Jeff Duntemann's Contrapositive Diary Rotating Header Image

social networking

Teergrubing Twitter

Jul 28th, 2020
by Jeff Duntemann.
8 comments

I’m of two minds about Twitter. Maybe three. Maybe seventeen. I’m on it, and I post regularly, typically two items per day. That’s just how it averages out; I don’t post for the sake of posting, but only when I find something worth linking to.

Why? Two reasons:

  1. More people are on Twitter than cruise blog posts. By posting on Twitter, I make more people aware of me than I do when I post things here on Contra. Every time I tweet a link to one of my books, I sell a few copies.
  2. It’s one of the most gruesomely fascinating phenomena to come out of tech since the Web itself, thirty years ago.

I’ve written about Twitter before. Back in 2019, my proposed solution to Twitter toxicity was to remove the retweet function. That would certainly help, but only to an extent, and not to the extent that I would like. I’ve spent more time on Twitter in the last 18 months since I posted that entry than I did in the 5 years before that. And in doing that…

…I’ve changed my mind.

I generally lurk, but occasionally I join a Twitter rumble to watch how it all unfolds. I never stoop to profanity or unhinged emoting. Here and there I have politely called a few people on their BS. In doing so I made an observation that bears on today’s entry: When I get involved in a ruckus, my follower count goes up. When I just post odd lots, my follower count decays. The reason is simple: Twitter has made itself into a sort of deranged video game. The Analytics panel shows you how many people have looked at your tweets, how many have mentioned you, how many followers you’ve gained or lost, and more. In the same way that Twitter as a whole is an outrage amplifier, the Analytics panel is a vanity amplifier. You have a “score.” The object of the game is to raise your score. And the best way to do that is to create or partake in a ruckus. In fact, the more rucki you launch or dive into, the higher your score will climb.

I’ve thought quite a bit about how Twitter would change simply by removing the Analytics panel, or any other stats on your activity. Even if Twitter would agree to do that (highly unlikely) it would reduce the number of neutrons only modestly. A ruckus feeds the ancient tribal impulse. Tribal Twitter is a game whether or not you have an explicit score.

Very briefly, I wondered how Twitter might change if the platform removed limits on (or at least greatly increased) the allowable size of tweets. Again, it would help a little by changing Twitter’s DNA to be a little more like Facebook or other social networks. Because it takes more time to write longer, more thoughtful entries, people would spend their energy doing that and not trying to destroy one another.

Maybe a little. Or maybe not. Which brings us to the heart of what I’m about to propose: slowing Twitter down. To return to the metaphor of nuclear fission, it would be about inserting a neutron moderator. I don’t mean a control rod (which eats neutrons) but something that merely slows them down and therefore reduces their energy.

I’m reminded of the spam wars before centralized spam suppressors appeared. The idea was to reduce the effectiveness of email spam by slowing the rate at which an email server would accept commands. There are several ways to do this, including sending nonsense packets to the system requesting connections. This was called tar-pitting, which translated directly to the charming German neologism, teergrubing. Spamming works by throwing out a boggling number of emails. Teergrubing fixes that by making the process of throwing out a financially workable number of spam messages too lengthy to bother with.

I don’t know to what extent teergrubing is done today. Doesn’t matter. What I’m suggesting is this: Build a delay into the process of accepting replies to any given tweet. Make this delay increase exponentially as the number of replies increases. First reply, one minute. Second reply, two minutes. Third reply, four minutes, and so on. Once you get up to the eighth reply, the delay is over two hours. By the time a tweet could go viral, the delay would be up in days, not hours. At that point, most of the original outragees would have lost interest and gone elsewhere. Most ordinary Twitter conversations generate only a few replies, and those would arrive in less than an hour, the first several in minutes.

Outrage addicts might try to finesse the system by replying to replies and not the original tweet. This would quickly reduce a ruckus to incoherence, since people trying to read the mess would not be able to tell what tweet a replier was replying to.

Or it might not work at all. Certainly Twitter would not consent to a change like this short of legal action. That legal action may someday arrive. My point is that the heat in any argument dies down when the back-and-forth slows down. Some few diehards may choose to sit out a several days’ delay just to get the last word. But if nobody reads that last word, having the last word loses a lot of its shine.

And Twitter is all about shine.

Taming Twitter

Jan 28th, 2019
by Jeff Duntemann.
9 comments

I knew Twitter was mostly useless before I ever got an account there. I got the account because the service seemed insanely popular, which I simply could not understand. My account is now four years old, and having mostly lurked in that time I think I finally understand what Twitter is for, and why it’s a problem.

This past week saw another instance of what many call a Twitter lynch mob: Hordes of tribalists, intoxicated with their own outrage, descended upon a group of Catholic high school boys who were waiting for a bus in DC when various kinds of hell broke loose around them. I won’t go over the details here; you can google as much as you like. The incident itself isn’t my point, and I will delete any arguments in the comments over whether they “deserved” the ill-treatment they got. (They did not. If you disagree, disagree in your own space, not mine.)

The point I’m actually making here is that this incident (and countless others like it) would not have happened without Twitter. I’ve been on LiveJournal since 2005, and on Facebook since 2009. I’ve never seen an online lynch mob on either service. I’ve seen plenty of arguments, some of them quite heated, a few of them absolutely insane. None of them “went viral” the way that Twitter lynch mobs go viral.

Part of the underlying problem is a lack of discipline among many journalists. Most of the money has gone out of mainstream news journalism over the past twenty years, and with it went the sort of disciplined, methodical reporting I took for granted before 2005 or so. When you have to get clicks to keep your job and pay your bills, “methodical reporting” means all the other starving journalists will get those clicks before you do.

But bad journalism is mostly an enabling factor. The real mechanism is Twitter’s ability to act as an amplifier of emotion. Until very recently, tweets were limited to 140 characters. That’s room enough to post a link to an Odd Lot (which is most of what I do) and not much else beyond quips, brief questions, quotations, short descriptions of photos and videos, and so on. This means that rational discussion doesn’t take place very often on Twitter. There just isn’t room. Sure, some people make their case using a number of independently posted tweets intended to be read in sequence. Megan McArdle of the Washington Post is very good at this. Alas, the process of creating such a thread sounds mighty tedious to me.

What’s left? Emotion. And what’s the emotion of the day, year, and decade hereabouts? Outrage. And while Twitter can amplify things like humor, cuteness, and gratitude (and occasionally real beauty) what it does best is outrage.

From a height, Twitter is an outrage amplifier. It starts with somebody posting something calculated to outrage a certain demographic. (Innocent posts sometimes trigger Twitter mobs, but they are uncommon.) Then begins a sort of emotional feedback loop: The outraged immediately retweet the reactions they’ve seen, so that their followers (who would not otherwise have seen the outrage tweet) get to see it. They retweet it to their followers, and so on, until millions of gasping outrage addicts are piling on without knowing anything at all about the original issue that caused the outrage.

The word “amplifier” may not be quite the right metaphor here. Most of us in the nerdiverse have seen videos of a common science demo consisting of a room full of set mousetraps, each with two ping-pong balls carefully placed on the bar. Toss a single ping-pong ball into the room, and it sets off whatever mousetrap it lands on. That moustrap launches two more balls, which set off two more mousetraps, and a few seconds later there is this chaotic cloud of ping-pong balls flying around the room, until the last mousetrap has been spring. This metaphor is a nuclear fission chain reaction, and I think it describes a Twitter mob very well.

So what do we do about Twitter mobs? We could encourage the victims to lawyer up and start suing the news organizations that tossed the original ping-pong ball, and perhaps Twitter itself. That process is evidently underway with the Covington Kids. But preventing Twitter mobs is simple, if difficult: All it would require is a single change to the Twitter software:

Eliminate retweets.

That’s all it would take. Really. The retweet function is like a neutron emitted by an unstable nucleus. (There are a lot of unstable nuclei in the Twitter system.) Chain reactions are easy to kick off, and difficult to suppress. But without the ability to instantly retweet some expression of outrage, the issue never goes critical. Sure, you can manually copy and paste somebody else’s tweet and tweet it to your own followers. But the sort of people who participate in Twitter mobs are impatient, and lazy. If copy/paste/tweet is work, well, their ADHD sends them on to something else.

Basically, eliminating retweets would turn Twitter from U-235 to U-238. U-238 is non-fissionable. Without retweets, Twitter would be non-fissionable too. Problem solved.

Of course, Twitter won’t voluntarily disable retweets. Without retweets, Twitter becomes just another microblogging social network. People would abandon it in droves. However, if a class action against Twitter mounted by victims of Twitter mobs ever got any traction, part of the settlement might include requiring Twitter to disable retweets. If I were the victim of a Twitter mob, that’s what I’d demand. Money wouldn’t hurt. But to fix the problem, retweets would have to go. If that in fact became the end of Twitter, I for one wouldn’t cry too hard.

Twitter is not a common carrier. It attempts to police its own content, though that policing is sparse and rather selectively applied. If it isn’t a common carrier, it can be held responsible for the actions of its members. If its members set out to deliberately destroy private citizens by retweeting slander and doxxing, Twitter should face the consequences. If it were forced to confront the possible consequences, who knows? Twitter might eliminate the Retweet button all by themselves.

Don’t wait up for it. But don’t count it out, either.

Odd Lots

Sep 29th, 2017
by Jeff Duntemann.
9 comments
  • I get asked several times a year: “What are your politics?” Tough question, given that I think that politics is filth. But now Jon Gabriel has answered the question for me: I do not join teams. I create my own. I’ve been doing this all my life. I’m not going to stop now.
  • Side note on Jon Gabriel: He used to work at Coriolis back in the day. So although I’ve been seeing him online for years, I never realized until very recently that he was our Jon Gabriel. (There is another who does diet books.)
  • Twitter is experimenting with doubling the size of tweets to 280 characters. I wonder if Gab had any least little bit to do with that?
  • Cirsova Magazine posted a short excerpt of something called the Denham Tracts from 1895 on Twitter, with a longish list of British supernatural beings, among which are “hobbits.” You can see the whole fascinating book on the Internet Archive. It was published by the Folk-Lore Society and it’s exactly that: Short notes on British folklore, including local saints, odd little ceremonies, song lyrics, and supernatural creatures I’ve never heard of, like the dudmen, wirrikows, gallytrots, miffies, and loads more. (The list starts on page 77.) Great fun!
  • At last, it looks like a popular treatment of sleep science is coming to us. Matthew Walker’s Why We Sleep will be released on October 3. This long-form piece provides some background. Walker is willing to say what I’ve been saying for decades: Do not short your sleep. Bad things will happen, including cancer, obesity, Alzheimer’s and who knows what else. Unlike me, Walker is an expert on the subject, so maybe you’ll believe him.
  • Lack of sleep can kill you. So, evidently, can low-fat diets, according a Canadian study of 135,000 adults in 18 countries, published by The Lancet. Note the reactions of NHS physicians, who aren’t convinced. (In their defense I will say that the Mayo Clinic is still pushing a low-fat diet in their newsletters.)
  • Here’s a long, rambling, but worthwhile discussion of how the fake science of fat demonization came about, and how, faced with the spectre of being shown to be wrong about something (impossible!) governments are doubling down on the fake anti-fat message. Government actions cause harm because we can’t throw the responsible parties in a cell and leave them there. The King, after all, can do no wrong.
  • Via Esther Schindler: The history of email.
  • I’d prefer that it be in Pascal, but so it goes: There is a Javascript code baby onesie. My grand-niece Molly is now a month old. Decisions, decisions…
  • In his will, philosopher Jeremy Bentham specified that he was to be mummified, dressed in his ordinary clothes, and put on display. So it was written. So it was done.