Indie Microblogging by Manton Reece

Popularity contests

“They’ve got you looking for any flaw, that after a while that’s all you see. For what it’s worth, I’m here to tell you that it is possible.” — Vincent from Gattaca

I wasn’t sure when first introducing Micro.blog without any follower counts whether it would work. Twitter has follower counts prominently shown on a user profile, and every Twitter-like service from Pownce to Mastodon has copied it.

But I had a theory that showing these counts brought an extra level of judgement. Does someone have only a few followers? Our first reaction is that maybe that person isn’t very interesting or worth following. But there are a lot of reasons for the number of followers, such as how long someone has been on the service.

I was relieved that it worked. Many users appreciated that they could choose to follow new users based on their interest in that user’s posts alone, not because they were in a popular list that everyone sees.

An article for Wired by Felix Salmon covered the influence of Twitter power users, celebrities who have accumulated massive followings:

Twitter is becoming increasingly concentrated on a tiny core of power users. It’s less and less a distributed mode of many-to-many communication, and more and more a broadcasting hub for the elite—a highly unequal place where their least-considered, Ambien-addled opinions get amplified to a global audience of millions.

The founders of Twitter courted celebrities early in the company’s history. Snoop Dog visited the Twitter HQ. Obama visited too. It was like a measure of success, giving validity to the service.

But increasingly users are realizing that they are happier to ignore follower counts. Rather than new year resolutions, Aleen Simms had a list of liberations for 2018: things to let go of and not worry about. In particular I like this one about not looking at numbers:

Twitter followers, podcast download stats, blog post views, the scale, whatever. Life isn’t a video game. Happiness doesn’t have a numerical value attached to it.

Now this is backed up by research too. In studies led by The University of Texas, researchers collected data on how receiving fewer “likes” affected adolescents emotionally:

Study participants helped test drive a new program that allowed them to create a profile and interact with same-age peers by viewing and “liking” one another’s profiles. Likes received were tallied, and a ranking of the various profiles displayed them in order of most to least liked. In actuality, likes were assigned by computer scripts. Participants were randomly assigned to receive either few likes or many likes relative to the other displayed profiles. In a post-task questionnaire, students in the fewer likes group reported more feelings of rejections and other negative emotions than those who received more likes.

It mirrors a philosophy we have with Micro.blog to launch without follower counts or public likes. Follower counts are not very useful for a new platform. They add anxiety and unavoidably lead to value judgements when considering whether to follow someone, instead of letting the quality of someone’s writing and photos speak for itself.

Andy Flisher posted to his microblog about how Facebook takes the opposite approach, encouraging the numbers game:

Facebook has genuinely ‘encouraged’ me to get more likes for my birthday than I did last year! 🤯 No wonder the youngsters are growing up so needy for likes and follows, not healthy 🙁

A social network doesn’t have to be like this. Micro.blog is a way to post to a web site that you control, and a place to discover and talk with other members of the community. Micro.blog is not a popularity contest.

This approach may become mainstream instead of the exception. The photo-sharing iOS app Glass launched in 2021 without public likes. In an interview with Om Malik, Glass co-founder Tom Watson said:

We’ve intentionally avoided any public counts. We don’t want Glass ever to become a popularity contest. We’re not home for influencers. We are a home for photographers.

Glass can do this because they are a paid subscription service. They need to make customers happy, but don’t need to focus on the kind of increased engagement that ad-based companies are always chasing.

Popularity contests are also easy to game. Researchers from the NATO Strategic Communications Center of Excellence ran a series of experiments to buy likes, comments, and clicks on social media posts. They paid companies in Russia and Europe hundreds of dollars to buy thousands of likes and followers, writing up a report on the results:

But the report also brings renewed attention to an often overlooked vulnerability for internet platforms: companies that sell clicks, likes and comments on social media networks. Many of the companies are in Russia, according to the researchers. Because the social networks’ software ranks posts in part by the amount of engagement they generate, the paid activity can lead to more prominent positions.

The researchers then actively notified the social media companies about the fake likes and tracked what action the tech companies took, if any. Most fake likes and accounts used for the experiment remained online a month after they were reported.

It is very difficult for a massive platform like Facebook or Twitter to catch everything. Instead of trying to “fix” fake likes that are purchased, the solution is to remove the reason someone would purchase likes to begin with. If like counts weren’t featured so prominently and used for surfacing content, there would be no incentive to try to game the system.

Instagram is also now testing hiding “likes”. Wired covered this with comments from the head of Instagram, Adam Mosseri:

Although Mosseri was careful to note that “bullying predates Instagram" and the internet, he did mention further measures that the platform is taking to improve the mental and emotional health of its users. The company is working with therapists and engineers to develop other tools to prevent and de-escalate bullying on the platform, such as figuring out a way to make users take a break when they need it.

This is good news and undoubtedly worthwhile for Instagram to experiment with, but it’s unlikely to be prioritized by Facebook leadership. Helping users “take a break” is at odds with their business of ads, fueled by measuring engagement and daily active users.


Tied in with popularity contests and follower counts is trending posts. Not just specific users or specific posts that are popular, but entire groups of related posts.

Brian Feldman, in an article for New York Magazine written after the Parkland shooting,

The first problem with “trending” is that it selects and highlights content with no eye toward accuracy, or quality. Automated trending systems are not equipped to make judgments; they can determine if things are being shared, but they cannot determine whether that content should be shared further.

YouTube’s trending algorithm surfaced a right-wing conspiracy theory video that was seen by hundreds of thousands of people before YouTube moderators took it down. Without the trending algorithm accelerating its spread, it would have been seen by fewer people.


Avoiding the numbers game in Micro.blog has also shined a light on other parts of the UI that can be rethought. Wherever there’s a number in the UI is an opportunity to question why it’s there.

Sometimes there’s a good reason. Unlike the count of people who follow you, which isn’t something you can control, how many people you follow is up to you, so there’s no harm in showing that number. But by paying attention to it, we found that even that could be deemphasized and improved.

If someone was following 500 people on Micro.blog, it used to include a “following 500” link in that persons profile. Clicking it would show a list of everyone that person was following.

But what someone is really after is to discover new people to follow. If out of that list of 500 people, you’re already following 400 of them, wading through the full list is a waste of time. So now the link shows “100 users you aren’t following”. It deemphasizes the total count and instead makes the feature much more useful.


The Atlantic, quoting Ben Grosser, after news that Jack Dorsey is reevaluating whether Twitter should even have likes:

“Part of what’s happening in spread of disinformation is that people can essentially repeat what someone else said and spread it to the world, the retweet has an effect well beyond the Like in that regard,” he said. Grosser also indicated that removing just the like button would only make the retweet more powerful. “I fear that if they remove the Like button the fact that there are other indicators that include metrics will just compel users to use those other indicators,” Grosser said.

Removing likes and retweets is of course old news to anyone who has been on Micro.blog. I wrote 2 years ago about the potential harm of retweets. It’s a common theme in the talks I’ve given and in other blog posts since.

When I first started rolling out Micro.blog to early supporters, not having public likes, retweets, or follower counts was a kind of controversial, risky decision. Now it’s almost boring. We’re in the middle of 2 complementary transitions: a move away from massive social networks, and smaller platforms providing the flexibility to remove features and algorithms in service to the community.


If we don’t algorithmically recommend users to follow, what do we do? There’s actually no shortage of ideas to help highlight users on Micro.blog when you think about human creation instead of algorithms.

Originally inspired by the “Follow Friday” tradition on Twitter, where Twitter users @-mention other users to follow, our community manager Jean MacDonald proposed Micro Monday. The idea was to encourage user recommendations with some context about why, not just a list of accounts to follow:

We suggest you make just one recommendation per week. Include a link to the account micro.blog/username to make it easy for people to click and follow, whether they see your recommendation on the Micro.blog timeline or on your blog. We highly recommend you give a short description of the reason for your recommendation.

Micro Monday has evolved into 3 separate things with the same name, each another way to promote what users are doing on Micro.blog:

  • Micro Monday user recommendations, where people can recommend someone to follow.
  • The podcast, where Jean interviews a member of the community about how they learned about Micro.blog and what they blog about.
  • The weekly email newsletter, where we link to the latest Micro.blog news and quotes from posts on Micro.blog.

We don’t do all of these every week, but when we do they serve as another human-curated view into activity on Micro.blog. It helps give recommendations more personal touch.

Any post that mentions Micro Monday is also collected in a special section in Discover. This is a good way for brand new users to scroll back through recommendations. And there have been similar community efforts such as as Smokey’s This Week in Micro.blog post.

Jean also brings people together around shared interests by starting “roll calls” on certain topics. People can reply if they are interested, and then by viewing the conversation you essentially get a list of users to consider following, people who might share some of the same interests as you.

It’s often said that the IndieWeb isn’t about protocols and tools. It’s about people. It’s about personal web sites and the interaction between people writing on their own blogs.

Welcoming new users by hand will always feel more meaningful than an algorithmic suggested users list. No one wants to talk to a computer.

Next: Banning users →