On the nature of comments

I've written before about how disappointing I find most of the Web 2.0 talk to be (I'm defining it in the most common way, as the new user-generated content of blogs, videos, comments, etc. that actually have the populace "creating" their own reality on the Internet - or something like that). I'm not an extremist, I grant that some of it has been useful in terms of forming communities and making information more broadly available, but many of its successes are not so much revolutionary as extensional. That Obama raised so much money over the Internet implies that his people saw it as a more effective marketing channel and made it happen, which does not reflect a re-ordering of reality.

Obviously I'm not going to say anything against blogging, but there is evidence that consolidation is occurring, that it's getting harder for voices to emerge from the vast pool and be heard. This is in no small part a matter of time; I barely have time now to post a time or two a day and read the blogs I already have in my feed reader and do the other things I need to do, so I'm certainly not out cruising Blogger for new people that I then would have to keep up with. It certainly seems clear that there is this big clump of sites in the middle of the Internet, or maybe it would be more accurate to say a clump of clumps differentiated by subject matter (I'm not sure how much intersection there is between the political blogs and the entertainment blogs). Those of us not in one of the clumps have a one-way relationship to the big boys and girls, and there's little hope of that direction changing, probably less of being invited to join the clump. (I've seen some graphs that purport to demonstrate the clumping, though I am unable to find any at the moment. What would be interesting is if they would show directionality, so that we could see quite clearly that Androcass is way more likely to link to Andrew Sullivan than Sullivan is to Androcass - oh, well, his loss.)

And comments, oh, comments, the vast cesspool of profanity and scatology and various unsavory -isms. Not all comments are vile or unhelpful, but far too many are, as I discussed here in a post about Chicago Tribune columnist Eric Zorn's decision to turn them off, citing the time it took to handle and reject the out-of-bounds ones.

However, even in their most benign form, comments suffer from some structural deficiencies that make them fail to live up to their promise to provide a forum where well-intended people come together to mull over the ideas expressed in the original post.

First, there is the possibility that the "relationship" between the original poster and the commenters will become unidirectional, with the poster never actually dealing with the comments that the post generated. For example, I have stopped by Charlie Rose's comments section several times, and never once have I seen someone from the show, let alone Charlie, respond to anything there. That may be excusable - would we rather have Charlie writing comments than preparing for his program, maybe wiping down his "table"? (I have wrestled with the priority dilemma myself, as I talked about here, and I don't get many comments at all.)

There are those who would say that it is unnecessary for the instigator to remain in the conversation, that the "community" can maintain the flow of ideas, but it still seems like a cheat somehow ("I've invited you to my home, now I'm going to go up to my study and lock the door...have fun!"). And, as I've pointed out, the "community" is far too often co-opted by toxic notions.

Second, there's the matter of thread structure. If you've ever tried to hold an organized discussion via e-mail, you've experienced this. Response A comes in, response B talks about some of the issues in A, response C was written before seeing B so it rehashes some of B's points and ignores others, and so on. By the time you get to response J, you have 20 issues on the table, but only three or so which are "hot," and you end up having to schedule a meeting to resolve these threads.

Some blog software tries to deal with this by allowing the discussion to be organized into formal threads, but this solution relies on the ability of commenters to self-organize, something that works far better in theory than in practice. Other blogs or sites rely on human intervention to spin off new threads whenever necessary, but any popular site soon becomes a nightmare for that human.

I bring this up because of a curious experience I had yesterday. Kevin Drum wrote a post in which he solicited comments on more blogs to read (I know, running counter to the trend I mentioned above, but Kevin may just have more time than I do). I saw it, listed a few of my favorites (sent another, because I forgot one, but that's not material to our story), and posted. His comments are moderated, so I expected to wait to see them go up, but I went back a few times, and didn't see them. In the meantime, comment after comment is going in. About four hours after the initial post, mine still aren't there, so I write another post to see if anyone has any insight and...boom!...it goes up immediately, no delay at all.

I figure, glitch (though this has happened a time or two before on Kevin's site), and I turn off the computer and go about my business. This morning I hop on, and there are my two posts, in the time sequence I wrote them, apparently finally pasting muster with the Drum process.

But here's the thing. No one is ever going to see them. In the example I used above, it's as if response D was written, then it mysteriously drifted about the ether while responses E through Q went up, then it's placed back in the queue as response D. If you have an interest in the discussion, you might monitor the posts to see what there is that's new, but you're not going to go back up to see if something's been inserted between C and E.

It would, of course, be equally strange to put new messages in the queue at the time they're approved, as they'll seem totally out of context, creating random discussions. About the only solution I can see would be to hold all messages in abeyance until the fate of each one is determined, in sequence. But then I can imagine a lot of people wandering away, figuring the discussion had come to an end.

I guess the real solution is not to take 4+ hours to decide whether a message is suitable, but to try to make that approval process happen in real time. So I wonder what Kevin is doing where that isn't happening. It all makes it pretty difficult to feel a part of a discussion, and, once again, demonstrates that we still have a long way to go before realizing the potential of Web 2.0