I'd be curious to hear an explanation of when you should or shouldn't do this. It's interesting to see stackoverflow as an example of bowing to community wants, because stackoverflow seems like the ultimate example of not doing that. The most upvoted topics are closed for being the wrong kind of content, and most of the questions that were, historically, the mosted upvoted would now be closed before they could pick up steam.
I can't be the only person who notices this; pretty much HN post that features SO includes a litany of complaints about them. Even if the original post isn't a complaint, the top thread in the post will probably be a complaint, regardless of how relevant that is to the topic.
My point isn't that what SO is doing is wrong. It's that what they're doing seems to go against the wishes of the vast majority of the community. Jeff even links to a post of his that basically says "you shouldn't always listen to the community". So, when do you listen and when do you ignore? Advice that consists of "sometimes you should do X and sometimes you shouldn't" seems trivially true for pretty much any real-world X.
I don't think Mr. Atwood is suggesting that you bow to what the community wants, but rather that you implement systems to facilitate what they're already doing anyway.
That aside, and regardless of if SO is going about topic policing in the right way, I think this bit from a guest post[1] by Randy Farmer might hint at why they're doing what they do.
"As user-generated content grows, content moderation of some sort is always required: typically, either employees scan every submission or the site’s operators deploy a reputation system to identify bad content. Simply removing the bad content isn’t usually good enough-most sites depend on search engine traffic, on advertising revenue, or both. To get search traffic, external sites must link to the content, and that means the quality of the content has to be high enough to earn those links."
It's the most virally popular off-topic content that you need to moderate. Off-topic content that doesn't get attention doesn't do much damage, and popular content that is in line with what you want the site to be is good. But popular questions that threaten to absorb huge amounts of attention at your site, promote unproductive threads of conversation, and attract karma whores and trolls, those need to be closed.
Ironically, I don't think Stackoverflow does this very well.
It's a good article, and the point of researching every wheel ever is a good point well made.
But as a long-time Stackoverflow user, I've been frustrated time and time again by cases where extremely valuable content emerges on StackOverflow, but then the moderators come along and kill it because "it is not the defined purpose of this site."
Now they have the right to define the rules & purpose for their own site. It wouldn't bother me so much if the content they were killing wasn't so fantastic. But I have seen so many deep, excellent, rich blobs of technical content get cut out and cast aside, for the sake of adherence to some superficial guideline.
It seems to be exactly the opposite of "seeing what your users are doing, then helping them do it."
I'm not sure if you're just intentionally being ridiculous or mistakenly thinking that you've got a good reductio ad absurdum argument going on. In any case, why exactly are you conflating "wildly popular" with "extremely valuable?"
Even if the grandparent post was about popularity, your statement is as unconstructive as this argument would be:
A: "You're intentionally putting way too many non-criminals in jail."
B: 'Ah, but for the purpose of setting boundaries, consider the extreme case where we don't put anyone in jail at all.'
Back in the day I spent way too much time keeping up with the MUD-DEV mailing list. There were a lot of discussions that touched on these topics, although not always as constructively as Atwood does here.
Here's an old post by Raph Koster that talks about how to deal with kewl d00ds. I was reminded of it because Koster referred to them as genetic algorithms with respect to bug/exploit finding. It's a bit of a tangent but there are some similarities.
"We noticed early in the Stack Overflow beta that users desperately wanted to reply to each other, and were cluttering up the system with "answers" that were, well, not answers to the question. Rather than chastize them for doing it wrong – stupid users! – we added the commenting system to give them a method of annotating answers and questions for clarifications, updates, and improvements."
This is excellent advice not just for online ventures, but for business in general. Too much time is spent making rules and regulations to prevent customers from taking certain actions, when a lot more can be gained by understanding what your customers are trying to do naturally, and finding a way to support that.
Curious, is there any video of expert D&D players playing? I used to be a big DM in my area back in the day (probably around 1981/82), but since the internet wasn't as prevelant there was a lot less communication about how the game was supposed to be played (and I wasn't in a large metro area -- so I pretty much set the rules for the town).
To this day, I'm curious as to how the game should actually be played.
I'd imagine you'd be hard pressed to find a video of "expert players" playing D&D. And even if you did, I really doubt it would be overly thrilling to watch.
For what it's worth, RPGs are a rather specific thing, since they can only work if the given group works as a group. This form of gaming is a communal effort, a DM who doesn't adjust to what his players wish for will soon be without a group. Both the DM and the players are supposed to have fun. They might need some nudging here or there, but overall, they are just as much part of the process.
Most questions in any RPG forum I've ever seen center either around meta-gaming questions (rules and how to enforce/circumvent them) or about how to deal with situations that arise from a specific group constellation. If you have a group that likes the fighting but dislikes riddles, a DM avoiding juicy fights and drowning them in riddles does them a disservice. Most problems however arise from less homogenous groups, where some people fancy one thing while others would like to focus on others.
Ultimately, every group has to find out for itself what does work and does not work. Other people can give you hints, propose changes, improvements or overall just tell you about their experiences and how they deal with problems, in order for you to be able to compare their problems and solutions to your own.
[edit] Something like stackoverflow works in much the same way. The community can propose solutions to a specific question you put forward, based on their own experience and ideas, however, you will ultimately be responsible to figure out whether the proposed answers are right for you and your specific contexts and experiences.
Wizards of the Coast often puts up videos of celebrity D&D games: http://www.youtube.com/user/DNDWizards. There's a televised game at PAX every year, for example.
I can't be the only person who notices this; pretty much HN post that features SO includes a litany of complaints about them. Even if the original post isn't a complaint, the top thread in the post will probably be a complaint, regardless of how relevant that is to the topic.
My point isn't that what SO is doing is wrong. It's that what they're doing seems to go against the wishes of the vast majority of the community. Jeff even links to a post of his that basically says "you shouldn't always listen to the community". So, when do you listen and when do you ignore? Advice that consists of "sometimes you should do X and sometimes you shouldn't" seems trivially true for pretty much any real-world X.