No, the arbitrary number is this 2-100

I think the basis for this range is valid in this vote. It covers all of the reasonable numbers that may be considered for MVT, even the totally wrong 50 option.

I would be interested to see how many posters there are in Buffy, Angel, Firefly, whatever. But that's just because I love data.

I feel like the most effective thing at this point would be to make your arguments for whatever number, and then people can read them, and if they buy them they can put that number.

I can see the argument for a 20ish number, but disagree with it. The real fact is, I can't imagine an issue that only 20 Buffistas will care enough about to pass judgement. We've regularly gotten 100+ votes in thread-naming polls. We love to express our opinion! Especially if it's as easy as a couple of clicks.

I think if only 20 people care enough to vote on something, there's something seriously wrong with us.

All right. What I would like to know is, on average (say for the past week) can we tell how many unique users posted at least once a day in a given thread. Or even, rather than average, pick a day out of the hat. On Tuesday, can we tell how many unique users posted at least once in Natter, Bitches, Buffy, Angel and Firefly?

I think if only 20 people care enough to vote on something, there's something seriously wrong with us.

But if we set up our system so that the conflict-averse among us don't even have to vote "no" to shut something down, I'm afraid that's what will happen.

I'm sure someone could find that out. It would probably take some time, and I'm not the person to do it, but I'm sure someone could. I mean, you could do it manually, by figuring out the post range you wanted, threadsucking, and then playing around with the posts you've just sucked.

I think if only 20 people care enough to vote on something, there's something seriously wrong with us.

What would be wrong with us, and if it were, how would one fix it?

Personally, I think if that thing is wrong with us, so be it.

ita, is it possible to tell how many user ids logged in? It makes sense for a regualr lurker to log in, and while this wouldn't tell us which threads they read, it would go some way towards 'regular user' numbers.

What would be wrong with us, and if it were, how would one fix it?

Personally, I think if that thing is wrong with us, so be it.

Yeah, I don't know. I was just making the argument. I don't have much deeper thoughts on the whole thing. I wanna see more arguments!

ita - an infinite number of posts back, didn't you think that if someone proposed we change the board's language to Swahili and only 8 of us could be arsed to vote, and 5 of us voted yes, then the board should be in Swahili?

I'm asking because, I think if we give Buffistas a range between 1 and 100 (or 2 and 100, or 10 and 100), we're going to come up with an average of 50 to 60. Theory of central tendency blah blah blah.

I don't know whether or not requiring 50 people to show up to vote before a vote can be counted is going to be a reasonable number. Maybe it is. Maybe it's an underestimate and maybe we have 200 unique users posting every day in the more popular threads. Maybe it's an overestimate, and we only have 50 unique users showing up in the popular threads. I have no confidence that averaging our favorite number out of one hundred is going to set a decent threshhold.

My own preference happens to skew low, but I would feel the same way if it skewed high.

Oh and the reason posters count more than lurkers in cases where we're voting on a new thread, is because lurkers don't post, and threads need posts to be active.