So I did some scraping and nerd stuff and came up with data about if we only have 20 active members. Anyone care to see my results?

Needs moar Astrochuck.

in all seriousness, my initial reaction is that there’s no one I find irreparably annoying (which means there aren’t some names here that used to be), so that’s good.

No shit.. 2 posts. What a tosser.

They turn seven, and it’s like you never existed.

If the list was started from the top down, taking the top 20 active Flutheronians, you would miss the cut by one member; did you realize that?

Nowt like facts to blow random guesswork out the window.
Had a feeling you’d come up with something more concrete, good stuff.

Man, that was the push I needed to go to the doctor about my insomnia.

HA! I’M THE THIRD!

Guess my machine gun did a swell job after all ~

Thanks @johnpowell. You completed the active jellies’ hall-of-fame for me.

You can run this same analysis at other times and plot member life changes. For example, I know one member who was here a lot but just had a child. My participation waxes and wanes depending upon how close I am to a significant delivery deadline. If you see a significant change you can infer something is going on.

This is another example of why I consider you one of the brightest bulbs on the tree.
Thanks!!!

Thanks @johnpowell. I’ve always imagined it would be quite satisfying to have direct access (read-only of course) to fluther’s db. The amount of analysis and reports we could generate pretty easily would be interesting (to me).

This board also confirms my hunch too: there are more active female members than males.

In the earlier thread, I estimated the number of active question askers to be 123. Your data includes both askers and answerers, so it is going to be much longer. The 123rd person in the list is Afos22 with 10 answers/comments. So, my estimate does not account for every user, but accounts for 90% of all activity.

I don’t recognise half of those names listed, not a clue who they are.

Love a geeky list, thanks JP

Okay, so here is some analysis of this data:

There were 322 users on this list.

The top 20 users are responsible for 48.8% of all posts. This is a pretty huge percentage.

Using the same estimation procedure I used in the previous thread, we can see how “complete” this list is. My sample size was 140, but this sample size is 9163 (and includes answers). With 9163 posts, 322 users, and 98 people who appeared once:

E = 322 / (1 – 98/9163)
E = 325.677

So, this list is pretty much completely inclusive of all active users, since the measured number of users (322) is about the same as the expected number of users (326).

We could take a larger sample, but doing so would involve sampling older and older threads, at which point the word “active” no longer applies. Eventually, the list would include all 30422 users who have ever answered a thread.

@johnpowell What is this other listing with my name?

1 users/Cruiser/Cruiser/100_0082

I see other Jellies listed again with 2 users and 3 users? Can you explain this?

@Cruiser That was when I logged in and answered for you. Someone has to agree with me every now and again.

@Cruiser There appear to be about 20 or so such errors in the data. I don’t know the exact reason, but I suppose it must be errors in the skimming. I assume that these errors must be fixed by hand.

Poor Mr Some_Ghost…

“Privacy Policy
...
We won’t share your info with any third parties without your explicit permission, unless the law or some imminent harm requires it.”

Just saying.

Thanks, John. It may not be perfect, but it is fun to see.

It confirms who the blabbermouths are. I fall in at the low end of the top twenty.. #18 and less than half the Q + A of the gold, silver and bronze medalists.

^^ #17

Blabbermouths unite, punk!

Nice follow up to my question a few days ago, @johnpowell

@Symbeline Haha, you outdone me by almost a hundred, always said women have bigger gobs :)

Where it says things like users/zenvelo/zenvelo/pepito_t that is just anomalies that the regex couldn’t clean up without doing more work than I wanted to.

@LuckyGuy :: I have a copy of all the questions ever asked, all 170K of them. Since there are timestamps you could graph a users activity over time. But that is way more work than I would want to do. And it would be so very computationally intensive.

I used GREP to basically pull out every instance that a persons avatar shows up. Then a regex to strip it down to just the username. That is why there are some errors. The thing is GREP starts shitting all over itself at about 10,000 files. I could do it in batches but I don’t care enough.

If anyone wants a zipped file of the HTML for all the questions ever asked I could upload it. It is rather large and would take me a few days.

I wonder if some of the people at the very top of the list may back off a little, at least for a while.
“Addicted to Fluther, who…me?”

I refuse. :p

@Symbeline I never said anyone should now did I? :)
Nowt wrong with being a hyperactive member no need for viagra but some might feel a little overexposed.

Yeah, but seriously I drank way too much coffee today. I think I should go for a walk or something haha.

Go barefoot, it’s good for the sole…err…

lol

I don’t think anyone is using the site to much. There is nothing wrong with being at the top of the list. People on the top make solid contributions and are probably responsible for keeping things going.

All I can say is thank god for the three questions a day limit.

Already got that covered.

@johnpowell Sooo…do us top 5 blabbermouths get some sort of major award?
I think you need to cough up some tickets to paradise here.

great effort! thanks.

Fluther T-Shirts all around @Coloma

I still miss SHIPPY! </3

Interesting stuff though jp! I feel like I’m dab stuck right in the middle

@johnpowell So if I understand how this works, say you redid the whole thing today, and still grabbed the last 500 questions, the numbers would change. (not asking you to redo it) Just curious.

Correct. When I grabbed the data I pulled in the last 500 questions. It would probably be very different if I redid it now. I know a good chunk of all mine came from only a few questions that might not be in the current last 500.

Good work, I’m trying to reduce my time to enjoy the summer so I should move down soon. ;)

WHOAH- I actually made it on the list.

You have all 170k questions? A lot of hidden gems in there…. Some seriously funny too.

@Dog See my power already? ~

This is so cool! I’m #49, and at 55 of 500 I still feel like I’m very active. So there’s definitely more than 50 active Jellies.

Aw, I don’t seem to be on there. (Sad trombone.)

Anyway, it’s really cool that you did this! Fluther is like an online version of the neighborhood watering hole. It’s like we’ve just seen everybody’s bar tabs.

No shame in being a big drunk.

And everyone knows who is bellying up to the bar regularly without seeing his bar tab.

Care to update this list? Do will still have a core of twenty?

I’m not even on there with a 1. That was during my 2-year hiatus, and I didn’t post at all. I don’t think I’d make the top 20 now, but I wouldn’t be a zero.

At a glance, though, there’s been a lot of turnover since then. It would be interesting to see how a new quasi-group census works out.