1. 9

When communities get too big, then they fall apart. Today, I was just thinking to myself: “What would happen if Lobsters had a user limit”? That got me poking around the Internet.

Lobsters has 3293 registered users today: http://lobste.rs/u

Here’s an interesting series of articles about online community size: http://www.lifewithalacrity.com/2009/03/power-laws.html

What do you all think community size? Obviously, too small is bad. But what size is too big?


  2. 8

    Lobsters will be at its best just before eternal September, just like every online community. Trying to prolong the inevitable - and it is inevitable, eternal September comes to everyone - will just snuff out the community in its prime.

    Everyone should take a page from MST3K… to paraphrase (bastardize):

    “And if you’re wondering how we stay on topic and other Lobster facts… Then repeat to yourself ‘it’s just a site, I should really just relax.’”

    In the not so distant future this same discussion will be happening somewhere else, so I say let’s just enjoy the ride while it lasts!

    (La la la)

    1. 9

      It doesn’t happen to everyone. Many communities fizzle before eternal september, many communities stay the same people they started with, and then there’s MeFi.

      I think the invite-only policy is doing its job properly, and we’ll be able to slowly add members for a long, long time.

    2. [Comment removed by author]

      1. 4

        I personally don’t think registration shutdown should ever happen, firstly, because the invite system works pretty well, and secondly because once it happens Lobsters will become a less open community and more of a clique. The way I see it the users will then begin to view themselves as somehow special because they are members of this site, that somehow that makes them smarter than their colleagues, when really all it means is that they managed to get in before the registration shutdown.

        Also, the current users will eventually move on, become inactive, die, or change interests, so Lobsters could die very quickly after the reg shutdown.

      2. 5

        The question is not ‘if’ or ‘what size’, it’s ‘when the community starts to degrade through having too many members, how can we restructure it to regain and restabilise this productive, friendly mode’. I’d suggest there’s only one answer to this, and it is mitosis, and this to needs to happen very close to what our ‘dunbar number’ is for this kind of community.

        If we don’t allow a community to split into more communities, each of which can re-kindle this ‘small community’ feeling, then reddit.

        1. 3

          It’s funny that you should mention reddit, because they have basically split their initial community into a large number of sub-communities (subreddits).

          1. 1

            Yeah, I was being a bit harsh, I guess I was thinking of the reddit home page, and how these communities that blow up out of proportion end up catering to the lowest-common-denominator. sub-reddits would be an interesting case study.

        2. 5

          Well, I’m inherently suspicious of any site that let me in. /s

          Actually, one of my favorite things about Lobsters so far is the low rate of submissions and commenting. Most news aggregators encourage people to submit and comment as much as possible. This has effects that are good for the site itself, but neutral for the culture. It would be interesting to study how incentives (read: software) can be used to guide users to cultivating the culture of the site itself, rather than gaining traction by low-quality content.

          1. 4

            One thing is, you can’t just go by users, you need active commentors or submitters, I think.

            1. 3

              I think that communities that keep people coming remain good communities even as they grow big. I’d be happy if Lobsters was as big as Slashdot, or bigger - it’s got a nice UI and policies that I approve of. Certainly I’d like there to be far more users than there are now - the comment threads are currently rather short in comparison to other sites.

              1. 2

                Ive been really happy with the size that lobsters currently is, the articles and discussions are all very interesting and the level of noise is very low compared to other places I go for for news and information.

                1. 2

                  One of the problems is that if you go through a period of rapid growth and your userbase quadruples in size, then 75% of your users don’t even know how much better the community was before they joined.

                  1. 1

                    Power Laws are all about Fractals.

                    ie. Every social media, online or meatspace, are subject to Power Laws governing activity and value of contribution. (Alas, highest activity and and value of contributions is only very loosely correlated.)

                    However, the same effects happen whether you clump or split the interactions. That’s one of the really really interesting things about Power Laws…. they arise mathematically from situations that are fundamentally fractal or “self similar at varying scales”.

                    So design your community to grow without bound, but allow it to be fractal.

                    1. Allow your users to split and/or aggregate it to suit them.
                    2. One man’s signal is another mans noise. The problem of “the forum is too large” is actually “the average man’s signal seems pretty much like lowest common denominator noise to me”.

                    The problem is the “like” button, it means the same thing to everyone, so LCD noise problem is inevitable.

                    Now imagine this resolution…

                    Imagine that every piece of content, whether a submission or comment had equal status. They were just “Vectors in a Very High Dimensional information space.”

                    Now imagine every username could associate a real number with every piece of content that indicated that username’s alignment with that content. (Or symmetrically, the alignment of that content with the users interests) (Vote inflation doesn’t matter, we will normalize the sum of squares to 1 anyway.)

                    Then a users interests are also merely a vector in the same very high dimensional space.

                    Then every user voting on every piece of content indicates his alignment relative to the content and to the other users.

                    There is a very nifty bit of mathematics to help us with this.

                    Singular Value Decomposition.

                    So the first component of all the User Vectors is merely the LCD hive mind view. The first component of the Content Vectors, will probably be Kittens with Captions.

                    This is the world of reddit.

                    Where it gets really very interesting is the long tail of lower order components.

                    Where your alignment gets weirder and weirder, the content that aligns with that component, gets more and more rejected by the main stream, who by down voting it, are actually are implicitly “upvoting” that content for those who align with the content.