Trackers like Lemmyverse.net and Fediverse Explorer are incentivizing instance owners to inflate numbers with bots
Trackers like Lemmyverse.net and Fediverse Explorer are incentivizing instance owners to inflate numbers with bots
cross-posted from: https://lemmy.world/post/358415
The vast majority of the instances in that screenshot have known jumps from 1~50 users to tens of thousands in less than a day. These instances also happen to not require a captcha on sign up.
It may very well be that instance owners are innocent as some have really been victims of bot attacks and simply forgot that you could enable captchas for sign-ups, nevertheless I think instance directories like Lemmyverse.net should start disincentivizing anyone from inflating his own instance with tens of thousands of bots in order to get on top of those "leaderboards".
What incentive is there to have a ton of users registered to a particular instance?
The larger instances get featured on websites like lemmyverse.net and get more visibility. The owners of those instances can then get free traffic which they can redirect wherever they want.
So they can scam and spam people effectively. As users, you try to block one spambot, and its a losing battle because there's 20,000 more accounts to take it's place.
If someone wants an audience, I imagine spamming activity in your own communities would be more effective. If most people subscribe to the largest community in the topic they are interested in that could give you a lot of new eyes across all the instances.
I own lemmyverse.net and I disagree. More users = more load. Users can pick an instance they are happy with.
Actually your "smart sort" looks like it successfully ignores spammy looking instances, but a sort by # of users (which is a natural reflex since people coming from Reddits would feel comfortable with a large instance, we're talking about people completely new to the Fediverse idea) gives this unknown "Coffee" instance as the 4th largest instance and its owner still hasn't enabled captchas.
