Skip Navigation

Posts
39
Comments
1,811
Joined
2 yr. ago

  • And healthy.community and viewfinder.pro are getting activity from other people who are just focused on doing their thing. What does this have to do with the overall point?

  • Piefed also has this Mastodon-esque tendency to implement features that only work on their system and are not interoperable with the rest of the ActivityPub software. Which is the kind of thing that is only "nice" until they are a minority player, but could make them one of the most hated systems if they start getting significant users.

  • Most people don’t pay for services on the Internet

    Yeah, but we are not "most people". I thought "we" understood if you are not paying for the product, then you are the product. I thought "we" understood that "Free software" was not a "free lunch".

    And if basic talk and text service was freely provided by volunteers, they’d milk those volunteer organizations dry, too.

    This is also why I think we should flip the script and stop cheering admins that run "free" instances. We should stop helping admins who can not make rent and we should start telling them to start valuing their work and demand proper compensation.

  • It’s too big to be a hobby, and too small to be a job.

    Facebook allegedly extracts $14/month of value from each of their US-based users, ~$12/european user, $7/month for Latin America and $4 from Southeast Asia.

    If each active user contributed $1/month for their instance and $1/month for the developer of the software they use, the Mastodon developers would have an operational budget of ~$800k per month, the Lemmy developers would have $50k/month.

    I don't think that the problem is we're "too small to be a job". I think that the problem is that the average "enthusiast" is an hypocrite. They will profess their hatred of the business practices of Big Tech, but they will look for any and every possible justification to excuse themselves to contributing to the pool.

    We have tens of thousands of people who (...) are mostly interested in consuming, not fighting for attention.

    Sure, but what I don't get is this: why is that people are absolutely fine with paying 10-20€/month (or $50-$70/month in the US) for their mobile phone service but expect that the server hosting service and software development service to fall from the sky?

  • You brought hypothetical scenarios. I was just pointing out a fact: the second largest Lemmy instance is going under and taking with it all the communities that were created there.

  • At one time I thought community spirit (for what that’s worth) would kind of tilt things in a long-term sustainable direction.

    Community is not enough. I wrote that in 2022 with Twitter and Mastodon in mind, but the same principle still applies for Reddit vs Lemmy.

    Lots of people say they want to "stick it to the man" but very few are actually going to put in the work and/or money required to actually succeed.

  • One scenario is hypothetical. The other really happened. It makes no sense to say "they are both true".

  • That was available to the lemm.ee admins in our scenario as well.

    It is a lot more difficult to get out of burn out than it is to avoid getting there in the first place.

  • They still shut down the instance.

    Only the users on lemm.ee are affected by it, in this scenario. It is bad, but the current scenario is much worse.

    Drama still happens involving lemm.ee users.

    User-only instances are less dependent on each other, defederation is not as big of an issue, so a lot of the drama would go away.

    Admins still get burnt out.

    Less communities on their instances means less traffic, less activities, less moderation reports (they would have to deal only with users on their own instances) and if even then they are overloaded with work, they could decide to scale down the operation before reaching burning-out point: close the instance for new registrations, make user registration conditional on payment/donation, etc.

  • This argument applies more to "instance with lots of users and groups" (what we have now) than "communities with lots of users on topic-specific instances", so I don't think this is the problem here.

  • We can not change "human behavior", so I don't see how/why we should expect things to "be different at .ee" compared to anywhere else.

  • Doesn’t really solve the issue, admins will all want to manage an instance for groups rather than instance for users as that would avoid much of the drama.

    And that is bad why...?

  • Not that these things didn’t happen to a significant extent, but it seems like a lot of .ee users and visitors, while willing to hang out at the place, were moreso just willing to soak up the content without putting in much effort to help make the place work.

    Blaming the community for that is not fair. It takes only a few rotten fruit to spoil the whole basket. Even if 99% of your userbase are model netizens who are supportive and only make positive contributions, the whole system can be brought down by a few dedicate trolls/losers.

    We need to build effective filtering mechanisms to get rid of abuse/spam and we need to maybe bring back the idea of Web of Trust. It's too easy to create an account and start polluting the fediverse.

  • Maybe we can take this as an opportunity to understand the importance of separating "instance for groups" and "instance for users"?

  • Can you describe how (if?) the flair part integrates with AP?

  • I'm downvoting because this community doesn't feel like the right place for any type of meta-discussion.

  • An oversimplification: wikidata is a graph database where people edit semantic triples (subject, predicate, object) instead of text articles.

    One could argue that the Fediverse is itself a graph database that anyone could edit, though current implementations are mostly focused on taking this abstract data and putting a usable shell that resembles specific applications.

  • Is every entry on a ibis an as:Article object? Could it be generalized for any type of Linked Data? This would make it possible to not only have a federated wikipedia but the whole wikidata project.

  • I'd guess that whoever would be willing to give you copies of legal documents just to join an "exclusive" instance would also be okay with contributing a few bucks per month to pay for its costs.