I'm not any kind of expert on this. I don't have personal experience (obviously I wouldn't admit it if I did, but still); I'm just making educated guesses based

I'm not any kind of expert on this. I don't have personal experience (obviously I wouldn't admit it if I did, but still); I'm just making educated guesses based on my knowledge technology, people, and the way that previous doxxing campaigns have succeeded. I hope someone else can answer your questions better; I don't want to mislead by sounding more authoritative than I should. I can't think offhand of someone in the Internet era who became famous (for writing/posting) under a pseudonym, where people had incentive to doxx them. That I can't think of such cases is very weak evidence they don't exist; the majority of cases would occur in non-English media which I don't read anyway. Scott Alexander is a good example of someone who has succeeded in having large and diverse online following behind a pseudonym. But he's not an example example of what you're looking for. He is very weakly pseudonymous, trivial to doxx. He often refers to his personal life on his blog (more so on his Tumblr), his pseudonym is linked to his real name, and many people know him in the flesh as the author of SSC. There is good information online on correctly, securely using Tor and VPNs to remain anonymous and on securing (and ideally segregating) the computer you're using for this. This information is often targeted to whistleblowers and to people whose thread model is their governments, but it works equally for anyone. Interacting with public comments on your own blog should be fine if your connection to it is securely anonymized. You can register a domain anonymously using bitcoin or prepaid CC without providing your contact info, but then the registrar is the owner of record of the domain and you have to trust them. The biggest danger may be that in the event of a dispute or wishing to transfer to a different registrar, you won't be able to assert control over the domain name without revealing your identity. See also: https://en.wikipedia.org/wiki/Domain_privacy https://en.wikipedia.org/wiki/Domain_privacy Making a list of things you're willing to reveal about yourself and sticking to it sounds like a good idea: precommit to clear rules, use checklists. The linguistic analysis questions depend heavily on both your threat model and on future technological development and I don't dare to try answer them.