Lately, I've been toying with Mastodon, an open source platform that's been positioned as a challenge to Twitter's hold on the market (such that it is) for micro-blogging. There was an initial flurry of interest back when the project launched a few years ago, but the relative complexity of setting up an account kept most people off. Since then, Mastodon has grown much more accessible and -- particularly since Twitter has made itself an exception to the general collapse of tolerance for Infowars -- more appealing to social media users looking for an alternative.
A major selling point for many is the relative paucity of white nationalists (or, if you prefer, Nazis) on Mastodon, which is only sort of the case. It's not that Mastodon takes a harder line on hate speech and racial animus -- the nature of the software prevents that -- but rather that, because there's no one single authority determining policy for the platform, independent administrators can effectively dissociate their servers from the servers that host hate groups. That may not be a perfect solution, but to many people who've been burned by Twitter's notoriously flawed policies on abuse, it's a vast improvement.
In truth, though, I don't think Mastodon is cut out for direct competition with Twitter, despite their prima facie similarities. The differences are largely the result of a number of thoughtful design decisions, like the one that makes possible the strategy of quarantining white nationalists from the more tolerant corners of the network. Though born independent of one another, Mastodon servers (called "instances") almost inevitably become linked to one another by a process of "federation." While the underlying principles are similar to the way that email serves interact with one another, federation on Mastodon is a much more overt process, and one that requires a degree of diplomacy between communities.
Along with federation, the contrast to Twitter that has me most interested in Mastodon is the trio of streams associated with each account. The one most akin to Twitter's unitary stream is "Home," which shows the messages posted by accounts you've elected to follow -- though, unlike Twitter's algorithmically adulterated timeline, Home is strictly chronological. But because each account is hosted by an instance, there's also a Local timeline, that shows all of the public posts make by accounts on the server. (In addition to public, Mastodon messages, unfortunately called "toots," can also be marked unlisted, private or direct, each resulting in a different degree of visibility.) More broadly, there's the Federated timeline, which shows all of the public messages broadcast by every account followed by every member of the local instance.
Many current instances are catch-alls, similar to Twitter in that they are less communities of interest than party lines bundling together everyone who happened to create an account on one server rather than another. But the Local timeline -- a feature afforded by the federated nature of Mastodon -- also opens up the possibility for genuine communities of interest of the sort that would be difficult to create or maintain on Twitter.
That, in part, is the plan for humanities.one, the Mastodon instance I recently launched. The initial idea was just to create a home for my personal account, one that revolved around a cluster of interests that weren't already being catered to by existing communities. (New social media platforms tend to attract STEM/tech types before the humanities set catches on.) On top of that, though, the goal is to set the conditions for a Local timeline where the talk reliably focuses on a specific range of topics, and so distinguishes itself not only from the Local timelines of its participants, and their joint Federated timeline, but also from the more general discussion found on the Local timelines of other instances.
The trick, it seems to me, is, first of all, to attract participants with a genuine interest in the designated topic (i.e. the humanities). Then, once they've created local accounts on the server, encourage everyone to mark off-topic messages as "unlisted." The result would be a Local timeline that, save for the occasional slip-up, is almost entirely concentrated on the humanities, while off-topic messages and banter show up only in the Home and Federated timelines.
Ultimately, the success of a platform like Mastodon may depend not on its capacity to challenge corporate-owned social media head-on, but on how it leverages the differences to foster different opportunities. The commercial incentives that drove the growth of Twitter don't necessarily apply when a company's brand presence can be excluded tout court by the administrator of an instance. But Mastodon may not need to attract commercial brands to build its own success. I can envision a future where academic departments, NGOs, think tanks, and other institutions launch their own instances, precisely because the structure of Mastodon affords better opportunities for separating out the proverbial signal from the noise.