Two weeks ago, I made a post that went viral on the Fediverse.
It got boosted 5,800 times.
What makes this interesting is where it went viral: on a single user Pleroma instance (atomicpoet.org).
This instance is hosted on a VPS with only 1 GB of RAM allocated.
Trust me, this has been a learning experience:
1. Small instances can handle handle viral posts
2. But can your web browser? Not necessarily
3. The biggest problem is moderation -- even on single user instances
Whenever I try to view that post in its entirety, my web browser slows to a crawl.
In order for Pleroma to run on minimal resources, it's apparent to me that it's offloading some work over to the client.
Which is fine if it's just a regular post not seen by many people.
But if literally thousands of people are interacting, things becomes a problem -- especially when you got to moderate comments.
Okay, so why do you have to moderate on a single-user instance:
1. If you're operating one, you're hosting whatever content is on it -- that includes comments
2. If you accept data from just *anyone*, stuff like comments becomes a vector for a DoS attack
3. Remember, everything that gets posted on your instance also gets sent to other instances
Thus, if you receive comments from a far-right troll (like I did) other people on other instances are receiving them too.
The other consideration about the Fediverse is that viral content is not a one shot deal.
On Twitter, you can assume that if something has like 5,000 retweets, it will be a big deal for a few days -- then will be quickly forgotten, no one remembering it.
This assumption does not apply to the Fediverse. Viral content has a long half-life.
Anything can be re-boosted at any moment, and now you have to moderate again.
The reality about viral content is "bandwagoning".
What do I mean about this?
Different parties often use viral content to bandwagon their pet causes.
If you make a viral post about puppies, someone will wander in and try to bandwagon about endangered European hamsters.
You might be okay with that. But the bigger and longer that puppy viral post happens, soon the trolls will pop in -- and they're going to want to bandwagon too.
I just had to interrupt this thread because an instance admin wandered in to tell me he doesn't believe in actively moderating unless the content he hosts is deemed illegal.
So I looked up his server rules, and he doesn't take a pro-active approach at all -- content is only removed if "local authorities" report it.
Yikes!
So I blocked his instance on mastodon.social, then reported his instance to Fediblock.
This is why moderation is a community effort.
@atomicpoet I’m new to this so hence the question: what’s fediblock and where do I find it?
@ronansoleste Fediblock is the hashtag used for instances to collectively block a bad instance. Please don't use that hashtag haphazardly :)
@atomicpoet @ronansoleste
Exactly. And I've blocked the instance in question already.
Chris Trottier did it well, providing clear evidence, including evidence from the rules page itself. (By the way, *that's the entire rules page* for megamast.io.)
Sometimes, it can be a good idea to do checking before trusting a random use of the fediblock hashtag, particularly if the reasons for blocking don't make sense, if it's from a person who isn't trustworthy, or if the report just doesn't seem correct in some way. I've seen people try to take advantage of the hashtag for less than honest purposes.
