I’ve seen lots of discussion on reddit of users trying to get others to join Lemmy and the prevailing reply is that it is too difficult to navigate and comprehend. Having to answer multiple questions and wait for manual verification is combersome and is limiting growth at a time when nothing should be standing in Lemmy’s way. Combine this with server/instance selection analysis paralysis, and you get my point.
The linked mastodon blog post sums up my thoughts, but the TLDR is essentially this:
Don’t let the perfect be the enemy of the good. Don’t let dreams of decentralization interfere with the greater goal of achieving the network effect.
We should all be telling people to go to lemmy.ml and sign up. The devs should be too, and they should rethink/remove the questions and waiting period. Hell, just put a captcha. Discussions about servers and analogies to email as an example of federated service we all already use is a waste of breath. We shouldn’t have barriers to entry.
Thoughts?
EDIT: I’ve just found kbin.social and find it has superior signup options. It’s just: make an account (email/password), or sign up with Google or Apple. No server talk. Upside is the layout is nice and it acts as a Lemmy instance (threads) as well as a mastodon instance (microblogging). Only downside currently is that their android/iOS app is in development and isn’t ready yet, so desktop only.
https://github.com/ernestwisniewski/kbin
I think this might be the better recommendation for newbies at the moment.
While this would get the effect of more users, it goes against what decentralized services are intended to do. One of the biggest things that decentralization brings is that Lemmy does not become another Reddit situation happening now in 7-8 years, if users are spread out over many instances, if Lemmy decides to pull a Reddit or a Digg, you can just go another instance instead of having to abandon it entirely.
Sadly it’s not a problem that can be easily solved by pointing users to a single instance, because then, ironically, you fragment the fragmented community into an “us vs them” situation against the lemmy.ml instance if anything were to happen, with lemmy.ml always winning because that’s where the users would be.
I think having a short list of general purpose instances, maybe 5-10 or so, where it chooses one of them at random and lists the others under an “other servers” button is the best compromise to this, as it spreads the load out across trusted instances, while also not leaving a single instance to become so big that they essentially control the entire network.
I completely agree. But lemmy.ml should sit in the default. Hell, I would even add a tool tip next to the drop down box for other servers that says, “Don’t know what this is? Click here to learn more about decentralization, or just leave it as the default and click continue”
I can see both of these ideas potentially working out, I just hope if Lemmy decides to do something like this that they ask the community first for ideas instead of just doing it, so they can work out a solution that works for everyone
I say they do this now, and in 5-7 years if Lemmy is lucky enough to be a strong reddit competitor, then they could do any number of things to solve the centralization problem. They could easily stop accepting signups to lemmy.ml and put a new default in its place to help spread the load.
Right now, lots of reddit users are bouncing here to check it out, and bouncing back due to the complexity of signup. That is wasted potential to onboard. Fear of centralization is not Lemmy’s primary issue at the moment. Lack of network effect is.
I agree that they should ask the community. But that’s kinda what I’m hoping to do for them with a post like this. If this post gains traction I hope they take notice.
Any large and old instance going out will cause significant damage. Spreading across multiple instances won’t solve that unless the instances approach the number of users. There’s another way to do this. It’s the Wikipedia way. Clearly license the user content posted on the large instance as CC and publish a regular backup or export of the data. That way anyone can bootstrap a copy of the main instance if it dies or goes rogue.