2 ms·
Recall that de-anonymization analysis is pretty effective these days. Even if you post under a made-up user name, there's a pretty good chance they can figure o
by CurtMonash 13y ago
Recall that de-anonymization analysis is pretty effective these days. Even if you post under a made-up user name, there's a pretty good chance they can figure out who you are.
That's one reason I post under my own name; anonymity wouldn't buy me much anyway. Even in forums where I'm technically anonymous, I don't try hard to preserve any secrecy about my identity. It's more a matter of "There's a culture here of intemperate posts protected by anonymity, so if you notice me posting there, please also understand that I might be responding in kind."
- wsxcde 13y agoDeanonymization that works well relies on: (i) correlating social graphs (ii) correlating likes/dislikes/reviews etc. across different networks. (iii) Lots of data to do (i) and (ii) And it's still difficult to do for random people on the internet (as opposed to the NSA or serious attackers such as those willing to put in the effort to crawl and analyze the entire linkedin graph.) I believe deanonymization based on just textual analysis is still a little bit of an academic effort. Anonymity does buy quite a bit - especially on a forum like HN - where there isn't a social graph and the like/dislike information is private.