Tl:dr: Remember the human, even if the project doesn't work, it wasn't as useless as it may seem, resources consumption may be concerning
Also disclaimer: I have no involvement in the Fediverser project other than following it from afar and discussing with the creator in a few comments.
Hello everyone,
As the other thread is already quite active and I guess my comment would probably be drown there, I open this new to bring an alternative perspective on the project.
Remember the human
First of all, could we please try to limit the hostility against the project creator? It's fine to disagree, to block, to defederate, but wording such as "hate", "screw the person" don't seem to align with "remember you will be interacting with actual, real people" and "Be respectful of others."
Now that this is out of the way, a few considerations to take into account:
The Network Effect - the issue that Fediverser is trying to solve
As most of you probably know, the network effect prevents most of the users of an existing platform to switch to another one. "Why would I go there where there will be no one, when all the people I want to interact with are here?"
It was the case for Mastodon until Twitter started to really become mediocre, and Signal still hasn't convinced most of the Whatsapp userbase to make the switch. Matrix is struggling to be a full Discord replacement, but has the benefits of having bridges with most of communication platforms (https://matrix.org/ecosystem/bridges/)
Those bridges can ease potential reluctant users to at least try out Matrix, as they can still access their previous network.
That was the whole idea behind Fediverser. I remember the initial plan being a two-ways bridge between Lemmy and Reddit, allowing people to see content from Reddit from Lemmy, interact with it, and having people on Reddit seeing responses too.
Added with all the Lemmy pros that we know (third-party clients, alternative front-ends, etc.), it could be a huge helper into bringing more people into Lemmy. Which brings us to the next question.
Do we need more users?
I know this is highly debatable, but I will try to bring some perspective on this.
I have been an active user on Lemmy for a few months now. I like it here, great apps, nice people, interesting discussions.
But still, I still to go Reddit too.
Why? Network effect. As much as Reddit sucks today, there is still content that is only posted there, and sometimes I just want to read that content. And I'm not talking about niche topics like obscure fandoms. Parenting, personal finance, relationship advice, fashion advice are topics that aren't very popular on Lemmy. And probably won't become anytime soon due to the network effect. Which is fine for me.
But the issue I see is that overtime, the migration might never really happen. We might be in a "next year is the year of the Linux Desktop" or a "Chrome vs Firefox" situation rather than a "Digg to Reddit migration". And I'm taking examples where the alternative is still widely used. Lemmy could actually become Diaspora, as over time, more and more people just think that the convenience of a Revanced third-party client is better than having to browse two platforms.
But to be fair, the future doesn't even matter that much. What I wanted to say here was that I understand why the Fediverser creator wanted to avoid that scenario, and tried to accelerate the process.
Resources consumption
The list of instances part of the Fediverse project can be found here: https://communick.news/c/communick_news_network. I had a look at two, https://level-up.zone/ which replicated a gaming sub, and https://selfhosted.forum/. While they are quite active, they don't seem to be that active (most of the threads have less than 5 comments, there are a few that high the hundreds, but they are quite rare).
I have seen several admins complaining about the system resources consumed by alien.top instances, "as much as the largest instances". Does that mean that if tomorrow reddit.old dies, we double or triple the number of users on Lemmy, instances would have to be shutdown? Can we afford a growth this large? The scalability issues have been mentioned since June, and it seemed that things had improved on that side, but should we be worried that Lemmy will hit a scalability ceiling at some point?
However, to be fair, I guess this point is mainly assessed as a "low return on investment" for the resource consumption. Which brings us to the previous point "What what Fediverser trying to solve".
As a conclusion, I hope this perspective might help people see why this project was made, and that maybe it does not deserve all the hostile reactions from the other thread.
That's it, thank you if you made it to the end. Looking forward having a discussion in the comments.
Have a good day.
Edit: I noticed I didn't mention the copyright issues in the comments, but to be fair I'm far from being knowledgeable on the question. It might however have a Streisand effect of having Reddit sue a single person over comments that are made for free by Reddit users. Is that worth being sued by them, I don't know (also, what about alternative front-ends like LibReddit, or archive websites?)
While we're at it: crossposts in general should go or he handled differently. The way things are, crossposts will create new posts with new comment sections like reddit did. Yet
a) if something is of interest to two communities, chances are, ppl who are interested are already in those two communities, so crossposts will only clutter their feed
b) it splits up the discussion unnecessarily
Why not have corrsposts as mere.links to the original post so everybody can join in? The way things are, each crosspost will turn into a circle jerk of all the same arguments regurgitated by different people in different communities.
#Kbin actually handles this really well. Whenever it detects 2 of the same #thread (at least that's how I think it works), it'll group them and only show one of the threads on your feed. If you click on it, right under the thread will be the other #cross-posted threads on other magazines, each of which has its own upvote/downvote and comment count. You can click on any of those threads to switch to a certain magazine's thread and see the #comments there. Also, when you comment, it only goes to the thread you currently have open.
Lemmy does the latter part as well. You just have a chance to see multiple as well on the frontpage if you're subscribed to both communities
Yeah, I just wanted to specify it isn't a mass comment kind of thing.
Interesting, thanks.
Kbin/Mbin seems more and more promising, that's good
That's the problem with federation though, you have 20 servers with 20 communities of the same thing, and there's not many people redirecting and curating, because everybody wants to be a powermod. When we had the reddit migration it started a chain reaction nightmare of creating an infinite number of dead, useless, redundant communities. I like to use sports as a good example. Fanaticus.social is designed to be the premiere sports instance, yet all the local instances, like .ca or midwest.social, also will have their requisite team pages.
Perhaps instead of mirrors to reddit, we should be working on linking those communities together.
It's happening already. You can see a few communities emerging as "the ones" on their topic
#Kbin has a #feature to kind of deal with this called #collections. Instead of subscribing to all #magazines of a same type or even name, you can put them in a collection (or find a collection where somebody already did that) and then favorite that collection see all of those magazines in your feed. Splitting up the discussion is still not ideal but at least this lets you see all of it at once and increases #discoverability.
I wish everything was a bit more standardized between kbin/mbin/Lemmy. It feels like we have these forks of the project that do different things because they emulate different behaviors of other sites, and reaching parity seems difficult without a lot of developer discussion.
I like a few things about kbin but for a while it was the instance causing the most spam on my feed because federated mod actions broke and spam cleaned up locally would not get cleaned on other instances. I saw Ernest back posting again so I guess development has resumed and some of those issues have been banged out.
Yeah, I believe Ernest had some serious irl issues come up and had to take a step back from development, which slowed progress down massively. But ever since he got back, he's been working really hard to #fix the major #issues and make the instance easier to use.