I just got a really good answer to a question of mine. Since the author was so knowledgeable (about vim in this case) and gave such a careful answer, I went to read other things he wrote. There was only one other question on unix.stackexchange.com, so I read it, and up-voted it since it was an excellent answer that might be useful to me some day.
I wondered though: at what point would this be considered serial up-voting? On the one hand, you could argue (I suppose) that I went to the questions because of the author, and I up-voted every one, and therefore it would be considered serial up-voting.
But on the other hand (let's be realistic here):
- I did actually read the questions in their entirety,
- I did honestly believe them to be high quality, and
- there was only one other question besides my own.
At what point would it be considered serial upvoting? Since the system is limited to using measurable factors to mark some votes as fraud, it can't "know" that I really thought the answers were high quality, so I'm assuming we're left with a practical rather than a principled way of distinguishing. What is that? What if I read all 1000 answers of a particular author, and thought that 950 of them were worthy of an up-vote?