Hey peeps. By now you all heard that Zuck's meta has started and in a defederated mode. This instance has already signed the Anti-Meta Fedi Pact and we plan to not help improve the value of Facebook, if and when they plan to open federation.
However I am thinking we can have a bit of fun beforehand.
You see, Threads is extremely sanitized in order to be "corporate friendly". Naturally this makes their service as exciting as a bowl of plain porridge. However in order to keep their corporate partners happy, they are also extremely cautious about anything controversial and will likely defederate from anything too spicy.
So why not use this an an opportunity to make Threads defederate from us?
What would you say that we leave federation up, but use the opportunity to share with the larger threads populace some important information about filesharing? Sharing is caring after all and we have plenty of knowledge to give.
Naturally, if this doesn't work, we will defed them ourselves soon after anyway, but I expect it will. What say you all?
Image made with Lucid Creations and SDXL
But they would if they’re allowed to federate. This is a company that builds shadow profiles on people that don’t even use fb. You really think they’re not going to try and get what they can get their hands on by any means necessary?
They could already do that by setting up a secret spying instance, or just use a webcrawler. You have to understand that anything you post online is out there forever.
You know I didn’t think of this. Does every new instance created, if it chooses to, federate and download data automatically from the other instances? Like upvotes downvotes etc, or is that only possible when something like kbin or lemmy.ml agree to federate with instance x?
When a new instance is created, its default allowing all other instances to federate. But nobody knows you exist and you don't know anyone else exist. It isn't until 2 instances interact, a "link" is established. When linked, the instance url will show up on each other's /instances page. (like lemmy.ml/instances just replace the domain with whatever) You establish the link by manually searching for a community. example: [email protected] and then a search will be conducted and withing a few minutes, it will be indexed and show up on the instance that conducted the search. Now you know each other exist, but you also have to manually search every community in order to find everything a instance has to offer. But to sync every post in the indexed communities, there must be at least one subscriber to a community.
So basically,
-Start instance
-Search for communities
-Subscribe to communities
And you're good to go.
if you want the target instance to see your instance's communities, you do step 2 and 3 on their instance using an account you created there.
What the other poster said. That information is already publicly available and can be collected without even having another instance just via the public api.
So I can create an instance and automatically start scraping people’s info? Or is an instance not even necessary? I understand how web crawlers and spiders will index but do they have access to the granular account data? If so how do we mitigate this?
You only have limited access to use info. You can't see emails or IPs etc. Content, votes etc you don't need an instance, you can just scrape directly.
There's no mitigation.