Conversation

Notices

1. Embed this notice
   Hilda Bastian (hildabast@mastodon.online)'s status on Friday, 13-Jan-2023 09:10:26 JST Hilda Bastian

   I'd fallen for, and helped spread, what I now realize was a myth about quote tweets: That there's almost no research on them. Big mistake! Now rectified.

   Here's my new post, digging into more than 30 studies with data on quote tweets. They don't settle all the questions. But there are answers, or at least partial ones, to most. It's still complicated, though.

   It's a * very * long read, but there are summary points ....1/2

   https://absolutelymaybe.plos.org/2023/01/12/quote-tweeting-over-30-studies-dispel-some-myths/

   #MissedQuoteBoost #QuoteBoost #QuoteTweets

   • clacke likes this.
   • clacke repeated this.
   • Embed this notice
     Jo (jo@blahaj.zone)'s status on Friday, 13-Jan-2023 09:10:18 JST Jo

     in reply to

     • Russ Belville 🇺🇸🌲🎙🎸🌹

     @RadicalRuss@mstdn.plus QTs are available on instances that run #calckey #akkoma etc. @hildabast@mastodon.online

     ThatOneCalculator :calcdumpy: :calckey: likes this.
   • Embed this notice
     Russ Belville 🇺🇸🌲🎙🎸🌹 (radicalruss@mstdn.plus)'s status on Friday, 13-Jan-2023 09:10:19 JST Russ Belville 🇺🇸🌲🎙🎸🌹

     in reply to

     @hildabast This is exactly the kind of post I would QT by telling my readers "Here's a great post on QTs. It's long, but generally supports my position that QTs are more good than bad."

     But instead, I have to write this as a reply, then boost it.

   • Embed this notice
     Carl T. Bergstrom (ct_bergstrom@fediscience.org)'s status on Saturday, 14-Jan-2023 17:24:00 JST Carl T. Bergstrom

     in reply to

     • Philip N Cohen
     • Drew Harwell

     @philipncohen @hildabast @drewharwell

     It also seems critical to be able to control quote tweets—without locking your account— as illustrated by this recent Scott Gottlieb thread. (CW: quote tweets of Scott's post include some really nasty stuff). He turned off replies, and people uses the quote-tweet feature instead to instigate a particularly venomous pile-on.

     https://twitter.com/ScottGottliebMD/status/1612548694762745856

     clacke likes this.
   • Embed this notice
     Carl T. Bergstrom (ct_bergstrom@fediscience.org)'s status on Saturday, 14-Jan-2023 17:24:01 JST Carl T. Bergstrom

     in reply to

     • Philip N Cohen
     • Drew Harwell

     @philipncohen @hildabast @drewharwell

     Hilda, this was a great post. Thank you.

     It's good to see that the majority use of quote tweets is not particularly problematic.

     At the same time, the median use of the feature may not be our biggest concern. Even if most quote tweets are mostly used in constructive ways, I believe that many public health professionals on twitter have received death threats when their posts were quote-tweeted by big right wing accounts. That's a serious problem.

   • Embed this notice
     Philip N Cohen (philipncohen@mastodon.social)'s status on Saturday, 14-Jan-2023 17:24:02 JST Philip N Cohen

     in reply to

     • Drew Harwell

     @hildabast @drewharwell thanks for sharing. I have been victim of a number of pile-ons that started with a famous person quote-tweeting me, but they've also done it in replies and with screenshots. I doubt the QT function itself was dispositive. On the plus side, of course, QT is a good tool for high-follower accounts to boost others, and bring more people to the discussion in a way that helps them build their audience.

   • Embed this notice
     Jon (jdp23@indieweb.social)'s status on Saturday, 14-Jan-2023 17:24:08 JST Jon

     in reply to

     @hildabast Also I updated the footnote in https://privacy.thenexus.today/black-twitter-quoting-and-white-toxicity-on-mastodon/ so now it links to your article. Recursion FTW!

     clacke likes this.
   • Embed this notice
     Jon (jdp23@indieweb.social)'s status on Saturday, 14-Jan-2023 17:24:10 JST Jon

     in reply to

     • Philip N Cohen
     • Carl T. Bergstrom
     • Drew Harwell

     Yeah really, great article @hildabast !

     And agreed that even though people who actively want to harass somebody will just do a screenshot-and-link if they can't QT, there's still potentially value in controlling who can QT and establishing norms around screenshot-and-link.

     Good to hear that they're finally implementing turning of replies, that's long overdue! I think the original bug was filed in 2018 or something like that.

     @ct_bergstrom @philipncohen @drewharwell

   • Embed this notice
     Hilda Bastian (hildabast@mastodon.online)'s status on Saturday, 14-Jan-2023 17:24:11 JST Hilda Bastian

     in reply to

     • Philip N Cohen
     • Carl T. Bergstrom
     • Drew Harwell
     • John (he/him) :vbike:

     @ct_bergstrom @philipncohen @drewharwell Yes, that's what Rochko's proposing - they're developing the turn off replies & it would use the same process.

     I quote a post at the end of my post that's important reading on this: https://write.as/collisiondomain/some-thoughts-on-quote-boosts by @cd24 Because you can circumvent it all with screenshots. Once there's a QT feature, he points out, the use of screenshots can become a flag to moderators to investigate.

   • Embed this notice
     Nika Shilobod (nikashilobod@fediscience.org)'s status on Saturday, 14-Jan-2023 17:24:19 JST Nika Shilobod

     in reply to

     • Philip N Cohen
     • Andrew Perfors
     • Carl T. Bergstrom
     • Drew Harwell

     @ct_bergstrom
     @hildabast @philipncohen @drewharwell @AndyPerfors

     People would probably chain accounts if there was a cap and it would give the illusion that more people agree with something than the reality in the long run. It would create a lot of problems.

     clacke likes this.
   • Embed this notice
     Carl T. Bergstrom (ct_bergstrom@fediscience.org)'s status on Saturday, 14-Jan-2023 17:24:20 JST Carl T. Bergstrom

     in reply to

     • Philip N Cohen
     • Andrew Perfors
     • Drew Harwell

     @hildabast @philipncohen @drewharwell

     "The problem is the type of user who issues death threats & the type of user who attracts large numbers of them"

     I 100% agree with you there.

     It does make me wonder about whether it would be useful to implement a proposal such as @AndyPerfors's to cap the number of followers that any one account can have.

   • Embed this notice
     Hilda Bastian (hildabast@mastodon.online)'s status on Saturday, 14-Jan-2023 17:24:21 JST Hilda Bastian

     in reply to

     • Philip N Cohen
     • Carl T. Bergstrom
     • Drew Harwell

     @ct_bergstrom @philipncohen @drewharwell Yes, it's a serious problem - but they only have to retweet something for it to be. The problem is the type of user who issues death threats & the type of user who attracts large numbers of them, & the hashtags that network them. That's a moderating issue, because they can do it with or without the QT. Indeed, it's often with screenshots in case the person deletes their tweet/post.

   • Embed this notice
     Nika Shilobod (nikashilobod@fediscience.org)'s status on Saturday, 14-Jan-2023 17:24:23 JST Nika Shilobod

     in reply to

     • Philip N Cohen
     • Andrew Perfors
     • Carl T. Bergstrom
     • Drew Harwell

     @AndyPerfors
     @ct_bergstrom @hildabast @philipncohen @drewharwell

     It would be fairly easy to create a bot that just repeated one account. People could create servers of them by the hundreds. It would be fairly easy to make a bot network here just like Instagram. It's full of them.

     clacke likes this.
   • Embed this notice
     Andrew Perfors (andyperfors@aus.social)'s status on Saturday, 14-Jan-2023 17:24:24 JST Andrew Perfors

     in reply to

     • Philip N Cohen
     • Carl T. Bergstrom
     • Drew Harwell
     • Nika Shilobod

     @NikaShilobod @ct_bergstrom @hildabast @philipncohen @drewharwell

     Yes, as I thought about it more, I think while I’m still convinced of the underlying problems that come with power-law distribution of follower numbers - I’m less convinced about the solution of just placing hard limits.

     That said, the “chaining accounts” thing isn’t the main problem with it IMHO. It would still make a big difference for mega accounts. e.g. consider someone with 1M followers. if follower numbers were capped at 10K you’d need 100 different accounts to get the same numbers and I don’t see how that is feasible at all for a person do to. Much less someone with 10M followers.

   • Embed this notice
     Hilda Bastian (hildabast@mastodon.online)'s status on Saturday, 14-Jan-2023 17:24:27 JST Hilda Bastian

     in reply to

     ...My take now? Making quoting easy is not a major vector for abuse or toxicity, and there are benefits.

     That’s not to say that the quote tweet can’t be abused. Every form of tweet/post can be, and is.

     The toxicity was already there, thriving in replies in particular, and amplified in all its forms by the algorithm, retweets, and hashtags.

     Did the quote tweet make it worse? If it did, it may not have been by much - and in a couple of datasets, it improved civility.
     2/2

     clacke likes this.
   • Embed this notice
     myrmepropagandist (futurebird@sauropods.win)'s status on Saturday, 14-Jan-2023 17:24:32 JST myrmepropagandist

     in reply to

     @hildabast Wow, science. Thanks for making this custom lit review for the fediverse community. Really interesting studies— but also helps illuminate the issue.

     As for why #qt would matter so much to #blacktwiter — well it’s the transition thing again but also trust. It really feels like trying to text with mittens on without it.

     (@BlackMastodon #blackmastodon see up thread.)

     clacke likes this.
   • Embed this notice
     Hilda Bastian (hildabast@mastodon.online)'s status on Saturday, 14-Jan-2023 17:24:37 JST Hilda Bastian

     in reply to

     • Jon

     @jdp23 Thanks! Called it "more than 30" because I didn't want it be wrong quickly - and I was right: authors of a couple of studies have already pinged me. Will update.

     clacke likes this.
   • Embed this notice
     Jon (jdp23@indieweb.social)'s status on Saturday, 14-Jan-2023 17:24:38 JST Jon

     in reply to

     @hildabast looks great ... thanks for doing all that digging!