|
namaste friends posted:loving lol at owlofcreamcheese defending pedophiles right to privacy I don't think pedophiles have any right to privacy, I think a well done program to catch and charge pedophiles would be amazing and great and that being possible is a great thing about technology. But this sort of proactive scanning of messages for crime is not the way any other major medium of communication works so someone has to figure out the exact details of the policies. No one has ever really done this before on a major form of communication. And once you decide to do it you gotta figure out what sort of public system or panel of experts or internal team is in charge of figuring out standards for what is flagged and what isn't. Like you can set up an automated scanning algorithm but you can't just talk into it and say "find pedophiles" you gotta actually type words and search terms and filters and scoring algorithms and stuff and somebody somewhere has to figure that out and facebook is asking "who should figure that out"?
|
#
?
Mar 6, 2018 03:14
|
|
|
|
| # ? Apr 28, 2024 01:06 |
|
|
check out my new AI ML algo that scans for retards beep your it
|
|
#
?
Mar 6, 2018 03:18
|
|
|
Before reading the article one thought that I had was that if you wanted to ask people some more nuanced questions as part of a survey, perhaps you might also ask less nuanced questions to add context around the other answers? E.g "well N% of respondents think doing X is okay, but 90% of them also said sexual harassment of children is ok, so maybe we ignore them" (perhaps weeding out junk answers?) Anyway the article didn't actually seem consistent with that guess, and I've also never designed a survey, so that's probably not it.
|
|
#
?
Mar 6, 2018 03:46
|
|
|
namaste friends posted:check out my new AI ML algo that scans for retards Like once you have decided 'detect and ban pedophiles" you still have to actually take actual steps to do it, it's not a button always had and just pick not to press. You have to actually find actual standards for what specific words and behaviors specifically have what consequences. If two 60 year old men say "you got a picture?" to a 12 year old girl and one is a weirdo that is creeping on random girl's profiles and the other is a grandpa responding to an excited grandchild talking about being the lead in a school play and her crazy costume she had to wear then you gotta develop the algorithm to detect that phrase and a flow chart for a moderator to quickly determine which category as they search through thousands of flags a day. And facebook is asking if that set of standards should be a thing the community somehow decides, if it's something a team of expert lawyers should write or if it's a thing facebook itself can internally decide.
|
|
#
?
Mar 6, 2018 03:57
|
|
|
actually it's a team of expert lawyers. not even, it's like a lawyer intern telling the shithead SDE 1 to just loving code it you idiot
|
|
#
?
Mar 6, 2018 03:58
|
|
|
Owlofcreamcheese posted:I don't think pedophiles have any right to privacy, I think a well done program to catch and charge pedophiles would be amazing and great and that being possible is a great thing about technology. But this sort of proactive scanning of messages for crime is not the way any other major medium of communication works so someone has to figure out the exact details of the policies. No one has ever really done this before on a major form of communication. And once you decide to do it you gotta figure out what sort of public system or panel of experts or internal team is in charge of figuring out standards for what is flagged and what isn't. Like you can set up an automated scanning algorithm but you can't just talk into it and say "find pedophiles" you gotta actually type words and search terms and filters and scoring algorithms and stuff and somebody somewhere has to figure that out and facebook is asking "who should figure that out"? There’s something wrong with you dude. Like, sure there’s a kernel of a discussion somewhere in your drivel, but this is a thread specifically about making fun of lovely companies led by autistic psychopaths like yourself. If you can’t laugh at the fact that a half-trillion dollar company is so thoroughly shook by bad actors exploiting their pollyanna hacker creed bullshit that they can’t even form an opinion on child porn, then maybe you need to log off and go outside for a few minutes.
|
|
#
?
Mar 6, 2018 05:00
|
|
|
namaste friends posted:check out my new AI ML algo that scans for retards Turn your monitor on 
|
|
#
?
Mar 6, 2018 05:33
|
|
|
I’m not defending pedophiles or saying they have privacy or freedom of speech or anything. Just that Facebook has millions of messages a day and any sort of automated or mass moderated system actually would require real thought to do well. Anyone they catch should be sent instantly to jail and sent to hell and given every penalty possible but like, catching people still requires someone to write a good policy. What words are flagged, who can determine ambitious phrasing, who is on the radar to be checked and someone has to write it in an iron clad way. And then once Facebook is in the business of monitoring messages for crime we need a method to prevent other countries demanding they extending that to things that aren’t good laws like Russia demanding the technology help arrests gays (and maybe thought on us using it to force correct values on countries or states that are still very gross about their age of concent and a policy on what to do in those places)
|
|
#
?
Mar 6, 2018 05:42
|
|
|
Owlofcreamcheese posted:I’m not defending pedophiles or saying they have privacy or freedom of speech or anything. Just that Facebook has millions of messages a day and any sort of automated or mass moderated system actually would require real thought to do well. Anyone they catch should be sent instantly to jail and sent to hell and given every penalty possible but like, catching people still requires someone to write a good policy. What words are flagged, who can determine ambitious phrasing, who is on the radar to be checked and someone has to write it in an iron clad way. And then once Facebook is in the business of monitoring messages for crime we need a method to prevent other countries demanding they extending that to things that aren’t good laws like Russia demanding the technology help arrests gays (and maybe thought on us using it to force correct values on countries or states that are still very gross about their age of concent and a policy on what to do in those places) hmm good point this is so hard good thing no one is even trying to block inappropriate content anywhere
|
|
#
?
Mar 6, 2018 05:48
|
|
|
I have a hard time believing that Facebook, Twitter, et al are not already scraping private communication for "demographic information", though there’s likely not a person watching the informantion.
|
|
#
?
Mar 6, 2018 05:55
|
|
|
Owlofcreamcheese posted:I’m not defending pedophiles or saying they have privacy or freedom of speech or anything. Just that Facebook has millions of messages a day and any sort of automated or mass moderated system actually would require real thought to do well. Anyone they catch should be sent instantly to jail and sent to hell and given every penalty possible but like, catching people still requires someone to write a good policy. What words are flagged, who can determine ambitious phrasing, who is on the radar to be checked and someone has to write it in an iron clad way. And then once Facebook is in the business of monitoring messages for crime we need a method to prevent other countries demanding they extending that to things that aren’t good laws like Russia demanding the technology help arrests gays (and maybe thought on us using it to force correct values on countries or states that are still very gross about their age of concent and a policy on what to do in those places) The question was in an ideal world. Obviously in an ideal world the algorithm is perfectly accurate with no false positives, so there's no reason to not have it block sexual harassers. It's really quite simple.
|
|
#
?
Mar 6, 2018 05:55
|
|
|
The broader question here is basically "Would you be okay if Facebook spied on your messages to monitor for illegal activity?" It's obvious that this snooping would rile up privacy advocates. This thread generally has cynical posters who already assume Facebook is spying on them, but for many people such a practice would be a shock and something to kick a fuss about. That's why, I assume, they're trying to erode the principle by starting with the extremely outrageous stuff. But they went too far that extreme and now people are mad that Facebook aren't already spying them for that poo poo. Really, whatever they do, they should just make a decision in-house. I don't think a survey on "Are you ok if we spy on you?" is ever going to end well.
|
|
#
?
Mar 6, 2018 06:06
|
|
|
Pretty sure Western Union would censor your telegrams.
|
|
#
?
Mar 6, 2018 06:47
|
|
|
Also Facebook is in trouble because most of the forms of content that they want to block are more subjective. Child porn is a relatively simple problem, technically speaking; there are clear rules for what defines it, there are no circumstances in which it's acceptable, and nobody cares that much if your detection algorithm is extra-sensitive and makes a bunch of type 1 errors. But Facebook is also under a lot of pressure to filter out "extremist content" and "fake news", and half their users will poo poo bricks and start screaming about leftist censorship if they actually block that stuff. Whole reason they had the fake news problem in the first place is because they had to get rid of their human review team due to reality having a liberal bias. Nobody was surprised when the news-checking algorithm they replaced the team with got systematically exploited.
|
|
#
?
Mar 6, 2018 07:13
|
|
|
I'm not buying that it's important to maintain some veneer of impartiality. I mean gently caress it's a veneer that I should pay any mind to a bunch of loving 4chan racists for the benefit of free speech like that Huffman rear end in a top hat from Reddit. Ask Breitbart of they could survive on ad revenue alone after a bunch of anonymous Twitter prior started shaming corporations with ad buys on their site.
|
|
#
?
Mar 6, 2018 07:19
|
|
|
It's not about impartiality. The people who say "technically speaking it's not child sexual assault" are a tiny minority. The people who say "I'm not doing hate speech, I'm just a bit white supremacist" are, like, half the country. The two are not equivalent. If Facebook draws a line anywhere, they have to draw it somewhere, and anywhere they draw the line on hate speech is going to piss off at least half their user base. If they really mess it up they'll piss off their entire user base. Facebook is screwed because their entire platform is based on enabling assholes.
|
|
#
?
Mar 6, 2018 07:50
|
|
|
One of the core issues with dealing with hate speech is that any line will end up with more punishment against righties than lefties. America's right wing is racist as hell and it's just coming further and further out into the open. The left is saying "hey racism is bad let's not do that" while the right is saying "we're OK with literal Nazis being on our side." The alt right is all about spreading white supremacy and misogyny as far and wide as they can. They're the ones screaming loudest about censorship so there's really on way Facebook and Twitter can do much about it without pissing off a significant chunk of the current right.
|
|
#
?
Mar 6, 2018 08:43
|
|
|
Avenging_Mikon posted:I have a hard time believing that Facebook, Twitter, et al are not already scraping private communication for "demographic information", though there’s likely not a person watching the informantion.
|
|
#
?
Mar 6, 2018 11:58
|
|
|
https://twitter.com/APompliano/status/970653385648205824 I screencapped one "thread", but all the responses are loving gold.  (For those tuning in, "longer time preference" means "white" in libertarian speak)
|
|
#
?
Mar 6, 2018 12:35
|
|
|
Absurd Alhazred posted:(For those tuning in, "longer time preference" means "white" in libertarian speak) That... does not even begin to make sense.
|
|
#
?
Mar 6, 2018 13:18
|
|
I CANNOT EJACULATE WITHOUT SEEING NATIVE AMERICANS BRUTALISED!
|
suck my woke dick posted:That... does not even begin to make sense. According to lolberts black people have a lower time preference than white people because ?????
|
|
#
?
Mar 6, 2018 13:25
|
|
|
BENGHAZI 2 posted:According to lolberts black people have a lower time preference than white people because ????? Assuming they mean "discounting future benefits/costs more heavily" with "lower time preference" instead of some lolbertarian-speak concept nobody else has ever heard of, I present exhibit A, every lolbertarian-led business ever.
|
|
#
?
Mar 6, 2018 13:27
|
|
|
suck my woke dick posted:Assuming they mean "discounting future benefits/costs more heavily" with "lower time preference" instead of some lolbertarian-speak concept nobody else has ever heard of, I present exhibit A, every lolbertarian-led business ever. I also guarantee you that OP doesn't understand a single one of the things he listed, either, because if you had a bit of understanding of anything you wouldn't be promoting cryptocurrencies (other than as a con-man). Libertarians don't suffer from too much self-reflection.
|
|
#
?
Mar 6, 2018 13:35
|
|
|
BENGHAZI 2 posted:According to lolberts black people have a lower time preference than white people because ????? It's referencing the stanford marshmallow experiment. That "you can have one candy now, or a bunch of candy in little while" thing where before a certain age every kid takes the bad deal and most people eventually grow up and take the better deal. Kids that grow up in stable and predictable environments learn to pre-plan early, people that grow up in unstable unpredictable environments (extreme poverty) don't learn to plan as young or as well. White kids tend to score high on it because they on average have the most stable environments provided for them. Racists make this into "whites are best" idea where it's inherent and not just another subtle privilege being white gets you.
|
|
#
?
Mar 6, 2018 13:51
|
|
|
I imagine the thought process is more along the lines of 'Yeah right, like you're going to actually give me the bag'. The disadvantaged grow up used to broken promises. Either that or 'I can waste time waiting for a bag of cheap candy or have one now and get out of here'.
|
|
#
?
Mar 6, 2018 13:53
|
|
|
Inescapable Duck posted:I imagine the thought process is more along the lines of 'Yeah right, like you're going to actually give me the bag'. The disadvantaged grow up used to broken promises. Yeah or "now is a calm time to eat a candy, in an hour who knows who is going to be yelling or storming around" or "maybe in an hour my awful drunk dad will eat the candy, he's not here now so I get this one" or "I'm 4 and I've moved 6 times, we might not even be in this city in a week" or just the general "chaos makes it harder to observe inputs giving outputs when you are 2 so you have to logic it out when you are older and not intuitively notice it as a baby". But because chaotic lives aren't distributed fairly in the US certain groups get them more than other groups and it's another thing racists can point to like it's a white supremacy thing instead of an effect white supremacy forces onto disadvantaged groups.
|
|
#
?
Mar 6, 2018 14:27
|
|
|
i dont see how studying history preferences you towards cryptocurrency, unless you think of it as like "history of filthy fiat" or something dumb and not "wow history is full of bullshit speculation scams for idiots"
|
|
#
?
Mar 6, 2018 16:55
|
|
boner confessor posted:i dont see how studying history preferences you towards cryptocurrency, unless you think of it as like "history of filthy fiat" or something dumb and not "wow history is full of bullshit speculation scams for idiots" those bullshit speculation scams for idiots made a bunch of people rich, and as libertarians will tell you they are too smart to fall into the same traps as the plebs
|
|
|
#
?
Mar 6, 2018 18:23
|
|
|
Ignatius M. Meen posted:those bullshit speculation scams for idiots made a bunch of people rich, and as libertarians will tell you they are too smart to fall into the same traps as the plebs *libertarian of 1920 standing waist deep in a florida swamp* "golly i'm glad i bought all this land, it will be worth millions some day! i just need to find some sucker to take some of it off my hands so i can pay my costs..."
|
|
#
?
Mar 6, 2018 18:25
|
|
|
Domino's is expanding it's autonomous vehicle pizza delivery beyond Ann Arbor now: http://www.mlive.com/business/ann-arbor/index.ssf/2018/03/dominos_pizza_expanding_self-d.html I know this thread thinks autonomous vehicles are somewhere between 50 years off and completely impossible, but they've been working pretty well in test environments. Like the article says, autonomous pizza delivery has been working in Ann Arbor for a while already. a foolish pianist fucked around with this message at 19:29 on Mar 6, 2018 |
|
#
?
Mar 6, 2018 19:23
|
|
|
a foolish pianist posted:I know this thread thinks autonomous vehicles are somewhere between 50 years off and completely impossible, but they've been working pretty well in test environments. it's just a marketing stunt on domino's part, the ideal form factor for a pizza delivery vehicle is a large quadcopter drone and not a full on car and this isn't indicative of whether or not full automation is possible
|
|
#
?
Mar 6, 2018 19:28
|
|
|
boner confessor posted:it's just a marketing stunt on domino's part, the ideal form factor for a pizza delivery vehicle is a large quadcopter drone and not a full on car The ideal form factor would some ballistic capsule that arrives on your porch, probably, but the current standard in pizza delivery is a car.
|
|
#
?
Mar 6, 2018 19:32
|
|
|
a foolish pianist posted:The ideal form factor would some ballistic capsule that arrives on your porch, probably, but the current standard in pizza delivery is a car. it is if you're trying to piggyback your self driving car test with a corporate partner to drum up publicity, yeah from an efficiency standpoint it makes very little sense to use a 2,000lb vehicle to deliver a pizza when you could use a 75lb vehicle
|
|
#
?
Mar 6, 2018 19:38
|
|
|
boner confessor posted:from an efficiency standpoint it makes very little sense to use a 2,000lb vehicle to deliver a pizza when you could use a 75lb vehicle It's weird to see you specifically trying to counter something as a techno fantasy stunt with some claim they would be better off doing something like 500x more fantastical.
|
|
#
?
Mar 6, 2018 19:44
|
|
|
Owlofcreamcheese posted:It's weird to see you specifically trying to counter something as a techno fantasy stunt with some claim they would be better off doing something like 500x more fantastical. the guy defending prestige chicken breeding and sniffing out pedophiles on social media thinks im doing something weird yall
|
|
#
?
Mar 6, 2018 19:45
|
|
|
boner confessor posted:it is if you're trying to piggyback your self driving car test with a corporate partner to drum up publicity, yeah From a risk management standpoint, a 2,000lb vehicle whose failure mode is stopping is better than a 75lb vehicle whose failure mode is falling out of the sky. Yeah, it's a stunt, and there are probably some clear rules about what conditions it can be used in, but that's okay. It's still pretty impressive. Baronash fucked around with this message at 19:52 on Mar 6, 2018 |
|
#
?
Mar 6, 2018 19:48
|
|
|
Baronash posted:From a risk management standpoint, a 2,000lb vehicle whose failure mode is stopping is better than a 75lb vehicle whose failure mode is falling out of the sky. Why not a small, robotic wheeled vehicle? If no one is going to be inside it, what are the regs on weight/size? Sort of a self-driving trike.
|
|
#
?
Mar 6, 2018 19:53
|
|
|
Metal Gears should be delivering pizza IMO
|
|
#
?
Mar 6, 2018 19:57
|
|
|
It should be a giant pizza-shaped wheel (also painted to look like a pizza) that rolls around like one of those single-wheel motorcycles with the giant wheel.
|
|
#
?
Mar 6, 2018 19:59
|
|
|
|
| # ? Apr 28, 2024 01:06 |
|
|
VideoGameVet posted:Why not a small, robotic wheeled vehicle? If no one is going to be inside it, what are the regs on weight/size? Sort of a self-driving trike. Because teenagers would be kicking them over to get at their soft pizza innards.
|
|
#
?
Mar 6, 2018 20:06
|
|
