All you need to know about Pirate-Community

Pirez said:
You're already in Brexitanny, Hunch : it's like the Bahamas, but the weather is always shit and the ladies aren't lovely.
Click to expand...
Well, for the second, I think it is quite the same in France and in Brexitanny.
Hunchman801 said:
Nope, the Bahamas have 0% personal income tax. Not quite the same here. :(
Click to expand...
I had bad times with the Panama Papers, Switzerland seems a better place to store his money now.
 
I think you mean "one's money". Funny how a mistranslation makes that post sound a lot more incriminating than it probably should!
 
Then I was right, "one's" was the word you were looking for.

(But apparently you actually were referring to yourself, to which I say: cool story bro.)
 
PluMGMK said:
Then I was right, "one's" was the word you were looking for.
Click to expand...
When Irish people are better at French than a native. :lol:

incognito said:
Pirez said:
You're already in Brexitanny, Hunch : it's like the Bahamas, but the weather is always shit and the ladies aren't lovely.
Click to expand...
Well, for the second, I think it is quite the same in France and in Brexitanny.
Click to expand...
What, women? I've lived in both countries and I can guarantee that there is a difference indeed.
 
Hunchman801 said:
PluMGMK said:
Then I was right, "one's" was the word you were looking for.
Click to expand...
When Irish people are better at French than a native. :lol:
Click to expand...
Not better at French, just at translating it into English.

BUMP: So I did a thing…
hist.jpg
 
Very interesting! That's quite a long tail.

Did you not try to match the data with a mathematical model? :P
 
Not yet! :P

Actually I should have used more bins, it's not very informative as-is. Also I think the code I used might have ignored zeros, I may need to revise it. I'll probably post another graph tonight.

EDIT: Yeah, the code I used to scrape the memberlist didn't pick up zeroes because they're not hyperlinks. Here's one with everything, and three times as many bins. As for model fitting, well, that can wait for another day!
hist3.jpg
 
The scale used in your chart is not very well chosen, maybe you could 'break' the first bar so that the others can be displayed at a visible scale?
 
I can't believe that's still the last post in this topic! Well, I was just thinking about redoing it at some point in the near future, so I'll see what I can do.
 
Well, I decided that given the quote in my signature, this would be a natural way to break up the histogram: :fou:
NonScavengers Histogram.png
Scavengers Histogram.png


I also plotted Tings versus Posts, to see how it compares to the "normal" Tings=Posts*5. As one would expect, the slope of a linear fit forced through the origin is more than 5. That outlier to the top-left is Canard WC.
Tings versus Posts.png


In both cases, the data have been partially corrected by taking into account some members who use multiple accounts. This doesn't include spiraldoor's (because even he couldn't keep track of them) and presumably some others. That doesn't really matter since I was mainly concerned with busting outliers in the Tings versus Posts plot, due to people like, say, iHeckler and Adsolution, who have lots of Tings stashed. Didn't seem to affect it much though!

(Also what the hell is going on with the scaling of the attachments??)
 
Nice job! It's great to be able to visualize all that data.

Just a couple remarks:
  • Your histograms could benefit from using shorter intervals, as they're not very descriptive for lower post counts.
  • The first chart would be more readable without users with zero posts.
  • Xowar is Raylex, you might want to take that into account.
 
I thought Xowar might be somebody all right!
Well, the good news is that I actually saved the dataset this time, so I don't need to re-scan the entire member list. The bad news is that I left it at home, so I can't make these adjustments till Friday night! :oops2:
 
You must log in or register to reply here.
Back
Top