Over at The Monkey Cage, Elizabeth Suhay wrote some much needed perspective, Comment Threads Are Messy, but so Is Democracy. There has been a push in a lot of quarters to get rid of user comments. It is understandable. A lot of commenters are repellent. But I think it’s a major mistake. What’s more, I think it reflects a kind of desire on the part of content providers to go back to the old days when they could sift through the letters and publish the ones that they wanted. It’s about control, and it doesn’t speak well of those who push it.
Some time back, the excellent blog The Incidental Economist stopped allowing comments. Now they treat their users the same way that magazines used to. If you have a comment, email them and if they think you are worthy, they will add the comment. In addition to the “Moses coming down Mount Sinai” arrogance of such a policy, it just isn’t practical. A few years back, I emailed one of their writers about the statistics in one of his articles. He took over a month to get back to me. But even if the response was timely, it would take at least a day before such comments would be posted — long after most people had read the article.
This approach also eliminates the possibility of what I consider the best part of comments: conversations. Comments to articles often end up being even more interesting than the articles themselves. They also make the readers more engaged with the material. There is no doubt that The Incidental Economist is no longer as exciting as it was when it had comments. Look at Eschaton: it is little but comments and is one of the most vibrant websites around.
Getting rid of comments strikes me as an overreaction to a problem. Sure, there are jerks who post comments. But the numbers are small. Suhay reported on some of her own research that found that “clearly disrespectful” comments only made up 10% of those found on sites like Daily Kos and only 4% of comments on sites like The New York Times. And for that, people want to get rid of all the good that comes from comments? That strikes me as, “Letting the terrorists win!”
What I think is going on is that content providers are thinking of what they do from their own perspective and not from that of their users (ie, customers). In fact, another article in The Monkey Cage found that comments make people trust articles less. The article concluded “news outlets that care about their reputations (including The Monkey Cage) should shut down their comments sections.” That shocks me. That’s such an authoritarian thought. Obviously, when a bunch of people openly debate an article, it is going to make that article seem less authoritative. And by and large that’s a good thing!
Under most circumstances, I don’t read comment threads. I have my own blog; if I want to comment, I will write an article. But I do find comments useful. If an article strikes me as using questionable logic or facts, reading the comments can be really helpful in corroborating or diffusing the article. And that’s especially true in reading about specialized subjects like economics. The comments are often of shockingly high value.
But I get it. Before moving to WordPress, despite my automatic filtering, I still had to manually remove a great deal of spam. Also, I had to explicitly approve all comments — even people who had commented before. But given all that is done by using free software, I don’t have any sympathy for far more successful blogs on this issue. And it is annoying to get certain kinds of comments. Personally, I don’t mind people yelling at me. The one thing that does bug me is when someone yells at me without having read (or understood, at least) the article I wrote. (Here’s my favorite example: Two Thoughts on Lars and the Real Girl.)
Ultimately, the push to destroy comments is just about the desire to control. I do understand why the folks at The Incidental Economist and Hullabaloo just wouldn’t want to deal with it. They can be forgiven. I think it is wrong to even discuss it at The Washington Post. But above all, it is dismissive of your readership.