For those that have poked around other fediverse stuff beyond Lemmy, and been around the spaces awhile, what’s stuck out to you as stumbling blocks, or basic user experience fumbles? Which parts do you think may be technical, and which may be cultural?
netsplits/defederation.
You can’t just tell someone to register for any server, and they will be able to see everything. So they then have to choose a server, which takes effort, and can cause analysis paralysis.
Maybe that is exactly what we need to do, to spare them from the indecision. Recommend them to a specific instance to sign up and follow you (if in doubt, the instance we use). I suppose we can mention there are lots of choices, and those who are inclined that way will want to explore other servers, many are not, and for them pointing them at a server may be best.
I’m just thinking that trying to say there are lots of networks, each with lots of servers etc, may be the problem.
Alternatively, should ask them some questions like do they want to post short format or long text format, and take into account a specific interest they have, and then we still recommend a server instance to them to join.
So for fellow ham radio operators, I just pointed them all to the ham radio Mastodon instance and said sign up there.
We can compose a list of instances with sane blocklists for each software and audit from time to time.
Yeah, but then the blocklists themselves become a centralized feature. I’m not saying “don’t block the fascists”, just that it’s going to be hard to maintain a blocklist.
I can totally see the Fediverse going the way of email, as in you need a reasonably large amount of capital to maintain a well-respected, not defederated-from server.
By ‘sane blocklists’ I meant small and auditable blocklists actually. There are instances like programming.dev, lemmy on sdf and the instance I’m on that don’t preemptively defederate from other instances. That’s what I meant.
https://fediseer.com/ can be used for something like that.
The ability to “float” between servers would go a long way to improve this. Make an account on one server, then come across another that you vibe with more, and single button press and you’ve transferred. All you subs, comment history etc are preserved (for overlapping federated servers). No idea how to implement this, but it feel achievable, perhaps with a quick step to set up a new password and username if it was taken.
Once the fediverse gains significant traction there will be a huge coordinated media smear campaign to associate it with extremism, CP, violent crime, terrorism etc… The “won’t anyone think of the children” attack will be leveraged to ““regulate”” the fediverse (probably by legacy social media like Meta alongside the security state), aka transform it into regular old centrally controlled social media. Upload filters, encryption back doors, know your customer laws etc… Non-regulated decentralized social media will be attempted to be made illegal, not sure if it will succeed.
So yeah eventually any true decentralized social media will have to move completely into the “darknet”.
- Onboarding
- It lacks the critical mass to accumulate more users
- Account migration / excessive defederation
Controversial and probably unpopular opinion:
- the biggest hurdle the fediverse faces is that it’s not run by a business with monetary incentives to make it more popular and doesn’t have any marketing / market research / product managers focused on gaining users.
I’m someone who hates advertising with a burning and seething passion, and I’m no lover of capitalism, but from a systemic standpoint there’s a reason most open source projects burn out and go nowhere, and for-profit businesses have a higher chance of survival, because there’s direct incentives (you know money/food) to keep making commercial software and increasing it’s user base, but there isn’t for hobbyist and open source software. Especially in the case of a social network that is only as valuable as the content and users on it, this might be a long term systemic issue.
most open source projects burn out and go nowhere, and for-profit businesses have a higher chance of survival
You know like 50% of new businesses fail within 5 years, right? I don’t have stats on open source projects, but it seems to me those are more likely to fail because they’re run by one person who loses interest than because they don’t have a profit motive.
You know like 50% of new businesses fail within 5 years, right?
Yes, that is a remarkably low failure rate. 99.9% of open source projects sit unused and abandoned after 5 years.
those are more likely to fail because they’re run by one person who loses interest than because they don’t have a profit motive.
They’re run by one person because they don’t have a profit motive, so they don’t need to hire QA, market research etc. etc. All the parts of a software company that help to keep continuously developing their software and make sure users are happy.
Dude, yes, they’re run by one person because it’s a hobby. This is like saying 99.9% of stories don’t get published because there was no profit motive. There usually isn’t when it starts, just a drive to create or fill a perceived void, or even just practice. I write damn near every day with zero profit motive.
Linux wasn’t started with a profit motive. None of the open source BSDs were either. As far as I can tell, they’re still not particularly profit motivated. Neither are a lot of other open source projects that have lasted ages. Where’s the profit motive behind Bash? It’s been around for 34 years.
An inability to pay bills can stop a person from working on a project, but at the end of the day it’s usually not profit that keeps an open source project alive. It’s popularity and passion.
