> Was this addressed in the article? Howard brought up a really interesting point, to me, and I just read the article, but I don't remember it mentioned there.
>
> But, basically, RinkWorks regulars are here for the other regulars. We love your site, Sam, but I don't think any single one of us (even Sam himself, if I can make that assertion) can say it's the sole reason we visit the chat room or the message board. We come here for each other.

This was said in the article in different words, and perhaps less directly. He gave the example of a Clairol (I'm not sure if it was a real or theoretical example) message board, set up for the purpose of discussing Clairol, owned and paid for by Clairol, but once a social community formulated (or would formulate), rules be darned, that social group is going to carve out a way to sustain itself or go somewhere else, and discussing Clairol and only Clairol is not sufficient. "The users are there for one another."

Well, originally this message forum was set up for the discussion of RinkWorks content, and if you look at the first couple dozen posts on the board, this leaning is apparent. But it took no time at all for that to change, and while I was not conscientious about this at the time, I wonder if I inadvertently sidestepped a pitfall by not insisting that posts remain "on topic." It actually never occurred to me that I might do that, especially since I had already set the precedent *before* the forum came into being by holding moderated discussions/surveys in the Site Journal that were unrelated to RinkWorks content.
I suppose it helped that the scope of RinkWorks is so broad anyhow, as you say.

Another thing that struck me as a reason the social aspect of RW has thrived is the heavy moderation. This was encouraging to me, because this is easily the most wearisome aspect of my duties here. I don't *like* deleting posts or (shudder) explaining to disruptive chatters why I'm kicking or banning them. But this article suggests that these actions are critical for the success of an online social group: the core group needs a way to defend itself. And so these actions of mine, and those of RinkChat ops, and the very fact that we have something as ugly as an op hierarchy, are probably critical factors in the RW communities continued success. I always thought so, but it's nice to see someone as studied as this guy confirming that this is the way to go.

Since reading that article, I've been thinking about how the technical aspects of things might be changed. I'd love to remove the op system and have an automatic way in which the chat engine itself dynamically determines who needs opping when. But as the article teaches, the technology cannot be separated from the social aspects, which means that such a thing left up to the chat engine is impossible to do correctly unless it bases its decisions on more than the chat engine can know (like when the user registered; how often the user logs in; how much the user participates, etc), and so it would have to be based, at least partially, on the input of other members of the core group. But who's the core group? I'd have to seed the system with who I think is core group and who isn't, and then statuses could fluctuate up or down from there. It all sounds too complicated to me, especially to fix something that, while it may not be perfect, essentially works. And one of my goals would be for me to skip out on making these types of calls in the first place, but it's fun to think about how other types of systems that both moderate and "reward the core group" might work.