously, somebody please explain why I shouldn't be allowed to stalk others on social media!

Mega Comrade posted:



Can't have people talking about Ukraine, might upset the people Elons trying to get to invest.

Without knowing anything about the bird code, all this suggests is that all those topics are essentially “nsfw” for people who have “safe mode” turned on. So ie you turn on safe mode and all those topics are hidden from you.

Which makes sense?

Adbot
ADBOT LOVES YOU

Nenonen
Oct 22, 2009

Mulla on aina kolkyt donaa taskussa

Boris Galerkin posted:

Without knowing anything about the bird code, all this suggests is that all those topics are essentially “nsfw” for people who have “safe mode” turned on. So ie you turn on safe mode and all those topics are hidden from you.

Which makes sense?

Does it make sense that one conflict is singled out? Or that information about Ukraine is likened to, say, hate speech or covid misinformation?

Boris Galerkin
Dec 17, 2011

I don't understand why I can't harass people online. Seriously, somebody please explain why I shouldn't be allowed to stalk others on social media!

Nenonen posted:

Does it make sense that one conflict is singled out? Or that information about Ukraine is likened to, say, hate speech or covid misinformation?

It makes sense that someone who has some version of "safe mode" turned on gets topics about the war in Ukraine blocked because, you know, it's a war. Also, I'm guessing it got singled out because there's a metric fuckton of tweets on the topic, so it's easy to categorize.

Believe it or not but a lot of people just want to follow funny people and not care about doom scrolling.

pumpinglemma
Apr 28, 2009

DD: Fondly regard abomination.

Boris Galerkin posted:

It makes sense that someone who has some version of "safe mode" turned on gets topics about the war in Ukraine blocked because, you know, it's a war. Also, I'm guessing it got singled out because there's a metric fuckton of tweets on the topic, so it's easy to categorize.

Believe it or not but a lot of people just want to follow funny people and not care about doom scrolling.
If that were the motivation you’d see tags for “war” or “politics” or “climate change” or “genocide” or literally any other topic that’s not outright lies or hate speech or copyright violation or porn.

Boris Galerkin
Dec 17, 2011

I don't understand why I can't harass people online. Seriously, somebody please explain why I shouldn't be allowed to stalk others on social media!
Cool, so do you volunteer to actually dig through the bird code and figure out what that list is for? Because I bet you the original twitter poster just did a ctrl-f "ukraine" and then went on and shouted about being silenced, after seeing other people do the same. Other people who also didn't bother to figure out what the code itself is actually doing. People like me, who also don't know what the code is doing, but is going off of variable names, like the original twitter poster.

Mister Facetious
Apr 21, 2007

I think I died and woke up in L.A.,
I don't know how I wound up in this place...

:canada:

Boris Galerkin posted:

Cool, so do you volunteer to actually dig through the bird code and figure out what that list is for?

How do you think this was found out? People are already doing it.

Boris Galerkin
Dec 17, 2011

I don't understand why I can't harass people online. Seriously, somebody please explain why I shouldn't be allowed to stalk others on social media!

Mister Facetious posted:

How do you think this was found out? People are already doing it.

I think Mr. Dog Poster above just did a ctrl-f for "ukraine" and tweeted about it to get internet points, after seeing other people with the same amount of understanding do the same.

Show me a tweet with someone who actually knows what the code is doing and I'll shut up about it.

(USER WAS PUT ON PROBATION FOR THIS POST)

pumpinglemma
Apr 28, 2009

DD: Fondly regard abomination.

You’re the one who can’t provide a plausible alternative explanation, why the gently caress should I do your legwork for you?

Nenonen
Oct 22, 2009

Mulla on aina kolkyt donaa taskussa

Boris Galerkin posted:

Cool, so do you volunteer to actually dig through the bird code and figure out what that list is for? Because I bet you the original twitter poster just did a ctrl-f "ukraine" and then went on and shouted about being silenced, after seeing other people do the same. Other people who also didn't bother to figure out what the code itself is actually doing. People like me, who also don't know what the code is doing, but is going off of variable names, like the original twitter poster.

The last lines on the list are labelled 'reserved16' and so on, we can easily tell that the previous 15 lines from DoNotAmplify to DoNotPublicPublish are the whole thing.

Boris Galerkin
Dec 17, 2011

I don't understand why I can't harass people online. Seriously, somebody please explain why I shouldn't be allowed to stalk others on social media!

pumpinglemma posted:

You’re the one who can’t provide a plausible alternative explanation, why the gently caress should I do your legwork for you?

I'm providing an explanation based on the screenshot of the names of the variables.

It's the exact same amount of legwork the dog twitter guy used when he tweeted his poo poo.

withoutclass
Nov 6, 2007

Resist the siren call of rhinocerosness

College Slice
It's pretty likely the Twitter employees that are left don't know what it does either.

Main Paineframe
Oct 27, 2010

Nenonen posted:

Does it make sense that one conflict is singled out? Or that information about Ukraine is likened to, say, hate speech or covid misinformation?

Absolutely. Ukraine war content gets special treatment from SA mods too, given how much misinformation, propaganda, blatant racism, and straight-up gore content is being posted from it. And most of that objectionable content being posted on SA has just been reposts from Twitter. I've never heard of any other conflict spawning as much goreposting as the Ukraine war did.

Nenonen
Oct 22, 2009

Mulla on aina kolkyt donaa taskussa

Main Paineframe posted:

Absolutely. Ukraine war content gets special treatment from SA mods too, given how much misinformation, propaganda, blatant racism, and straight-up gore content is being posted from it. And most of that objectionable content being posted on SA has just been reposts from Twitter. I've never heard of any other conflict spawning as much goreposting as the Ukraine war did.

It still makes zero sense to single it out as a topic when you also get such material from Syria, Afghanistan, US schools and dozens of other areas with armed conflict.

Main Paineframe
Oct 27, 2010

Nenonen posted:

It still makes zero sense to single it out as a topic when you also get such material from Syria, Afghanistan, US schools and dozens of other areas with armed conflict.

Not really, no. SA's never had a problem with people regularly sharing videos of soldiers killing themselves or videos of people getting shot at close range in Syria or Afghanistan or US schools. The Ukraine threads have drawn more QCS attention over the least year or two than all three of those subjects have drawn in SA's entire history.

It really is a unique (so far) conflict, not only because there's so much more horrible content coming out from it, but also because Westerners are so much more willing to watch and reshare it with racist screeds or :yeshaha: commentary attached.

duz
Jul 11, 2005

Come on Ilhan, lets go bag us a shitpost


Nenonen posted:

The last lines on the list are labelled 'reserved16' and so on, we can easily tell that the previous 15 lines from DoNotAmplify to DoNotPublicPublish are the whole thing.

At the top of the picture you can see what the list belongs to:
code:
  private lazy val thriftToModelMap: Map[s.SpaceSafetyLabelType, SpaceSafetyLabelType] = Map(
"DoNotAmplify" is one of the subcategories of the safety tag, like things that violate the Ukraine war policy.



Nenonen posted:

It still makes zero sense to single it out as a topic when you also get such material from Syria, Afghanistan, US schools and dozens of other areas with armed conflict.

Those probably fall under the generic categories listed elsewhere since Twitter didn't need to write a specific policy about them
code:
    MisinfoCivic -> SafetyResultReason.MisinfoCivic,
    MisinfoCrisis -> SafetyResultReason.MisinfoCrisis,
    MisinfoGeneric -> SafetyResultReason.MisinfoGeneric,
    MisinfoMedical -> SafetyResultReason.MisinfoMedical,

duz fucked around with this message at 17:14 on Apr 2, 2023

Vegetable
Oct 22, 2010

Most code makes little to no sense unless you’re looking at the broader infrastructure. It’s really silly to jump to conclusions based on a ctrl + f.

StumblyWumbly
Sep 12, 2007

Batmanticore!

Vegetable posted:

Most code makes little to no sense unless you’re looking at the broader infrastructure. It’s really silly to jump to conclusions based on a ctrl + f.

On the other hand, Musk absolutely does not deserve the benefit of the doubt. He's unbanned and offered personal customer service to Nazis, he's warped Twitter around his own ego, and he's said he may start charging Ukraine for Starlink, despite the US Gov't already paying him a ton.

If he wants folks to trust that he's doing the right thing, he needs to explain more shady poo poo.

Boris Galerkin
Dec 17, 2011

I don't understand why I can't harass people online. Seriously, somebody please explain why I shouldn't be allowed to stalk others on social media!

StumblyWumbly posted:

On the other hand, Musk absolutely does not deserve the benefit of the doubt. He's unbanned and offered personal customer service to Nazis, he's warped Twitter around his own ego, and he's said he may start charging Ukraine for Starlink, despite the US Gov't already paying him a ton.

If he wants folks to trust that he's doing the right thing, he needs to explain more shady poo poo.

Like I asked up thread, I'll ask again: show me an analysis from someone who has actually dove into the code and can explain what the code is doing, not someone who did a ctrl-f "ukraine" and posted a picture on twitter to get internet points. The code is supposedly publicly available and someone above has said this has already been done so it shouldn't be too hard right?

Because as alluded to by forums poster duz, the code shown in the screenshot in the tweet is just a list of safety(/sensitive?) topics. That's literally all it is.

Absurd Alhazred
Mar 27, 2010

by Athanatos
A breakdown of the code and what it means was posted in the Social Media Companies thread:

Shrecknet
Jan 2, 2005


Boris Galerkin posted:

There is no world that visa/mastercard and any legitimate payment processor is going to provide services for a website that generates deepfake porn based off of real people. This leaves the shady payment processors based in like Russia or cryptocurrency. Someone who’s savvy enough to convert cash to buttcoins just to buy illegal deepfake porn is probably a lost cause in regulation and the people who enter their credit card info into realvisapayment.ru are gonna get hosed with their identity being stolen eventually.

Next to the payment processors, you’ve got domain registrars, servers, cdns, etc, all of who are also not going to go anywhere near illegal deepfake porn.

Then you’ve got laws like revenge porn laws that already exist today. Not to mention CSAM related laws that will absolutely gently caress people over.

You’re right, deepfake porn is probably going to be a thing if it isn’t already. But it’s going to be relegated to the dark corners of the internet much in the same way CSAM is, and if they can pass laws for those types of contents I’m sure they can regulate deepfake porn under similar frameworks.

I made a thread to argue about the future of porn

Motronic
Nov 6, 2009

StumblyWumbly posted:

On the other hand, Musk absolutely does not deserve the benefit of the doubt.

This isn't giving him the benefit of the doubt. It about someone who doesn't understand what they're looking at or its context making wild assumptions and looking like an idiot in the process. Because that's simply not how variable names work or what they can tell you.

We've got a bunch of first day experts looking at code for the very first time doing a lot of CRTL-Fing and convinced this makes them very smart boys. Hey, that sounds familiar, almost like someone we've all heard of who pretends he know how to code but really doesn't know poo poo about so he asks his employees to do stupid poo poo like print out all their code on paper.

StumblyWumbly
Sep 12, 2007

Batmanticore!

Motronic posted:

This isn't giving him the benefit of the doubt. It about someone who doesn't understand what they're looking at or its context making wild assumptions and looking like an idiot in the process. Because that's simply not how variable names work or what they can tell you.

We've got a bunch of first day experts looking at code for the very first time doing a lot of CRTL-Fing and convinced this makes them very smart boys. Hey, that sounds familiar, almost like someone we've all heard of who pretends he know how to code but really doesn't know poo poo about so he asks his employees to do stupid poo poo like print out all their code on paper.
It is absolutely giving him the benefit of the doubt to say "maybe this code that looks like its lumping the Ukraine Crisis in with misinformation and hateful content is good, actually".

Motronic
Nov 6, 2009

StumblyWumbly posted:

It is absolutely giving him the benefit of the doubt to say "maybe this code that looks like its lumping the Ukraine Crisis in with misinformation and hateful content is good, actually".

This would be a true statement if you think that's what that code snippet looks like. And what I'm telling you is anyone who knows what they are talking about wouldn't make such an assumption based on a variable name a that small code snippet.

So again, a bunch of first day experts that don't know anything are the sum total of people who would form this opinion based on what was presented there.

Jose Valasquez
Apr 8, 2005

The code appears to be specific to Spaces, and it may not be used anymore, but the intention seems to be to hide spaces related to Ukraine, not anything related to "safe search" which seems marked by LoggedOutOrViewerOptInFiltering

The ones marked nsfw seem to be the only ones related to safe search. The Violence tags for example does not have the LoggedOutOrViewerOptInFiltering tag.

Tweets don't seem to have a similarly named label, but there's an opaque DoNotAmplify label that probably gets applied to specific topics defined elsewhere.

Jose Valasquez
Apr 8, 2005

Jose Valasquez posted:

The code appears to be specific to Spaces, and it may not be used anymore, but the intention seems to be to hide spaces related to Ukraine, not anything related to "safe search" which seems marked by LoggedOutOrViewerOptInFiltering

The ones marked nsfw seem to be the only ones related to safe search. The Violence tags for example does not have the LoggedOutOrViewerOptInFiltering tag.

Tweets don't seem to have a similarly named label, but there's an opaque DoNotAmplify label that probably gets applied to specific topics defined elsewhere.

There is a separate Tweet policy for AbusePolicyUkraineCrisisMisinformation to receive LimitedEngagement. This could be good or bad depending on who determines what is misinformation (It's Elon Musk so almost certainly bad)

Mega Comrade
Apr 22, 2004

Listen buddy, we all got problems!

Motronic posted:

This would be a true statement if you think that's what that code snippet looks like. And what I'm telling you is anyone who knows what they are talking about wouldn't make such an assumption based on a variable name a that small code snippet.

So again, a bunch of first day experts that don't know anything are the sum total of people who would form this opinion based on what was presented there.

I'll be totally honest, I barely looked at it. Thought it was funny, posted it. Didn't realise it would get such a reaction. Looking at the full code in more detail it still 'looks' like it's supressing Ukraine topics but Scala isn't a language I know so I could be wrong.

Twitter GitHub posted:

SafetyLabelType

Describes a particular policy violation for a given noun instance, and usually leads to reduced visibility of the labeled entity in product surfaces. There are many deprecated, and experimental safety label types. Labels with these safety label types have no effect on VF. Additionally, some safety label types are not used, and not designed for VF.


The OP though isn't some rando day 1 expert, he's a well known cyber security expert, reading others peoples code and figuring out what it is doing is his daily job, so yes, I did take his judgement at face value.

Mega Comrade fucked around with this message at 10:40 on Apr 3, 2023

Senor Tron
May 26, 2006


Main Paineframe posted:

Not really, no. SA's never had a problem with people regularly sharing videos of soldiers killing themselves or videos of people getting shot at close range in Syria or Afghanistan or US schools. The Ukraine threads have drawn more QCS attention over the least year or two than all three of those subjects have drawn in SA's entire history.

It really is a unique (so far) conflict, not only because there's so much more horrible content coming out from it, but also because Westerners are so much more willing to watch and reshare it with racist screeds or :yeshaha: commentary attached.

That sort of material has been getting shared increasingly commonly for most conflicts as the internet has grown. The big different for western audiences is that it's the first big conflict since the mainstream adoption of the internet that is both perceived as "them vs us" and "they" having been the ones who clearly started it, making some people more justified in enjoying seeing them suffer.

Boris Galerkin
Dec 17, 2011

I don't understand why I can't harass people online. Seriously, somebody please explain why I shouldn't be allowed to stalk others on social media!

Boris Galerkin posted:

I think Mr. Dog Poster above just did a ctrl-f for "ukraine" and tweeted about it to get internet points, after seeing other people with the same amount of understanding do the same.

Show me a tweet with someone who actually knows what the code is doing and I'll shut up about it.

(USER WAS PUT ON PROBATION FOR THIS POST)

Did I seriously eat a 12h probe for posting about a twitter user? The gently caress? Elon is that you?

Rocko Bonaparte
Mar 12, 2002

Every day is Friday!
Maybe? Somebody must of thought you were making GBS threads on a goon in-thread or something, but I have no idea which poster that could have even been. I'm kind of lost on the discussion now.

silence_kit
Jul 14, 2011

by the sex ghost

Boris Galerkin posted:

Did I seriously eat a 12h probe for posting about a twitter user? The gently caress? Elon is that you?

Honestly in D&D you should post with the expectation that if you post a minority opinion, you will be prosecuted to the full extent of the law.

Also, IMO you should also have the expectation that you’ll sometimes get wrongly convicted as well. Moderators don’t read threads, and often will take the tattletale’s account of your post at face value.

Rocko Bonaparte
Mar 12, 2002

Every day is Friday!
I thought the dog poster was one of the Twitter accounts so I am really confused. Like, I don't know who was talking about what due to that. Did somebody think they were getting dunked and reported it?

silence_kit
Jul 14, 2011

by the sex ghost
I’m obviously speculating here, since I’m not able to read reports, but I suspect that someone who didn’t like the opinion expressed in the post, tattled on the post using the post reporting system. The moderator took the report at face value, didn’t read the thread, and probated the poster.

Boris Galerkin
Dec 17, 2011

I don't understand why I can't harass people online. Seriously, somebody please explain why I shouldn't be allowed to stalk others on social media!
Yeah by "Mr Dog Poster" I was referring to the twitter/mastodon account with the handle "@GossiTheDog". I have/had no idea what his credentials are since it wasn't a link and I have no idea how to look up his account details on mastodon (nor would I have tbh).

Boris Galerkin
Dec 17, 2011

I don't understand why I can't harass people online. Seriously, somebody please explain why I shouldn't be allowed to stalk others on social media!

Absurd Alhazred posted:

A breakdown of the code and what it means was posted in the Social Media Companies thread:

Anyway, this is good information thanks. Although a lot of it is still just speculation since he doesn't show how these lists are being used and whatnot.

Either way, I still don't see what the big fuss is to be honest. The Ukraine War topic is a major current event happening right now, and there's a lot of tweets and discussion on it. It is probably easier to just create a new "nsfw" category on this topic than to lump it into one of the existing generic ones. In fact, it might even be better to single the topic out with its own label so that it can be treated differently, be that positively or negatively, by giving it different weights and/or exemptions for example.

I just wanna reiterate that the only thing that code shows is that there is a category specifically about the war in Ukraine, but the code as posted does not show how that category is being used.

e: Also I'm not giving Musk any benefit of doubt. I'm just looking at this from a realistic point of view and as someone with some experience/knowledge of programming.

Boris Galerkin fucked around with this message at 17:09 on Apr 3, 2023

withoutclass
Nov 6, 2007

Resist the siren call of rhinocerosness

College Slice
If someone attaches themselves to an idea and then you propose a position that undermines it they're going to feel like they themselves are under attack, no matter how reasonable.

pumpinglemma
Apr 28, 2009

DD: Fondly regard abomination.

Boris Galerkin posted:

Anyway, this is good information thanks. Although a lot of it is still just speculation since he doesn't show how these lists are being used and whatnot.

Either way, I still don't see what the big fuss is to be honest. The Ukraine War topic is a major current event happening right now, and there's a lot of tweets and discussion on it. It is probably easier to just create a new "nsfw" category on this topic than to lump it into one of the existing generic ones. In fact, it might even be better to single the topic out with its own label so that it can be treated differently, be that positively or negatively, by giving it different weights and/or exemptions for example.

I just wanna reiterate that the only thing that code shows is that there is a category specifically about the war in Ukraine, but the code as posted does not show how that category is being used.

e: Also I'm not giving Musk any benefit of doubt. I'm just looking at this from a realistic point of view and as someone with some experience/knowledge of programming.
Come on dude, there is literally an NSFW violence category that would neatly apply to everything NSFW about the war pictured in that code snippet. This is also not just speculation - the full code is literally public and that tweet thread is a cybersecurity expert talking about it with full context. We also now have two goons itt who have looked at, again, the actual full code - not tweets of snippets - and reckon it looks like suppression rather than NSFWing as well. Please either look at the code yourself and come back with actual evidence, or do some googling and find some experts who disagree, or stop meeting effort with low-effort.

HelloSailorSign
Jan 27, 2011

Senor Tron posted:

That sort of material has been getting shared increasingly commonly for most conflicts as the internet has grown. The big different for western audiences is that it's the first big conflict since the mainstream adoption of the internet that is both perceived as "them vs us" and "they" having been the ones who clearly started it, making some people more justified in enjoying seeing them suffer.

Libyan and Syrian civil war threads I remember some :stare: content going on, but maybe people were only interested because we had Goon on the Ground Caro, lol

Some of the ISIS videos in particular, gently caress man.

Boris Galerkin
Dec 17, 2011

I don't understand why I can't harass people online. Seriously, somebody please explain why I shouldn't be allowed to stalk others on social media!

pumpinglemma posted:

Come on dude, there is literally an NSFW violence category that would neatly apply to everything NSFW about the war pictured in that code snippet. This is also not just speculation - the full code is literally public and that tweet thread is a cybersecurity expert talking about it with full context. We also now have two goons itt who have looked at, again, the actual full code - not tweets of snippets - and reckon it looks like suppression rather than NSFWing as well. Please either look at the code yourself and come back with actual evidence, or do some googling and find some experts who disagree, or stop meeting effort with low-effort.

Yes I read the quoted tweet thread during my 12h probe. He doesn't actually get into how the code is actually being used, just speculates because he says that the full code isn't actually public. So once again, he's speculating based on variable names.

e: Most of the stuff he goes into is how tweets are amplified. He never gets into what those SafetyLabel categories are used for and/or how they are used. I did, in fact, take a look through the GitHub repository myself. There is no function that calls any of this poo poo being discussed.

e2: As far as the categories themselves, you're basically saying "there's already a 'history' section in the library, why does there need to be a separate 'european history' section?" Because sometimes it's not enough to use a generic label. Hope that helps.

Boris Galerkin fucked around with this message at 19:17 on Apr 3, 2023

Vegetable
Oct 22, 2010

pumpinglemma posted:

Come on dude, there is literally an NSFW violence category that would neatly apply to everything NSFW about the war pictured in that code snippet. This is also not just speculation - the full code is literally public and that tweet thread is a cybersecurity expert talking about it with full context. We also now have two goons itt who have looked at, again, the actual full code - not tweets of snippets - and reckon it looks like suppression rather than NSFWing as well. Please either look at the code yourself and come back with actual evidence, or do some googling and find some experts who disagree, or stop meeting effort with low-effort.
I don’t think anyone here who has looked at the code has said it’s suppression.

Adbot
ADBOT LOVES YOU

Blue Footed Booby
Oct 4, 2006

got those happy feet

HelloSailorSign posted:

Libyan and Syrian civil war threads I remember some :stare: content going on, but maybe people were only interested because we had Goon on the Ground Caro, lol

Some of the ISIS videos in particular, gently caress man.

I followed the Syria thread fairly obsessively for a while. Highlights include:
-various videos of ISIL destroying priceless artifacts, like the winged bull statues that once guarded the gates of Ninevah
-"unboxing" videos of captured poo poo, including a crate of STG-44s for some reason
-ISIL tweeting that they'd taken various hills, buildings, etc, including pictures. Then no-poo poo US state department accounts tweeting back FLIR videos of those same hills and buildings exploding, often with visible casualties. It was just a bit shy of "hey we killed ur doods" :psyduck:
-videos of regime tanks trundling through open plains completely unescorted, then getting hit with ATGMs and popping their turret. It was mostly posted in the context of, eg, "now the rebels have TOWs," "why does the Syrian army keep doing this," and the inevitable "why do the turrets do that"
-links to the beheading videos, discussion of said videos and whether they should or shouldn't be viewed. I think this is the one category where mods got involved, because people were traumatizing themselves in some misguided quest to stay informed

In general, literally everyone agreed on ISIL needing to eat poo poo on like a cosmic level, most folks wanted Assad to get merced, and there were various strong opinions for or against various levels of intervention, but I don't remember there being much ghoulish cheerleading or what have you. Nor do I recall goons looking to paint anyone who voluntarily views war footage as bloodthirsty and deranged.

Blue Footed Booby fucked around with this message at 19:29 on Apr 3, 2023

  • 1
  • 2
  • 3
  • 4
  • 5
  • Post
  • Reply