I think you've missed the point. I'm not disagreeing with your desire to not read what he writes. I'm saying that your claim to be able to objectively quantify that in a way independent of your own opinions is empty. You're just taking the abstract notion of objectivity and replacing it with unknown quantities. You're trying to elevate your existing opinion to the status of a statistical inference without having to show your work. I think that's deeply mistaken and moreover unnecessary. Your opinion is fine as it is. Everyone's is. But it's not a fact.
Of course, everyone's views on Android are different. While I may have some threshold for the number of positive articles that would be necessary for Gruber's writing to be useful, someone else may have some other threshold depending on their priors. For me, the threshold is quite low because I do not like Android at all, and Gruber falls short of even that low bar. I don't think there are many who can make significant updates based on Gruber's work, since P(negative review of Android feature) is nearly 100%, even for features most people consider good.
In other words, Gruber's writing is entertainment, not evidence. The post you were originally commenting on, as I read it, was based on the assumption that Gruber's writing ought to aspire to be evidence, and thus lamenting the fact that it's not useful for most people. Personally I don't believe in the idea of "ought," but I think that asking for a quota is a sort of intentional missing of the point that the previous person was trying to make.
