Abstract
This paper examines the potential of the Fediverse, a federated network of social media and content platforms, to counter the centralization and dominance of commercial platforms on the social Web. We gather evidence from the technology powering the Fediverse (especially the ActivityPub protocol), current statistical data regarding Fediverse user distribution over instances, and the status of two older, similar, decentralized technologies: e-mail and the Web. Our findings suggest that Fediverse will face significant challenges in fulfilling its decentralization promises, potentially hindering its ability to positively impact the social Web on a large scale.
Some challenges mentioned in the paper:
- Discoverability as there is no central or unified index
- Complicated moderation efforts due to its decentralized nature
- Interoperability between instances of different types (e.g., Lemmy and Funkwhale)
- Concentration on a small number of large instances
- The risk of commercial capture by Big Tech
What are your thoughts on this? And how could we make the Fediverse a better place for all to stay?
It’s a reasonable MVP, but still needs major progress around discoverability.
I’ve tried dragging some friends in, and they’re not opposed to leaving reddit, but the refrain is always ‘topic x isn’t there’ or ‘nobody is posting anything’.
You can easily pick the “wrong” community on a topic, and see zero content when there’s another one full of posts but… how do you know which community on what server is the “right” one without finding a 3rd party tool that can provide metrics and you’ve lost most of the people who were interested at that point.
And, since Lemmy is quite a lot slower, it’s also hard to tell if you’re in the right place but nobody is posting or if you’re in the wrong one and nobody is ever going to post anything, which kinda makes this require a lot more time than just picking the single subreddit that looks like what you’re after.
Discovery should not be a function of the Fediverse itself. It should be a tool, app, or otherwise layer on top of a baseline. Discovery is an opinionated service offering—if it’s baked in, it’s not a good thing IMO.
I think that depends on your goals and concerns more than anything.
I was merely relating my anecdotal experience with about 10 people who would be Lemmy users, but bounced off so hard they’re not even interested in considering it ever again because finding anything interesting was more effort than it was worth, so they’ve literally written off the whole Fediverse as a dumb idea for nerds.
I can’t say I blame them because finding anything worth interacting with on Lemmy or the microblog platforms or whatever is worse than the commercial platforms, usually by design and, mostly, I don’t disagree that killing the algorithmic feed is a good thing.
But, for something like Lemmy? It should be trivial to find out where people are talking about a topic you’re interested in, and not require knowing someone who knows what tools are updated and current and accurate and how to use the tool and THEN how to take what the tool tells you and subscribe to a community since even that is obtuse and complex.
I think piefed’s community grouping thing is a step in the right direction, but probably could go further with being able to find and share groups between instances.
I’m so torn on this. I’ve tried looking at Mastodon but the chronological feed is just not giving me what I want. Maybe I’m permanently ruined by algorithms. Firefish with its antennas sounded like an interesting compromise but that project turned messy and I don’t really know where to go from there. Wait for the Iceshrimp.net code rewrite?
I actually have a Firefish instance I also admin, and with enough relays and antennas, I’ve stuffed together a feed of people that mostly post interesting stuff, with the occasional derail into politics.
It’s not quite perfect, but I’m also fine with it being interesting but having an end and having to quit and go do something else over endless scrolling.
I don’t know, getting off TikTok was probably good for me so maybe I shouldn’t pursue this algorithmic feed diversion desire I have. I just have an itch sometimes.
Are you happy with Firefish in its current state or are you looking at any of the forks or rewrites?
It works perfectly fine for what I’m after, it’s stable, my instance has been around for over a year with no issues, and Firefish still has a dev team that’s stayed on top of fixes and security issues.
Right now I can’t see ANY reason that could possibly make me nuke and re-do everything on different software since migrating accounts and data isn’t exactly straightforward and breaks follows/followers still. Feels like it would be a huge waste of time given I can already scroll people posting about retro games and computers and shit.
That’s fair. I remember Firefish had a bunch of drama a while back and the main Firefish.social instance died, but I’m glad it has a stable and active dev team now. Someone else I was talking about this with recommended the upcoming Iceshrimp .NET rewrite, but I don’t know enough about these things to know the distinct advantages. Makes sense not to bother if migrating still breaks followers.
Yeah the developer and firefish.social admin team had some sort of breakdown and basically fell off the internet and shit died.
The Iceshrimp rewrite is theoretically great, but I’ve been in IT long enough that someone says “rewrite” I hear “might be useful in 3 or 4 years”.
And yeah, migrating between platforms still basically doesn’t happen with activitypub stuff and I’d be surprised if that ever really gets fixed in a timeline that matters.
(If you’re interested, you’re welcome to see if Firefish does the needful for you as my instance is open signups.)
@[email protected] , our discussion earlier today 😄
Link to comment thread.
Thanks!