My prediction is, as new members join, new levels of stratification will develop. There will always be the top tier which we see beginning to form now for a few reasons. The obvious cause is that they are top tier for a reason- they post a lot, and they are usually quality posts. This draws in more followers, and thus they stay on top. Another reason is related, though somewhat different. Due to the amount of posts they make, they are often the first people new users encounter. These new users are typically looking for people to fill up their feed, and as such they follow whomever they first encounter. Likewise, there will always be the lower tier of people for much the opposite reasons as I just listed. While this tier of users may still post and comment, it will not be the same quality or quantity as do the top tierers. Some of this tier may not even post and instead just use hubski to follow people, which is fine. I predict that intermediate tiers will develop. This tier will consist of people who post similar content as the top tier, but does so in a limited interest range. Obviously if someone is posting large amounts of quality content related to both science AND politics, they will acquire followers from both interest fields. However, if a user were to post in only one of these fields, it stands to reason that they would only attract followers from that field. It is these types of users that will make up the second tier. What remains to be seen is if mobility through these tiers is easy, or if new users remain in the bottom tier unless they become superusers.
what I would like to see is a non-small world graph.
I hope you reexamine the data in several months. We are gaining new members at a pace that we haven't seen in our history. This may be a phase transition in which the data haven't found their water level yet. My assumption is that they will smooth out in time.
There's not much that would make me happier than to have thousands of people with more followers than me on Hubski.
Big reddit post brought in a lot of new folks, folks who were looking for reddit names to follow, thusly we get kleinb100, syncretic, possibly Saydrah, and definitely not myself getting a lot of followers, changing the curve a bit.
kleinbl00 seems to be the biggest "beneficiary" of the reddit injection and is the distorter of the curve.
I think that this has more to do with the fact that Hubski doesn't have a lot of ways to find new people to follow who match you. At the moment, since mk, thenewgreen, kleinbl00 and a few others have the most popular posts to share theirs are the most widely visible, so when a wave of people joins they're the first faces. I'm also not sure if this kind of stratification is necessarily a bad thing.
Hey, this is a side question: Are your observations taken from actual data? If so, how did you get it, and can you release it? If not, how are you drawing your conclusions? I'd be interested in how you came to your observations.
Awesome! On a side note, I am somehow the second most popular commenter behind Saydrah! What an interesting discovery! I really don't quite know how I ended up with that :)