experimenting in groups: the quiet voices.

One thing Laurie and I have been investigating over the years is how to bring forward the quieter voices in a community. In most communities, there’s a group of folks with fairly thick skins and a willingness to enter the mosh pit to get their comments heard — and a second group with lots of interesting things to say, but not as willing to elbow in and make themselves heard over the noise…

One thing Laurie and I have been investigating over the years is how to bring forward the quieter voices in a community. In most communities, there’s a group of folks with fairly thick skins and a willingness to enter the mosh pit to get their comments heard — and a second group with lots of interesting things to say, but not as willing to elbow in and make themselves heard over the noise…

We’ve found that second population to be as knowledgable and interesting as the primary population — just quieter. So one of the things we’ve looked for is ways to bring them forward into the community discussions. (it should be noted that when we’ve talked about this issue in our communities, the normal response is some variation of it’s no big deal, just start talking — and that normal response comes from the folks who don’t mind the mosh pit atmosphere in the first place, of course…)

One of our more successful ways we’ve found to draw these people out, at least temporarily, is a concept we’ve called lurker day. It’s fairly simple, at least conceptually. On a specific day, the regulars haul themselves off to the sidelines and sit on their hands for 24 hours, and leave the group to others. Our guideline for lurker is pretty flexible (averages 1 posting a month or less), and is generally self-policed.

it works quite well — you get a whole new set of views of the topics of the day, and the regular posters (for the most part) take the vacation with good spirits. Occassionally this encourages a lurker to join the mosh pit, but for the most part, they enjoy their day on the list, then head back to the sidelines.

For me, that’s a frustration. lurker days work — but you have to schedule and manage them, and in my mind, it doesn’t really solve the problem. It merely makes it clear there’s a problem to solve, which is how you get those people contributing on a regular basis, not on lurker days. And how do you do that without screwing up what you already have? Because it’s not broken, after all, you’re looking for improvements, not changes.

I like to use real-world analogies to help define virtual situations, because it helps me translate what we’re trying to build into known environments. Analogies illuminate both by where they work and by how they fail — and where the analogy doesn’t work is where I try to focus my ideas and experiments.

My favorite analogies for a mailing list (especially since many of mine are sports oriented) is the sports bar and the cocktail party. I especially like the cocktail party concept because it includes the concept of small groups of users
having independent but (sometimes) interrelated discussions. The advantages of a mailing list is that you can listen in to all of these discussions at once, not just the one you’re in (and you can be in multiple conversations at once!). the negative aspect of a mailing list is that, effectively, when one group is having a discussion, all of the other groups have to shut up and more or less pretend they’re not listening — good conversation requires a sense of intimacy, which is at odds with the realities of a mailing list. That’s one reason people lurk — they can never get comfortable with the facade of intimacy, they always see the audience out beyond the lights. Most conversations are between two and six people, from what I’ve seen. While there are always one or two people who seem to revel in being in almost all conversations (I certainly wouldn’t ahem know about that…), you’ll find if you study list traffic that conversations tend to be serial, and different conversations tend to attract different sub-groups, and once a conversation starts, the other chatter usually dies out and waits for it to finish.

So ultimately, instead of a cocktail party of intermingling conversations, a mailing list is more a serialized conversation among intersecting sub-groups. In essence, it’s a cocktail party where conversation can only take place in the talking-place, only one group can be in it at a time, and everyone jostles around to step in when an existing conversation ends.

(this might explain a syndrome I’ve always wondered about: when a group of people on a list get into a topic and really start a lively conversation, someone invariably pops up and tells them all to shut up; it’s as if they think the list is there to be subscribed to, but not really used. But perhaps, what they’re doing is reacting to people staying in the conversation pit too long, and trying to move them along, albeit in a non-polite way…..)

So what we’ve been talking about the last week or so is ways to create more “talking pits”, ways to get rid of that first-in-first-out aspect of a mailing list, without actually screwing up what’s working.

Lurker day has proven there are groups of people who can really contribute to the discussion — if we can find a way to allow them to contribute on their terms; create them a conversation pit away from the mosh pit of the main list.

A concept we came up with over the weekend that I think has promise is this: to create a series of “extra” mailing lists (call them, say, sharks-1 through sharks-6). Members could subscribe to any, or all, or whatever. The use of those lists would be undefined, allowing users to figure out what they want to do with them.

The only binding rules are this: (1) no cross-posting among lists, and no re-posting from list to list. The idea is to create independent places for multiple conversations, as if you’ve taken the cocktail party and spread it among different rooms in the house. Allowing cross posting defeats the purpose, the same way a PA system would defeat it in the real house. (2) you can’t tell others to move to another room. you can only offer to take your own conversation to another room.

My hope/belief is that over time, different groups would migrate into different lists and self-define what those lists are about. I’d expect the main list to continue to be the driving force in the environment, the room with the bar, the band, the dance floor and the mosh pit. less enthusiastic users wouldn’t ahve to wait for the band to take a break to spread out onto the floor and talk any more. Instead, they’d wander off to one of the side rooms. I wouldn’t be suprised to see three or four sub-lists to be more or less permanenty co-opted by sub-groups for various purposes, with the last couple of lists being left for ad-hoc discussions (“hey! let’s take this over to sharks-5 and hash it out”).

and for people who just don’t want all that complexity, the original list is still there. It only changes to the degree people find the sub-lists useful and productive and worth the hassle of populating and using. so it seems like a low-risk experiment on top of everything else. If it works, you end up with a much larger, vibrant community. If it doesn’t, the sub-lists wither and die, and you can prune them without any impact.

I’m curious what people think about this idea. has anyone tried something like this? Has it worked? Anyone know of any research or variants on this kind of setup I can research? does it seem like it might be a useful thing to try?

I’m going to propose this to our sharks list (the official guinea pig list of plaidworks) and see what the reaction is. I’m curious how they, as a potential user base of this idea, will react to it…

(a final note: this is something I’d consider for large, busy lists that “need more space”. It’s not appropriate for smaller, quieter lists. One thing I’ve noticed over the years is that a list has to have a consistent level of postings to it to remind people it exists and is there to be used. If your user base is too small, you have trouble sustaining that and people get out of the habit of thinking about (and using) the list — and then it dies. It’s better to have one list with three small and somewhat related groups on it than three lists with three tiny populations, because those three lists will have great trouble building any kind of community due to the infrequent postings inherent in a tiny audience…)