[betwixthewires]

One thing I think ignored in the article is the rise of FOSS federated servers. There are a few prominent examples to those who follow these sorts of developments, but there are two in particular that I find interesting: Lemmy and Littr.me.

Lemmy is basically a clone of reddit's functionality that federates over ActivityPub. It is a very high quality piece of software. A single server can host multiple communities and users, just like reddit. You can follow and comment on posts and communities on external servers using your account on whichever server you call home.

Littr.me is more like hackernews, a single community server, but one that will federate with other servers using the ActivityPub protocol, like Lemmy. The development is slower, but it is still interesting and the maintainer seems to be committed to building it out. Something like this allows what the author of this article talks about, single community sites for different niche communities, but also allows for intercommunity interaction.

With either of these, a user can tailor their feed to subscribe to any number of communities on any number of servers, and lurk, or interact if they like. With tools like this, the compromise of eliminating friction by centralizing communities is no longer necessary. I personally believe federating protocols rather than central servers are the future of online social interaction, and I am very excited about it.

One thing I think needs to exist is a message board/forum server that federates. The link aggregator UX paradigm is not optimal for all online communities. Forums are very useful and as of right now I know of no effort to implement federation in an existing FOSS forum server or build one with federation in mind.

[howlgarnish]

Is this meaningfully different from the user's point of view than just subscribing to a set of different subreddits? What does the user or the host get out of using federated servers?

From my POV this seems to be the same issue that has prevented federated social networks or chat services from taking off: there's no meaningful, concrete advantage to it.

[mariusor]

I'm the developer behind littr.me. In my mind, the best reasons for migrating to your own service as a reddit community is the fact that you get ownership of it. You can monetize, you can enforce rules that are different to reddit's, you can focus your interface more specifically towards custom UI, etc.

The main downside of moving off reddit would be that you lose access to the larger pool of reddit users and communities. That however is mitigated by the fact that federated communities can still interact with each other despite being independent services.

Personally this would be my main goal for the project: managing to get one of the cool subreddits to leave reddit in favour of starting their own independent community using my code. However some of the things that are made easier by the centralized model of reddit are quite more difficult in the federated case (moderation as an example), and I don't feel confident (yet) to push for this.

[betwixthewires]

Consider as a part of user experience the ability to select your own admin and mod teams and terms of service by choosing a server on which to participate, or even be your own admin and mod team by launching your own server, without having to sacrifice access to a larger userbase. This changes the user experience considerably and is a definite draw.