I uploaded the png file with the histogram to http://xs.to/,
which is a free image hosting server.
However, I don't know for how long it can be kept there. If you have a permanent web server, please copy the image there and change the link. Thanks.
gedh: Is the graph really number of players v. rank, or number of "accounts" v. rank?
Reuven: Number of accounts is more likely as there's no way to keep track of users. Actually I found it surprising - Most people I know, took a couple of months (for the more serious players but nobody took more than half a year) to go past 10k - People don't stay 15k for too long according to my expirience which'd hint that most of the players/accounts on kgs are cycling - Old users leaving, making room for new ones... In other words, my view of the ranking world has shattered as it's extreamly unlikely...
In a way I was lucky enough to be in a light-go-school-like envirement on the server for a while but it only seemed natural as we were trying to spread it (Somewhat in the way the kgs teaching ladder works)...
So the progress is quite slow on kgs in average... Is it only kgs that's like that? How does it compare with other servers, real life and perhaps go schools? It'd be kinda interesting if somebody would write about their expirience in different envirments...
ggleblanc I've been on KGS (as GilbertLB) for over five years, and I'm still a 16k. Not everyone marches upward through the ranks.
I agree that new Go players come to KGS regularly.
tderz: Median, I assume, that you produced the diagram, and Nachtrabe, that you improved it aesthetically. How did you produce the diagrams form the raw data? Which tool did you use?
nachtrabe: I used R to both create the graph and find the median value. This is the basic code for the graph:
barplot(ratings[1:39,1], col="white", axes=F, space=0)->a axis(1, at=c(a, a, a, a, a, a, a, a ), labels=c("30k", "25k", "20k", "15k", "10k", "5k", "1d", "6d") ) axis(2, at=c(51, 250, 500, 750, 1000, 1250, 1500, 1746) )
Note that 51 and 1746 are the minimum and maximum values, respectively. Also note that the data table "ratings" has both the zero rating and the NULL rating groups removed.
tderz: Sorry for my ignorance: What and where is that program R?
melgo?: this discussion is hopelessly outdated. it should probably be removed.