25 ms·

So car manufacturers need to stop making cars because someone might drive drunk? Kitchen knife manufacturers need to somehow prevent their products being used a

So car manufacturers need to stop making cars because someone might drive drunk? Kitchen knife manufacturers need to somehow prevent their products being used as weapons? This line of thinking is appropriate to a degree, but not really all that useful. Many products can be abused, and there's no feasible way to prevent that abuse.

yawpitch 3y ago
Seems to me there is an obvious qualitative difference between a company that manufacturers cars or kitchen knives and a company that creates a service that uses random matchmaking to repeatedly introduce a serial pedophile to unsuspecting underage victims. Sure, people _can_ abuse any service, this seems like a service that wasn’t just ripe for abuse, it was essentially perfectly designed to enable it. Moreover in this case there absolutely appears to be numerous feasible ways to have headed off that particular avenue of abuse. This is the same ludicrously weak argument that is constantly and erroneously applied to guns… neither a car nor a kitchen knife are primarily (much less exclusively) instruments of harm, guns are. Well, turns out, so are random matchmaking services that link adult users with children and let them view each other, and that appears kind of obvious in hindsight.
- Manuel_D 3y ago
  Ford might manufacture a car driven by a serial drunk driver. Perhaps they need to install breathalyzers in all their cars, by default. I'm not really sure how much more ripe for abuse Omegle was compared to, say, Discord. Pretty much any video chat service can be abused to send or receive illegal content, and to abuse and manipulate other people. These are risks inherent to anything enabling communication. Short of a panopticon where all communications are manually approved by a human moderator, there's no sure way to prevent abuse (and even then human moderators are fallible). There ought to be some reasonable attempts to mitigate abuse, like a reporting functionality. But beyond that I don't see much more Omegle could have reasonably done.
  - yawpitch 3y ago
    Again the pointless and frankly silly comparison to cars… they’re categorically unrelated product classes: Ford‘s cars didn’t intentionally, as a feature of the vehicle, randomly put you into head on collision situations with others, which is essentially what Omegle did by design, while also NOT (in any sense) being fruitfully comparable to a manufacturer of vehicles. Same with the equally pointless comparison to Discord… Omegle wasn’t merely a video chat service, it made random matches that the user could narrow by identifying their own interests; an adult male user identifying as being deeply interested in things only children would be interested in could readily and easily (and obviously) weaponize the platform, and Omegle absolutely could have (and should have) used the many and obvious means available for profiling and identifying such incongruous users, which (sure) would include human moderators. There’s an enormous ethical difference between not doing anything whatsoever to prevent abuse and perfectly preventing abuse, and you seem to think they had no obligation to prevent any because they couldn’t prevent all… they don’t exist any more (thankfully) because lawyers started to (correctly) point out that that isn’t how either ethics or tort law work.
    - Manuel_D 3y ago
      Omegle didn't intentionally put people on a collision course with abusers either. If Omegle was intentionally facilitating abuse as you put it, then so is IRC and effectively any other public communications mechanism: because anyone could be an abuser. Even just preventing 1% of abuse would probably have been beyond the capabilities of this site. You write that they should flag adult men listing interest in topics associated with children. How are they supposed to identify the gender and age of users? People under 18 are prohibited from the site, yet that clearly failed. Human moderation can't even monitor a fraction of one percent The "many obvious" ways of preventing abuse were in fact attempted [1]: > Omegle implemented a "monitored" video chat, to monitor misbehavior and protect people under the age of 18 from potentially harmful content, including nudity or sexual content. However, the monitoring is not very effective, and users can often skirt around bans. Sure, Omegle "randomly put you into head on collision situations with others", but so is every other public communications: IRC, discord, Xbox Live, pretty much anywhere you can meet random people on the Internet fits into this category. 1. https://en.m.wikipedia.org/wiki/Omegle https://en.m.wikipedia.org/wiki/Omegle
- rockskon 3y ago
  What ways could've headed off abuse thar wouldn't have defeated the entire purpose of the service - matching random people with each other?
  - yawpitch 3y ago
    Age verifying every user and not allowing children on the service at all, or only matching children with children, as an obvious first. Alternatively, maybe just randomly sampling video feeds and running them through an ML classifier to see if, I dunno, “adult male penis” was a high probability on one side and “extremely uncomfortable looking child” was a high probability on the other? In hindsight the entire purpose of the service was a bad idea, absent minimal efforts to avoid its trivial weaponization by users with an obvious motive for using the means and opportunity the service was providing by design.
    - rockskon 3y ago
      So one of your solutions to a service aimed at randomly match anonymous people is to get rid of anonymity? I asked you how do you solve the problem without defeating the purpose of Omegle. Your solution is the equivalent to someone asking how to solve world hunger and you responding with "Just feed them. Duh."
meepmorp 3y ago
> So car manufacturers need to stop making cars because someone might drive drunk? In the US, at least, bars can be found liable for patrons drunk driving. That's probably a closer analogy to Omegle than a car manufacturer, since its patrons hang out at the "establishment," engaging in potentially risky behavior. That's not a comment on the validity of the lawsuit, but the situation isn't as simple as you make out.
- Manuel_D 3y ago
  https://www.washingtoninjurylaw.com/can-bartenders-be-liable-for-a-patrons-dui-in-washington/ https://www.washingtoninjurylaw.com/can-bartenders-be-liable... > The drunk driver can sue the bar or bartender for allowing them to become intoxicated to a dangerous level. Individuals may file a lawsuit, but this does not always mean it will hold up in court. Dram shop laws in most states make clear definitions to avoid false liability. Washington is one of those states. It looks like it's more the case that the drunk driver tries to sue the bar, not that the government is pursuing the bar. Furthermore, some States specifically forbid attempts to hold the bar liable for drunk drivers. And even if it were, it's vastly easier for a bar to monitor the conduct of patrons than a web service. The scale of the latter is too great to make non-automated moderation feasible.

yawpitch 3y ago

Seems to me there is an obvious qualitative difference between a company that manufacturers cars or kitchen knives and a company that creates a service that uses random matchmaking to repeatedly introduce a serial pedophile to unsuspecting underage victims. Sure, people _can_ abuse any service, this seems like a service that wasn’t just ripe for abuse, it was essentially perfectly designed to enable it. Moreover in this case there absolutely appears to be numerous feasible ways to have headed off that particular avenue of abuse. This is the same ludicrously weak argument that is constantly and erroneously applied to guns… neither a car nor a kitchen knife are primarily (much less exclusively) instruments of harm, guns are. Well, turns out, so are random matchmaking services that link adult users with children and let them view each other, and that appears kind of obvious in hindsight.

Manuel_D 3y ago
Ford might manufacture a car driven by a serial drunk driver. Perhaps they need to install breathalyzers in all their cars, by default. I'm not really sure how much more ripe for abuse Omegle was compared to, say, Discord. Pretty much any video chat service can be abused to send or receive illegal content, and to abuse and manipulate other people. These are risks inherent to anything enabling communication. Short of a panopticon where all communications are manually approved by a human moderator, there's no sure way to prevent abuse (and even then human moderators are fallible). There ought to be some reasonable attempts to mitigate abuse, like a reporting functionality. But beyond that I don't see much more Omegle could have reasonably done.
- yawpitch 3y ago
  Again the pointless and frankly silly comparison to cars… they’re categorically unrelated product classes: Ford‘s cars didn’t intentionally, as a feature of the vehicle, randomly put you into head on collision situations with others, which is essentially what Omegle did by design, while also NOT (in any sense) being fruitfully comparable to a manufacturer of vehicles. Same with the equally pointless comparison to Discord… Omegle wasn’t merely a video chat service, it made random matches that the user could narrow by identifying their own interests; an adult male user identifying as being deeply interested in things only children would be interested in could readily and easily (and obviously) weaponize the platform, and Omegle absolutely could have (and should have) used the many and obvious means available for profiling and identifying such incongruous users, which (sure) would include human moderators. There’s an enormous ethical difference between not doing anything whatsoever to prevent abuse and perfectly preventing abuse, and you seem to think they had no obligation to prevent any because they couldn’t prevent all… they don’t exist any more (thankfully) because lawyers started to (correctly) point out that that isn’t how either ethics or tort law work.
  - Manuel_D 3y ago
    Omegle didn't intentionally put people on a collision course with abusers either. If Omegle was intentionally facilitating abuse as you put it, then so is IRC and effectively any other public communications mechanism: because anyone could be an abuser. Even just preventing 1% of abuse would probably have been beyond the capabilities of this site. You write that they should flag adult men listing interest in topics associated with children. How are they supposed to identify the gender and age of users? People under 18 are prohibited from the site, yet that clearly failed. Human moderation can't even monitor a fraction of one percent The "many obvious" ways of preventing abuse were in fact attempted [1]: > Omegle implemented a "monitored" video chat, to monitor misbehavior and protect people under the age of 18 from potentially harmful content, including nudity or sexual content. However, the monitoring is not very effective, and users can often skirt around bans. Sure, Omegle "randomly put you into head on collision situations with others", but so is every other public communications: IRC, discord, Xbox Live, pretty much anywhere you can meet random people on the Internet fits into this category. 1. https://en.m.wikipedia.org/wiki/Omegle https://en.m.wikipedia.org/wiki/Omegle
rockskon 3y ago
What ways could've headed off abuse thar wouldn't have defeated the entire purpose of the service - matching random people with each other?
- yawpitch 3y ago
  Age verifying every user and not allowing children on the service at all, or only matching children with children, as an obvious first. Alternatively, maybe just randomly sampling video feeds and running them through an ML classifier to see if, I dunno, “adult male penis” was a high probability on one side and “extremely uncomfortable looking child” was a high probability on the other? In hindsight the entire purpose of the service was a bad idea, absent minimal efforts to avoid its trivial weaponization by users with an obvious motive for using the means and opportunity the service was providing by design.
  - rockskon 3y ago
    So one of your solutions to a service aimed at randomly match anonymous people is to get rid of anonymity? I asked you how do you solve the problem without defeating the purpose of Omegle. Your solution is the equivalent to someone asking how to solve world hunger and you responding with "Just feed them. Duh."

meepmorp 3y ago

> So car manufacturers need to stop making cars because someone might drive drunk? In the US, at least, bars can be found liable for patrons drunk driving. That's probably a closer analogy to Omegle than a car manufacturer, since its patrons hang out at the "establishment," engaging in potentially risky behavior. That's not a comment on the validity of the lawsuit, but the situation isn't as simple as you make out.

Manuel_D 3y ago
https://www.washingtoninjurylaw.com/can-bartenders-be-liable-for-a-patrons-dui-in-washington/ https://www.washingtoninjurylaw.com/can-bartenders-be-liable... > The drunk driver can sue the bar or bartender for allowing them to become intoxicated to a dangerous level. Individuals may file a lawsuit, but this does not always mean it will hold up in court. Dram shop laws in most states make clear definitions to avoid false liability. Washington is one of those states. It looks like it's more the case that the drunk driver tries to sue the bar, not that the government is pursuing the bar. Furthermore, some States specifically forbid attempts to hold the bar liable for drunk drivers. And even if it were, it's vastly easier for a bar to monitor the conduct of patrons than a web service. The scale of the latter is too great to make non-automated moderation feasible.