In Defense of Comments.

Excuse our ignorance, but at what point did it become apparent that “inline blog comments are going the way of the BBS and Gopher sites of yore”? We understand there are certain inefficiencies in the traditional way of handling comments, including but not limited to spam, follow-ups, digressions, fragmentation, trolls, and idiots. Yet we fail to see how the tweet-as-comment paradigm resolves any of those issues.

If building a better comment system was the goal, iterating on the progress made by dedicated comment sites like Hacker News, Slashdot, Digg, Reddit, et al seems a much savvier plan than blowing it up and starting over. While we stop short of praising comment threading, voting, and flagging as panaceas, it does seem undeniable that Twitter solves none of those issues, introducing instead a host of its own.

By replacing comments with tweets, Happy Cog twice undermines its audience, first by contributing to the general Internet noise pollution (as well as the specific article comment thread noise) and second by trivializing the resulting discussion. Artificial brevity is a flaw not a feature. It makes substantive conversation if not impossible at least heavily discouraged. Ex-communicating the indefinite length, local comment as the fundamental unit of a larger intellectual discussion is inimical to Internet culture and, most importantly, learning — which we presume is a core value of any blog post.

But perhaps we’re unfair. We tremble at the thought of standing in the way of progress and stick-in-the-mud traditionalism is hardly our philosophy. Maybe the goal was to simply make it easier for people to leave feedback on the article. In that case, Cognition has succeeded wildly. It is a bold move (though perhaps a bit shy of “brilliant” as many commenters tweeters gush)  worthy of applause. If upsetting the rotting apple cart of blog commenting is one outcome, it won’t all have been for naught.

Our metaphorical fingers are crossed that Happy Cog instituted this system not as part of a grand PR scheme but in the sincere belief that it offered at least the chance of a better model by holding commenters accountable and elevating responding blogs to first-class comment citizens. If that is the case, we humbly suggest supplementing Twitter and blogs with good, old-fashioned comments à la Disqus because, on some level, a well-considered comment system is irreducibly complex. It requires a spectrum of in situ response lengths to adequately simulate real conversation.1 By striving for “simple” we fear Happy Cog has strayed into “simplistic.” Unless changes are made, commenting on a Cognition post is an exercise in futility, home only to well-meaning pats-on-the-back and vapid gestures.

Let us be clear: we have no objection to decentralizing, extending, or otherwise improving the conversation. We do, however, oppose its infantilization.

In Zeldman we trust.

  1. And, if beggars were choosers, things like voting, threading, burying, author highlighting, comment email or RSS subscriptions, etc. There is much room for improvement. []

Here Be Comments