4 ms·

My analysis of abuse in social media: it's an unavoidable side-product of buzz mechanics. Buzz exists in social media because the ad-based company that invented

My analysis of abuse in social media: it's an unavoidable side-product of buzz mechanics. Buzz exists in social media because the ad-based company that invented it needed a slot-machine-like mechanism to keep users addicted (has already been discussed here recently). As such, for me the root of both buzz and abuse are features like shared global news-feed and real-time trending topics (or network-wide search by some popularity-like criterion in general). Abuse mostly works in medium to large groups and it crucially depends on dehumanization of the target. Afaik a classical way to unspring abuse IRL is to organize small isolated confrontations, exactly the opposite of doing an epic moral speech in front of the crowd because one cannot have reasonable and logical arguments in a big group. So in the end a global search ranked by recent popularity is exactly the opposite of what is needed for peacemaking: any heated argument will attract more and more bystanders and there is no way to reverse the steam. So what solution? My guess is that only whitelisting makes sense and with a granularity lower than instances. One could also provide tools for instance moderators to make a particular thread private to isolate it. We should also not provide any popularity based ranking, leaving that to specific structures dedicated to it, with their own rules and a well-defined editorial policy (like hacker news or other). Of course we cannot just stop people from ranking things by popularity, we have to actually make it impossible. One way for that is that links really be directed: retooting something should not inform the target, making the retoot-count a hard edge-reversal problem (which would require full network crawling). We could even make crawling harder by a small rate limiting or proof a work in the query protocol (eg just making the query expensive for the client). Does this sound sensible? edit: actually i somewhat just expanded on this comment below: https://news.ycombinator.com/item?id=17895159 https://news.ycombinator.com/item?id=17895159

joering2 8y ago
Yes but I think there are two problems with your ideas, at least when it comes to Twitter: > organize small isolated confrontations, exactly the opposite of doing an epic moral speech in front of the crowd because one cannot have reasonable and logical arguments in a big group. That won't work. Open any Trump tweet and click on any person whether he/she is for or against president. You will see never-ending stream of tweets back and forth at each other, one party tries to convince other that Trump cheated on his wife, meanwhile other tries to convince that Clinton had her server hacked. Eventually such stream boils down to competition of who is more aggressive, rude and provocative against each other. Many such conversations ends with Hitler being injected, but of course no party will ever see the other side. I literally monitored 3 such streams of women wasting 10 days of their lives going after each other with frequency of one tweet per 45 minutes. Your idea to break it down and sort of "mediate" between two people that honestly are not interested in even hearing what the other side has to say, but are only engaged to vent off and blindly cheer whatever cool aid they been already sold on long time ago (i.e. that Trump is incredible president, not a con man, or Clinton is amazing women, not a crook) won't work. Second and finally: > My guess is that only whitelisting makes sense and with a granularity lower than instances. Twitter has signed off its death sentence when it went public. Its a tectonic shift within your organization when you move from serving users for the good sake of the whole system, into serving stockholders that only spend 15 minutes with you, once a year at the end of financial period and will only look into one-page accounting statements of profits and loss. While twitter users and even owners may want the most happy, prosperous and healthy environment in which everyone thrives and enjoy spending their evening reading and responding to tweets, the truth is the only people Twitter answers to is stockholders who only wants to squeeze as many ads into your timeline as possible. I'll say it even more: if Twitter starts fixing its feeds and doing further account purges, they will get sued by stockholders because obviously less users on platform equals to less ads-engagement. And its irrelevant whether you engaged with trollbots or real genuine people; it only matters you seen (and hopefully clicked) on ads. Does this sound about right?
- lapinot 8y ago
  Yeah, not sure. For the first point, actually i don't really care: eg the action i would like to take is simply to ignore the whole thread for myself (and thus in a sensible distributed system where people might make their policy depend on mine because they trust me, it would also lower the rank of this thread and in the end isolate it in some limbo). And after some time, one can hope that a long-term effect of discouraging abuse, shitstorm and humor/meme buzz will educate users and make "interesting" uses take over (eg these people will either get off this particular network or stop participating in flame wars because there are less of them and that there is no social reward on the network). And for the second point, i completely agree with you: there is no way that twitter makes the right thing. But i was mainly talking about what "a hypothetical well-run probably federated or even fully distributed social network" (possibly mastodon but not restricted to it).
  - joering2 8y ago
    > And after some time, one can hope that [...] because there are less of them and that there is no social reward on the network). That's a wishful thinking! One could say the same about first color TV when first comedy serial was aired; that eventually people will stop wasting their time watching silly shows (point given you cant argue with TV-set) and eventually move on to be more productive etc. To the contrary I don't think we seen a mountain top just yet; still I meet people who don't know what Twitter is; their path to being daily engaged and being dragged into trollbots' wars is probably few years down the road.
    - lovich 8y ago
      Eh, I've met a few people who had only heard of the name and wondered what it was all about. When I showed them how to sign up and look for hashtags, they quickly gave up between the flame wars and the bad ui. I don't think Twitter is going to reach anywhere near the ubiquity of tv or movies for entertainment. Most tv shows and movies don't have a good chance of making you feel angry, frustrated, or depressed after watching them

joering2 8y ago

Yes but I think there are two problems with your ideas, at least when it comes to Twitter: > organize small isolated confrontations, exactly the opposite of doing an epic moral speech in front of the crowd because one cannot have reasonable and logical arguments in a big group. That won't work. Open any Trump tweet and click on any person whether he/she is for or against president. You will see never-ending stream of tweets back and forth at each other, one party tries to convince other that Trump cheated on his wife, meanwhile other tries to convince that Clinton had her server hacked. Eventually such stream boils down to competition of who is more aggressive, rude and provocative against each other. Many such conversations ends with Hitler being injected, but of course no party will ever see the other side. I literally monitored 3 such streams of women wasting 10 days of their lives going after each other with frequency of one tweet per 45 minutes. Your idea to break it down and sort of "mediate" between two people that honestly are not interested in even hearing what the other side has to say, but are only engaged to vent off and blindly cheer whatever cool aid they been already sold on long time ago (i.e. that Trump is incredible president, not a con man, or Clinton is amazing women, not a crook) won't work. Second and finally: > My guess is that only whitelisting makes sense and with a granularity lower than instances. Twitter has signed off its death sentence when it went public. Its a tectonic shift within your organization when you move from serving users for the good sake of the whole system, into serving stockholders that only spend 15 minutes with you, once a year at the end of financial period and will only look into one-page accounting statements of profits and loss. While twitter users and even owners may want the most happy, prosperous and healthy environment in which everyone thrives and enjoy spending their evening reading and responding to tweets, the truth is the only people Twitter answers to is stockholders who only wants to squeeze as many ads into your timeline as possible. I'll say it even more: if Twitter starts fixing its feeds and doing further account purges, they will get sued by stockholders because obviously less users on platform equals to less ads-engagement. And its irrelevant whether you engaged with trollbots or real genuine people; it only matters you seen (and hopefully clicked) on ads. Does this sound about right?

lapinot 8y ago
Yeah, not sure. For the first point, actually i don't really care: eg the action i would like to take is simply to ignore the whole thread for myself (and thus in a sensible distributed system where people might make their policy depend on mine because they trust me, it would also lower the rank of this thread and in the end isolate it in some limbo). And after some time, one can hope that a long-term effect of discouraging abuse, shitstorm and humor/meme buzz will educate users and make "interesting" uses take over (eg these people will either get off this particular network or stop participating in flame wars because there are less of them and that there is no social reward on the network). And for the second point, i completely agree with you: there is no way that twitter makes the right thing. But i was mainly talking about what "a hypothetical well-run probably federated or even fully distributed social network" (possibly mastodon but not restricted to it).
- joering2 8y ago
  > And after some time, one can hope that [...] because there are less of them and that there is no social reward on the network). That's a wishful thinking! One could say the same about first color TV when first comedy serial was aired; that eventually people will stop wasting their time watching silly shows (point given you cant argue with TV-set) and eventually move on to be more productive etc. To the contrary I don't think we seen a mountain top just yet; still I meet people who don't know what Twitter is; their path to being daily engaged and being dragged into trollbots' wars is probably few years down the road.
  - lovich 8y ago
    Eh, I've met a few people who had only heard of the name and wondered what it was all about. When I showed them how to sign up and look for hashtags, they quickly gave up between the flame wars and the bad ui. I don't think Twitter is going to reach anywhere near the ubiquity of tv or movies for entertainment. Most tv shows and movies don't have a good chance of making you feel angry, frustrated, or depressed after watching them