Well, I'd guess it represents a much larger segment than you guess. Neither of us has any data, nor does the article. So the only rational conclusion is to acce

Well, I'd guess it represents a much larger segment than you guess. Neither of us has any data, nor does the article. So the only rational conclusion is to accept our ignorance, right? Or at least one should carefully document every single wild guess being made when coming up with some numbers, and carefully identify them as very plausibly being very large sources of error (so that no readers mistakenly doubt that the numbers might be off by a factor of 10 and think it's actually based on hard data).