In the last week I have seen solar punk go down and aussie.zone owner talk about stepping away.

I've not seen two places down at once that has made it to my feed in the last 3 years.

Is the self hosting aspect of lemmy a detriment to the entire environment? Admins getting burnt out or the costs becoming too high is absolutely an issue.

I know this is kind of rambling but what would happen to Lemmy if the major hosters all went down, and we all spun up our own personal services?

all 31 comments

sorted by: hot top controversial new old
[–] 19 points 5 hours ago

I really can't blame people for getting burned out--hardware costs are expensive as fuck right now, software exploits are more urgent than ever, the entire Internet is endless sea of bots, and there is no mechanism in place to offset any of the associated costs or work to the community. As things are right now, a small minority of fediverse users are shouldering all of the burden of various costs so that the rest of us can use this infrastructure for free.

From a money and time standpoint, it seems mostly unsustainable, and it's not something that can go on forever or scale up and up.

I think there are ways to make it more sustainable, like quotas, subscriptions and ads... But they are pretty unpopular.

Unlike corporate social media, we can't rely on leveraging other people's money in the form of VC or stocks, and so we are left mainly with voluntary donations.

But in the end, the health of the fediverse is arguably in the breadth and diversity of its servers. While it's good to have some newbie friendly flagship servers, it's probably better if everybody isn't concentrated on the same few mega servers.

And it's a shame when a server goes down because the person running it burns out or can't afford it. But this usually ends up being a social problem moreso than a technical one. And social problems can be mitigated (or avoided) by improving communicating and making sure that we aren't putting all of the load on any one person. There are cultural changes that we can make to improve the health of specific communities and instances.

  • source
  • [–] 22 points 5 hours ago (4 children)

    ...solar punk...

    I've been reading it as slurp-nik in my head

  • source
  • hideshow 4 child comments
  • [–] 5 points 3 hours ago (2 children)

    Sounds like a Soviet space program. Or submarine program?

  • source
  • parent
  • hideshow 2 child comments
  • [–] 38 points 6 hours ago (2 children)

    Is self hosting a problem for Lemmy? Yes and no.

    Self hosting for personal or small groups is a great feature. It allows communities to be independently responsible and has a relatively low barrier to entry.

    I think Lemmy needs to have a certification process though. People not looking to self-hosting should be guided to instances that have documented availability, backup, recovery, and backend support.

    You shouldn’t find out that your instance is running off an single drive in sombody’s mothers basement and they are out of town for 3 months by having problems logging in.

  • source
  • hideshow 2 child comments
  • [–] 11 points 5 hours ago (1 child)

    The developers have made it clear that Lemmy isn't designed to scale and that has been made apparent in their design philosophy and what they focus on.

    Also, users minimize the work that it takes to moderate a decently sized community or administer an instance. The design philosophy bakes in Reddit's role hierarchy while trying to present Lemmy as being more open and free.

    Beehaw admins pointed out the design issues a while ago and at one point considered building its own backend to replace Lemmy for their use.

    Sadly, building a large community isn't just a technical issue.

  • source
  • hideshow 1 child comment
  • [–] 1 point 1 hour ago

    Isn’t Lemmy capable of being scaled horizontally now? Of course the DB could still be a bottle neck if it doesn’t work with something horizontally scalable like Citus, Cockroach or Yugabyte (or if the schema doesn’t lend itself towards sharding).

  • source
  • parent
  • [–] 24 points 6 hours ago* (4 children)

    If we all spun up our own personal services, there would just be that many more shutting down.

    I love the fediverse but the whole "there's so many TLDs, you have to use a third party link service that you need to share links through and that the person clicking the links needs to configure on their end before it will forward them to a link where they're still logged in and can see the content" is absolutely laughable to me. I am always amazed that nobody from lemmy or piefed have attempted to make the linking process make at least some sense.

  • source
  • hideshow 4 child comments
  • [–] 17 points 5 hours ago (3 children)

    I'm not gonna lie. I stumbled on lemme by mistake, looking for a way around Reddit IP ban.

    I've enjoyed the hell out of y'all's company and conversations. Truly. Not everyone here likes me, but for most part I don't feel like any one here is a bigot of any kind.

    But...dear fucking God I do not understand a goddamn thing about any of the shit y'all talk about when it comes to any of that.

  • source
  • parent
  • hideshow 3 child comments
  • [–] 1 point 1 hour ago*

    Just try to picture email. The different providers on different domains; different people have different addresses (after the @). Yet one can send to the other and back. That’s federation. The protocol behind Lemmy (ActivityPub) does something similar.

  • source
  • parent
  • [–] 17 points 6 hours ago (2 children)

    I look at it like this - the self hosting nature makes the whole system more robust because there's no single point of failure. Lemmy as a whole will be fine. We will lose an instance or two every once in a while (RIP lemm.ee) just as you lose a schoolmate or coworker, but the system/environment lives on.

  • source
  • hideshow 2 child comments
  • [–] 3 points 1 hour ago (1 child)

    Depends how you define “point of failure”. Technically Reddit (or the other big techs) are much more robust. They are geographically distributed, have backups, multiple datacenters etc.

    But organizationally you depend on that one legal entity. If they shut their big system down, it’s gone.

    So I think in regards to unexpected/technical failures, big tech wins the reliability race. But in regards to legal dependence… federated self hosted wins.

  • source
  • parent
  • hideshow 1 child comment
  • [–] 12 points 6 hours ago* (2 children)

    Why would anyone who wasn't delusionally idealistic want to provide that service?

    Do you really want to spend hours of your life per day trying to prevent people from spamming their insane propaganda to your instance and endlessly being harassed and doxxed by internet strangers because you blocked them from posting pedophilia and other illegal and violent content?

    Self hosting is great for a small private community, not so much for something that is akin to public infrastructure, let alone the type of infrastructure that attracts the type of nutjobs that make up most prolific users in fediverse.

  • source
  • hideshow 2 child comments
  • [–] 7 points 6 hours ago

    I don't like how the huge instances have such control over the content. I like having all sizes. If we lose a few, we lose a few.

  • source
  • [–] 8 points 6 hours ago

    I think this continues to point at some fundamental design flaws in lemmy we still haven't addressed: provenance, and migration.

  • source
  • [–] 7 points 6 hours ago* (6 children)

    I think the fedi project that it most interesting in this regard is Holos, where your fedi server runs on your device. Having an app that runs on your phone that “just works” is probably the best solution long-term, and it feels like Holos has a reasonable chance of achieving it.

  • source
  • hideshow 6 child comments
  • [–] 3 points 4 hours ago

    Ooh, this looks fantastic. My first thought was "doesn't this just move all the costs to the relay?" But the relay has a 30-day retention limit, so that'd keep storage costs well under control there.

  • source
  • parent
  • [–] [S] 3 points 6 hours ago (4 children)

    That seems super interesting, I wonder though what it would do to data costs on a phone though

  • source
  • parent
  • hideshow 4 child comments
  • [–] 4 points 6 hours ago (3 children)

    Holos is designed to be data-efficient. Text activities are tiny (a few KB). Media (images, videos) are loaded on demand and cached. The relay handles delivery fan-out to your followers, so your device doesn't need to contact each one individually.

  • source
  • parent
  • hideshow 3 child comments
  • [–] 1 point 4 hours ago (2 children)

    So there’s a central failure point…

  • source
  • parent
  • hideshow 2 child comments
  • [–] 1 point 3 hours ago (1 child)

    Do you mean each person's relay? Isn't that the case for each instance as well?

  • source
  • parent
  • hideshow 1 child comment
  • [–] 2 points 1 hour ago

    I didn’t read the link but the text said “the person’s device doesn’t need to handle” the fan out, which means there’s a server somewhere. If it’s designed to avoid the pitfalls of self hosting then it’s centralized server tech.

  • source
  • parent