NY Times doubles down on Internet comments

Weeks after NPR decided to give up on the notion that its online audience can intelligently discuss the day’s news via comments, the New York Times, which generally has a comment section worth reading, is trying another approach to elevate the scene.

It writes today that until now, it wasn’t “economically feasible” to moderate comments quickly. So the robots will now be in charge.

It provides a “test” to allow readers to see what it’s like to be a comment moderator.

What the New York Times community desk demands most of all is that some effort is made to justify your views. The desk must also be convinced that your intention is to inform and convince rather than to insult and enrage.

Like what? Here’s the answer code.

Rejected: A comment referring to “Repugnicans”:

Typically, insults against entire classes of people will be rejected by moderators. Times moderators often walk a fine line between encouraging open discussion and keeping the comments from devolving into the food fights you see on many other sites.

Rejected: A comment in which the writer says he/she will keep her thoughts to herself:

For example, comments that state only “Why is this news?” or “Who cares?” are rejected if they lack any further explanation. A commenter should tell readers why they believe what they do. Criticism of reporters or of The Times is encouraged, but should apply directly to the article in question. This way, we can prevent comment threads from becoming so repetitive that they discourage people from reading them.

Approved: A comment on same-sex marriage violating God’s intention:

We would have approved this comment. The commenter is making an argument against gay marriage that reflects common language used in the public sphere.

While this comment does contain content that may be insulting to a broad class of people, we must allow reasonable space for the expression of every point of view that is addressed by our reporting. This policy allowing for every point of view to be represented can lead to unexpected consequences. For example, The Times does not typically allow conspiracy theories in comments — unless the article is about conspiracy theories. Article comments are meant to further conversation and debate on the foremost disagreements about our society. Moderators should try to avoid making moral judgements on readers’ arguments.

Rejected: A comment that the Middle East should be sterilized.

We would have rejected this comment. It is effectively an insult to an entire class of people, and recommends presumably violent actions against them.

To be considered at all, this comment should explain why the commentor believes the people of these places should be sterilized. However, a comment making a detailed argument for something that might typically be considered racist may be published in some cases due to our internal mandate not to edit the conversations that take place on the site, but to protect them. This comment does not reach that standard. In many cases, the approval or rejection of a comment relies on the writer’s specificity. For example, a comment saying “Antarcticans are too violent when they protest, and it’s a sign of their leaders’ irresponsibility. These people have no real goals” would be rejected as an insult against a class. But if that comment began, “The Antarctic Liberation Front is …” It would be accepted.

Approved: A comment that women are insecure and fight biology via monogamy.

We would have approved. Whether we agree with a comment, we want to provide a space for interesting interactions to occur.

Put simply, the replies to this comment may prove to be enlightening for many readers. The comment does not criticize a specific woman, so it is not a personal attack. While it might be considered an insult against a class, alleging an “obsession” is not a particularly inflammatory insult. We don’t seek to police opinions, only to help as many people as possible express themselves on nytimes.com.

For the record, that’s the only one I got wrong. Or the only one the Times got wrong, depending on your view.

The Times calculated it would take me 71 hours to moderate its daily volume of comments, thus the need to go to a Google-based robot.

And, unlike NPR, the Times will be expanding the amount of its content subject to commenting, although it didn’t say how.

Ironically, people could not comment on today’s article.