It doesn’t take a genius to realize that a long-running account on any platform can easily be used to build a profile of someone.
Since many discussions on Lemmy and other platforms can often cause someone to write about their job, family, hobbies, where they live (city/community), etc., there’s a lot of concern about non-private post history being used against someone.
Other than using fake names, throwaway emails, etc. are there any other best practices for handling this?
Should we be creating new accounts/profiles every once in a while?
I’d say no. Your fooling yourself if you think you can. The best you can do is make it harder, but that is only true compared to computing power. Problems that were hard, are now possible because computing power has increased.
Any one profile could give enough to out you. A textual analysis can tie separate text posted by different accounts to the same person. (Varying degrees of accuracy for this. If anyone would do parse all Lemmy comments to do this is separate issue.)
In short, don’t post anything that might compromise you. But how fruitful are conversations if you’re always hiding anything that might out you? Stick to only non personal stuff? Information about local shops, movies, etc etc can all narrow down your location. But even with that, narrowing down to a town with 50 people in it, is different to a town with 1 million people.
It isn’t only what you post. Someone else can also out you. Someone who knows you, or someone that you don’t know. Your PC can be hacked, etc etc. The options go on and on.
Define what your risk is. Define who your adversaries are. Then plan how to operate with acceptable risk tolerance.