First Experience with Mastodon

Recently, I had a more in-depth experience with a product called "Mastodon". Regarding the name, there was actually a mix-up. Originally, it was called Mastodon (Mastodon), but it was mistakenly translated as "长毛象" (Long-haired Elephant). By the time the mistranslation was discovered, it had already spread, and this translated name was quite cute, so it continued to be used.


Mastodon is a social platform similar to Weibo and Twitter. It has all the features of dynamic information flow, comments, reposts, likes, bookmarks, private messages, lists, etc. However, Mastodon has a fundamental difference from them—it is a federated product without a central server. Due to its open-source technology, anyone can create their own site (officially called an instance). Different instances can communicate with each other because they are based on the same ActivityPub protocol.


This may still be somewhat abstract, so let me give you a simple example: suppose you follow some friends who are active on other platforms, and these platforms also support ActivityPub. In that case, you may see updates from your ex-girlfriend on Weibo and videos from your ex-ex-girlfriend on Bilibili on your homepage. Isn't it magical to think about?

You don't have to register for a bunch of platform accounts just to follow certain people. It gives you a feeling of "having one account in hand, traveling the world". The social world that Mastodon wants to create may be like this.

Today, based on its open-source technology, hundreds of instances have been created worldwide, and over eight million registered users are active in various instances.

Why Pay Attention to Mastodon?#

Actually, I noticed this product a few years ago, but it wasn't very popular at the time, so I didn't pay much attention to it. However, recently it has gained popularity again, thanks to Elon Musk's support: the acquisition of Twitter triggered a wave of migration, and a few days ago, a measure to ban competitors was introduced (which seems to have been withdrawn), once again causing a large-scale migration. In addition to Musk, the increasingly strict censorship in China has also forced a large number of users to flee to Mastodon.

Initial Observations and Thoughts#

After observing for a while, I found it quite interesting. Here are my observations:

  1. User Base and Growth
    Currently, there are over 8 million registered users, with around 30,000 new registrations per day (data from official bots), and about 2 million monthly active users. The registration growth rate looks optimistic, but for a social product that relies solely on word-of-mouth without much promotion, the future of Mastodon is still worth paying attention to.

  2. Observations on Instance Dynamics, using the largest Chinese instance, "草莓县" (Strawberry County, with 16,000 active users), as an example:

    1. User Dynamics: After observing for a few days and sampling over 300 posts, the average posting frequency is about 4 posts per minute. Considering the lower frequency at night, the overall average is about 2 posts per minute. Most of the posts are made by highly active users, with an average of 2 posts per day per person. Therefore, it can be inferred that the approximately 2,880 posts per day in Strawberry County are mainly contributed by around 1,440 active users (excluding comments and other interactions). Based on the observed posts, the main types of posts are: daily life updates > current affairs commentary > book and movie recommendations. It is also observed that over 50% of the sampled users are from Douban (a Chinese social platform for books, movies, and music) and Twitter, indicating that they are "refugees" from other websites.
    2. Characteristics of Popular Posts: After observing over 100 popular posts, the ratio of post types is approximately: opinions (political/social events): daily life updates = 80:20. The factors that contribute to posts becoming popular are likely related to interactions, so users with more followers have an advantage. However, the preferences of users on the platform are also an important factor. From the categories of popular posts, it can be seen that users on the platform tend to criticize government actions and focus on social events. This confirms the previous observation that users have migrated to Mastodon to escape excessive censorship on other platforms.
    3. Trends: The participation in popular topics is approximately 10-30 people every 2 days. For example, within the past 2 days, only 27 people participated in the topic of the winter solstice. This indicates that users on the platform have a relatively low level of engagement with topics. They tend to post updates in their own way. However, it also indicates that the Strawberry County community has almost no operational activities and is entirely driven by user-generated content.
  3. Business Model

    1. Currently, there is no mature commercialization model, and instances are generally of a non-profit nature. The operators of different instances are usually funded by foundations, crowdfunding, or personal contributions.
    2. The currently mature advertising revenue model may be feasible on Mastodon, but its effectiveness will not be very good. The fundamental reason lies in the platform's decentralized mechanism, which makes it difficult to distribute traffic and provide precise ad recommendations.
    3. Subscription fees may be a way forward. It depends on what Mastodon will develop into in the future and whether the value it provides will be compelling enough for users to pay, similar to current email services or phone services.

What Do I Use It For?#


Mastodon is suitable as a "tree hole" where you can share stress-free updates. I generally use it to post some rants, random thoughts, and book/movie recommendations.

Speaking of book/movie recommendations, I have already exported all my data from Douban using the "豆伴" (DouBan Companion) tool and imported it into the NeoDB instance. Let me introduce NeoDB, a tool for tagging books, movies, music, and other things. It is designed for "federated universe" platforms like Mastodon and supports logging in with a Mastodon account. It also supports synchronizing tagged book/movie updates to Mastodon.


If you are just starting to use Mastodon, the decentralized and algorithm-free design makes it difficult to have a smooth start. Finding interesting people to follow becomes a challenge. Here's a method I suggest:

  1. Go to NeoDB and find books/movies you like, then check out users who have similar tags. Visit their profiles and follow them.
  2. Check the follow lists of the users you follow to find interesting users to follow. (Note: Some users may hide their follow lists.)


I believe Mastodon is an idealistic product that is user-friendly but not operator-friendly. In the short term, it is unlikely to replace Twitter, mainly due to its decentralized design, which naturally leads to difficulties in cold-starting and commercialization.

However, in the long run, I think it may create a different form of social networking—users can encounter interesting people without having to register for additional accounts on different platforms. Of course, this premise is that there are a sufficient number of users, especially influential users, and more and more communities and tools support the ActivityPub protocol. However, the biggest question is why big companies would do something that harms their own interests. If a way to achieve a balance of interests can be found, the above scenario may become a reality.

Ownership of this post data is guaranteed by blockchain and smart contracts to the creator alone.