A couple of very interesting complementary items appear on the blogroll:
Fergus has a lament that is very similar to the ones that finally got me starting blogging, here. It appears that reason is not enough.
Overcoming Bias is a site that advocates that people think more like statisticians. This can be difficult. Statisticians tend to be more like snipers than collaborators sometimes, alas. It's very easy to get criticism from them, but much harder to get good advice. I suspect that the nature of the field makes bemused contempt a natural response. Nevertheless, I certainly aspire to think like a statistician, and one of the better pieces of advocacy for that approach is currently on offer at Overcoming Bias.
Both pieces seem to come to more or less the same place via very different routes. Probably the best we can do is to be open-minded ourselves, they suggest. Maybe so, but I'm afraid that may be too little, too late. We need to change the culture at large, not to be greener or kinder or gentler, but to revive the value of intellectual honesty.
Not entirely tangentially, one of the reasons I advocate programming as part of basic high school education is that you cannot cajole, bully, bribe or whine your algorithm into submission. I wish more people simply had the experience of right versus wrong at this fundamental level. We now have a tool at our disposal to make that possible.