Free Speech, Identity and Mastodon

There’s a new social nework on the block: Mastodon.
Or rather, it’s a social media technology. When we funnel all our conversations through the servers of a big company like Facebook or Twitter, we grant them enormous power. They control the extent of our privacy and of our free speech, and that power can be abused in ways that are both legal and not. The companies can sell our data to third parties (a process made much easier by the US Congress last week); they can reveal our data to the security agencies of nefarious regimes; and they can throttle or shut down our free speech if they so desire, without going via a court.
Decentralising the way in which we converse online means we can reclaim some of that power. A few years ago I posted a link to a blog post on Dave Winter’s Scripting News which sets out the practical and political importance of this idea: by spreading out, we’re harder to stop.
Mastodon is an open source project, so anyone can install it on a server and run a Mastodon ‘instance’. The software uses a principle called ‘federation’ to allow users to see messages posted on other instances of the software. So people who signed up on (say) mastodon.social can view and respond to messages posted to octagon.social (which is the version I signed up to with the username @robertsharp).
Problem solved, then? Not really.
In an interesting Medium post, John Henry says that Mastodon is ‘dead in the water’ because it doesn’t quite solve the issues of identity and moderation that are probably the most difficult problems on the web right now.

• Identity is crucial if you want to build a following, user-base or simply a network for your messages (whatever they may be)
• Moderation is crucial for the health of the conversation and the success of the community you are trying to build… but of course different communities need different policies. Moderation is incredibly time intensive work and even Facebook and Twitter are failing in this area. Trolls flourish while legitimate content gets taken down.

Henry suggests that Mastodon (in its current form) will fail because entire instances can be blocked from the federated network, breaking the connections between swathes of people. The network will never reach global proportions without the moderation and identity problems becoming critical. He also lists the problems with all the other kinds of identity solutions that companies and individuals have come up with. None are perfect.
While the problems John Henry identifies are real, I don’t see why that should mean Mastodon is D.O.A. Imperfect software still gets adopted and can become wildly popular. Facebook and Twitter thrive despite their faults. While we complain about their moderation approach and their propensity to allow ‘fake news’ into the ecosystem, no-one seriously suggests that we shut down the sites and return to extremely limited methods of communication that existed prior to their invention.
In the case of Mastodon, the limits of the federation model might be a strength. A semi-permeable social network, which restricts access to certain instances, might be perfect for children and schools, for example. And of course there will be iterations to the software that improve on the current solution and perhaps give more control to individual users.
I don’t pretend to have all the answers (indeed, any of the answers) to this problem. This blog post is really just an extended bookmark for John Henry’s post, which is enjoyed because of the way it swings wildly between an exceedingly technical discussion about internet protocols, and broad philosophical ideas about freedom of expression. It shows us how seemingly innocuous technical issues, which few people understand and even fewer people are working on, can have huge implications for democracy.