> If that's so, doesn't the defection to an opposing camp make his message even here more credible? All you really need to judge a Tom Friedman article is the

> If that's so, doesn't the defection to an opposing camp make his message even here more credible? All you really need to judge a Tom Friedman article is the "Opinion columnist" badge by his name. What does he produce? He produces opinion. By what metric are his opinion pieces judged? Engagement. Tom Friedman - and for that matter, most other opinion columnists (excepting those who are practicing domain experts who occasionally write about those specific subjects) - isn't paid to be right. He can be wrong over, and over, and over again. If the orthodoxy of the group who employs him changes (hey, look who's buying NYT subscriptions these days...) he can be wrong in novel ways. He doesn't have to "show his work." He never suffers any consequences. He might have re-evaluated things! He might even be right - because hey, a stopped clock is right twice a day! But knowing what his job is and how he's performed in the past, it should be apparent that there's no particular reason one should believe him over, say, a random commenter on HackerNews such as myself. He has no incentive, credentials, or history that suggest a reader could usefully appeal to his authority.