And that's not how I or fillippone see it; it's the way the protocol sees it. If you have a private key, you can sign messages with it. You're, therefore, the signer and, ultimately, the legitimate owner of that key.
I don’t think I would agree with that broad or a statement. For example, if someone uses deception or exploits a weakness in security to install malware on a computer to obtain a private key, I would not consider the person the “legitimate” owner of the key. Similarly, if someone uses a $5 wrench attack to obtain a private key, they are not the legitimate owner of said key.
After reading the thread in question, it is clear that fillippone was not using deception, nor force upon any third party to try to find a private key originally used by someone else. As such, I don’t think it is accurate for nullis to describe fillippone as a “wallet thief”, wannabe or otherwise.
fillippone was using what amounts to be brute force to look for used private keys. If however there was a more efficient means to generate an already used private key, I would expect many people to generate these keys, which is why I desire bitcoin to use robust means of cryptography to go from private key, ultimately to address. I would equate calling someone looking for weakness in the cryptography used in bitcoin as a “wallet thief” the same as discouraging looking for weaknesses in cryptography, and that will ultimately lead to negative results.