Opinion: Here’s a simple way Elon Musk could make Twitter better

Final month, Twitter consumer Qasim Rashid tweeted the next:

These numbers aren’t correct. The typical worth of West Texas Intermediate crude oil in June 2008 was $134, not $181.58. In March 2022, it was $108, not $99.76. Gas prices have been $4.05 in June 2008 and $4.22 in March 2022. So the markup on gasoline has elevated modestly since 2008, however not almost as a lot as this tweet suggests.

Even so, Rashid’s tweet has racked up 18,000 retweets. As of publication time, it’s nonetheless on Twitter.

Tweets like this one are on my thoughts as I take into consideration Twitter’s Monday announcement that it had accepted a deal for Elon Musk to purchase Twitter for $44 billion.

“Free speech is the bedrock of a functioning democracy, and Twitter is the digital city sq. the place issues very important to the way forward for humanity are debated,” Musk stated within the press launch saying the acquisition.

In recent times, Twitter has developed an more and more elaborate system for eradicating varied forms of dangerous and low-quality content material from Twitter, corresponding to hate speech, vaccine misinformation, and former President Donald Trump’s tweets tacitly endorsing the January 6 riot on the US Capitol.

Rashid’s tweet apparently doesn’t run afoul of any of Twitter’s guidelines. However garden-variety misinformation clearly isn’t useful to a functioning democracy.

Conversations about this challenge have a tendency to interrupt down alongside now-familiar partisan strains, with of us on the left demanding that social media platforms do extra to battle misinformation and hate speech and folk on the best decrying that as censorship. Musk has thrown his weight behind the free-speech facet of the argument; there’s little likelihood that Twitter will do extra content material moderation with Musk on the helm.

However there are alternatives different than simply taking down misinformation or leaving it up. place to begin can be for Twitter to work tougher to not actively promote misinformation. That oil tweet wound up with 18,000 retweets as a result of Twitter is designed to maximise the distribution of extremely “participating” tweets. And interesting tweets are sometimes dangerous tweets.

The difficulty with algorithmic information feeds

After I joined Twitter in 2008, the location confirmed customers each tweet by individuals they adopted in strictly chronological order. In 2016, Twitter introduced a brand new algorithmic feed that prioritized tweets Twitter thought customers have been more likely to care about. This alteration met vital resistance from customers, and Twitter initially portrayed it as non-compulsory. However over time, Twitter has more and more pushed users to switch. As we speak, the algorithmic feed is the default view.

It’s straightforward to see the shift as an innocuous enchancment to the consumer expertise. If Twitter is aware of which tweets I’m more likely to discover most attention-grabbing, why not present these first? However the swap had profound penalties for the type of platform Twitter would turn out to be.

In 2015, I had sufficient Twitter followers that I might depend on each tweet getting no less than just a few reactions from followers. Some tweets obtained extra reactions than others, and I normally hoped that my tweets would “go viral.” However my essential motivation was to share stuff I assumed was attention-grabbing with my direct followers.

However just a few years later, I observed a rising variation within the degree of response to my tweets. If I wrote a few extremely participating subject (say, US politics), I might typically get a bunch of likes and a few retweets. But when I tweeted a few much less thrilling subject, engagement can be very low. Generally, I’d tweet and get no reactions in any respect.

The primary few instances this occurred, I questioned if I’d written an particularly boring tweet. However now, I feel the extra seemingly clarification is that hardly anybody sees these types of tweets. As soon as Twitter’s algorithm decides a tweet isn’t participating sufficient, it stops placing the tweet into individuals’s newsfeeds.

The sensible result’s that Twitter’s software program is “coaching” all of us on the type of tweets to put in writing. No person prevents us from writing tweets on non-engaging matters, however once we do, it’s like shouting right into a void. Over time, we study to put in writing in a extra “participating” approach—which regularly means writing tweets which are partisan, inflammatory, or pandering to the biases of our current followers.

And since a lot of our public discourse occurs on Twitter, I feel the transfer has had a non-trivial affect on our political tradition. Twitter is feeding individuals tweets that affirm their current biases and make them indignant or fearful. Once we see tweets from the “different facet,” it’s typically somebody saying one thing outrageous, accompanied by dunks from our personal facet. We’re much less more likely to see tweets that problem our prejudices or introduce us to matters we didn’t know we have been occupied with.

This fundamental perception isn’t new, in fact. It has been a typical criticism of social media since no less than 2010, when creator and activist Eli Pariser coined the time period “filter bubble” to explain the phenomenon. However the rise of algorithmic feeds over the past decade has made the issue a lot worse. A standard prescription for escaping filter bubbles is to intentionally observe individuals with ideological views completely different from your personal. However this doesn’t assist if Twitter’s algorithm notices you don’t interact very a lot with their tweets and stops displaying them to you.