Over the past few days, I’ve witnessed a remarkable surge in the number of communities on browse.feddit.de. What started with 2k communities quickly grew to 4k, and now it has reached an astonishing 8k. While this exponential growth signifies a thriving platform, it also brings forth challenges such as increased fragmentation and the emergence of echo chambers. To tackle these issues, I propose the implementation of a Cross-Instance Automatic Multireddit feature within Lemmy. This feature aims to consolidate posts from communities with similar topics across all federated instances into a centralized location. By doing so, we can mitigate community fragmentation, counter the formation of echo chambers, and ultimately foster stronger community engagement. I welcome any insights or recommendations regarding the optimal implementation of this feature to ensure its effectiveness and success.
I like multis and I think discoveribility is a bottleneck, but I’m very wary of this idea. If you merge communities together like this, you essentially multiply the users in that community. Moderation isn’t 4 small instances anymore - it’s one large one with 4 separate mod teams each handling a quarter of the posts
I think this is more likely to lead to polarization and eventually echo chambers than if you kept them separate - outrage drives engagement more than anything else, and explosive growth is a great way for a fraction of the group to dominate the first few pages of comments, which turns off moderate voices, which works like confirmation bias to make the outraged believe they’re the prevailing voice of the community, which again drives them to post more incendiary comments, and the whole thing spirals
If you want to avoid echo chambers, the best way is to throw a small group together and make them get along through mods that are involved in the community
But then you’d probably end up with most members of one community slowly joining the rest, which is a healthier growth model, but still not great
My intuition is that the ideal solution involves encouraging users to join a single smaller group, but being exposed to top posts from sister groups to avoid fomo. Possibly through something like the way Reddit handled crossposts, where you get the post but not the comments, and a small link to the discussion in other communities. It could be automated if the post crossed a certain threshold of votes, keyed to a certain deviation above the daily average of the original group and optionally with a minimum up/down vote ratio.
This would help keep moderation ahead of participation, and hopefully build a tighter knit community - people are less willing to be jerks to people they recognize than strangers you get in a larger population. By encouraging users into one small random group instead of shopping around for the one that best fits their view, I think we could resist natural grouping by beliefs.
To go further, if this works we could consider a mechanism for “mitosis”, a splitting of a group when the mod team feels the culture of the group is getting past their ability to manage in a nuanced way
The goal is decentralization after all, not distributed centralized groups
Make it user specific. Feeds are combined solely from the individual user’s perspective. Consumption would be easier but submissions are still federated.
I think this was the proposal - but problem is still doing this automatically.
It’s not posts I’m worried about, it’s comments. Comments are where the discussion takes place and the culture develops
Why not just allow the option to subscribe to them as a multi or subscribe to them individually, leave it up to the user to decide?
Because humans are barely sapient animals with limited understanding of ourselves and little to no awareness of the long-term consequences of our actions.
We don’t operate in our own best interest or the best interest of the group, we’re built on the assumption that the environment and our local community will moderate our actions. There’s natural limits to physical actions, natural repercussions to social ones when everyone knows each other
Technology doesn’t have these limits. Things made of code can scale past human comprehension in seconds. And it changes it’s users
Part of the ethics of software development is to carefully consider the ramifications of what you bring into the world.
The public can’t make an informed choice, because they lack both the nuanced understanding of the tech, and every choice has a cognitive load. It’s up to you to make it safe and healthy or to inform them of the consequences, and you can’t just put up a 26 research papers on the psychological and solciological considerations for hitting a button… No one is going to read that.
You also can’t have booby traps - anything a user can do inadvertently or accidentally shouldn’t have serious consequences.
There’s some room for debate, but it all comes down to this: you’re responsible for how an average user is going to use your technology. You should do all you can to make it easy to use the tech safely, you should add covers over the buttons that do something with consequences, and things with deeper ramifications should only be available to power users who presumably have the technical knowledge to make an informed choice themselves.
So onto this situation. Say you make this button “sub to /c/_____ and all sister communities”. That’s not really a choice - it’s like you go to McDonald’s and order a burger, and they say “for the same price, I’ll give you 3 additional burgers with different options”. Some people would say no, but they wanted a thing and you offer them more of the thing. If they haven’t tried them before, there’s fomo - what if one of the other burgers is better? And it’s not like they couldn’t just stop eating.
The majority will accept 4 burgers, because they don’t see the hidden consequences. There’s no world where the average person sits down with 4 burgers and eats less than they would if they had 1 - it’s human nature, studied and documented… Giving someone more food leads to them eating more, because we judge the amount we’re eating in large part visually. Put it on a larger plate or pile it higher, and we underestimate how much we’ve eaten. Put it on a small plate, and we eat less.
Sure, there’s people who understand this - those of us who’ve struggled with weight or food scarcity are either not going to accept the burgers, or we’ll set 3 aside for later. There’s people who might benefit from eating 4 burgers - someone who burns 10k calories a day needs that kind of intake (even though they’d be better off with more variety).
Good or bad, you’ve increased consumption based on how you’ve presented this choice. The outcome was a statistical certainty, but technically it was a choice. It’s just a choice that every human would naturally answer the same way if they went in blind - do you want only the thing you asked for, or that plus more free stuff.
So if you make this a button, it’ll overwhelm the single sub option. And there’s a game theory aspect to this - I’d likely hit the button too, because individual choices here don’t matter, it’s a matter of speed and volume of users subbing and unsubbing