MetaChat - A data point on the number of active users.

MetaChat is an informal place for MeFites to touch base and post, discuss and chatter about topics that may not belong on MetaFilter. Questions? Check the FAQ. Please note: This is important.

27 November 2007

A data point on the number of active users. During weeks 30-39, there were 1657 posts made by 233 users.
→[More:]Here's a breakdown of the number of users who made how many posts:
58 users made 1 post 36 users made 2 posts 20 users made 3 posts 27 users made 4 posts 15 users made 5 posts 14 users made 6 posts 4 users made 7 posts 6 users made 8 posts 8 users made 9 posts 3 users made 10 posts 3 users made 11 posts 2 users made 12 posts 6 users made 13 posts 4 users made 14 posts 3 users made 15 posts 1 users made 16 posts 3 users made 17 posts 1 users made 18 posts 1 users made 19 posts 1 users made 20 posts 1 users made 21 posts 1 users made 22 posts 1 users made 26 posts 1 users made 27 posts 2 users made 28 posts 1 users made 29 posts 1 users made 30 posts 1 users made 31 posts 2 users made 32 posts 1 users made 36 posts 1 users made 43 posts 1 users made 45 posts 1 users made 56 posts 1 users made 72 posts 1 user made 104 posts
Assuming of course, that I haven't made an error somewhere.

I didn't try to get the
number of commenters during those weeks, as I didn't want to hit the servers
too hard.

post by: DarkForest at: 09:13 | 78 comments

Oooh, data, we loves it, we do...

≡ Click to see image ≡

http://www.flickr.com/photos/theophileescargot/2068933742/

posted by TheophileEscargot 27 November | 09:32

This is data for MeCha?

posted by chewatadistance 27 November | 09:34

It's Mechadata.

posted by Daniel Charms 27 November | 09:35

This is data for MeCha?

Yes, derived from the weekly pages you can get in the archives. The xhtml was formatted nucely to easily extract this without any xml parsing.

posted by DarkForest 27 November | 09:38

1 user made 104 posts

Keeeen-o.

I kid. I love you all.

posted by chrismear 27 November | 09:41

104 posts? Hee. I wonder who that could have been. ;P

That's really interesting, DarkForest...I love data and numbers. Just FYI, a super quick read through of registered members tells me that the newest member has a number assignation of 2,255, which equates to that many usernames that have been registered since day 1. That would include the occasional banned member, and from what I can see, a handful of people who registered once and then, some time later, registered again, either because they forgot they registered in the first place, or because they forgot their password, or any other reason. A few registered under one name and then switched to a different name. Many hundreds have registered and then have gone right into lurk mode, never commenting. So, how many unique users there are, I can only speculate, but it's certainly less than 2,000. How many active users there are? Much less than that. A few hundred at most at any given time. Not exactly science on my part, just a semi-edumacated guess.

posted by iconomy 27 November | 09:45

You had to do it DarkForest. You had to blow the lid off of this whole thing... You know how many others had to "disappear" so this wouldn't get out. Fingers, Harold Hammers, My own wife... Have you heard of The Trinity Consortium? Do you know how much of the internet they own?

Do you remember filtermetatalkyweb.com DarkForest? They published some data too. Have you tried googling for them recently? Have you, god damn you, have you tried googling them recently?

*sobs. walks away slowly. single shot rings out*

posted by seanyboy 27 November | 09:49

I imagine there are a lot of commenters around who rarely or never make a post. I wonder if there are many who read regularly but never comment?

posted by DarkForest 27 November | 10:02

I did something similar recently, intending to add some data for the GirlZone discussion, although unfortunately I had to leave it and get on with real work.

In a spurt of intense geekiness, I intended to ask:

if women post more,
if women's posts get more comments,
if posts at certain times get more comments,
if women post more chatty posts (i.e. posts without links),
if chatty posts get more comments, and
how these effects interact.

I wrote a crude GreaseMonkey script that runs on archive pages and outputs information that can be copied into Excel.

Of 917 posts, 467 were by female posters (51%), and 44% of users who posted were female (there were very few users whose gender I couldn't work out; I randomly assigned them M/F). Here's a graph of posts by time (0: midnight MeCha time, 0.5: noon; ignore the stuff on the right).

I have a spreadsheet of this data, and not enough time to do what I intended with it (probably half an hour's work, tops), so if anyone has any use for it, let me know. I was going to do a probit regression to answer the questions I listed above, but there are probably lots of things you can do with it.

posted by matthewr 27 November | 10:03

Do you remember filtermetatalkyweb.com DarkForest?

Okay, Google is scary fast these days.

posted by chrismear 27 November | 10:05

LOL @ iconomy!

DarkForest (I keep trying to call you Deep Forest) it would be interesting to see lurker data, though I have no clue how it would be measured.

posted by chewatadistance 27 November | 10:07

I wonder if there are many who read regularly but never comment?

Yes. There are. They're watching you.

posted by iconomy 27 November | 10:12

interrrresting. The [ghost of] seanyboy (male) and I (female) are now 50% alive and 50% assumed dead, which may skew future results, so try to correct for that in any future or ongoing results, scientists of Metachat.

posted by taz 27 November | 10:16

I'd be curious to hear who the top 5 posters were. I might be up there....

posted by ThePinkSuperhero 27 November | 10:20

Would lurker data be expressed as shadows?

posted by msali 27 November | 10:24

I'd be curious to hear who the top 5 posters were. I might be up there....

You were #4 on the list. No bronze medal for you. Sorry.

posted by DarkForest 27 November | 10:25

I didn't try to get the number of commenters during those weeks, as I didn't want to hit the servers too hard.

If you allow a few seconds between each hit, I don't think requesting 1657 pages would do the server any harm. In general, though, it's a good idea to set the User-Agent of your HTTP request to something informative, like DarkForestBot, so an admin could block your script if it caused problems.

posted by matthewr 27 November | 10:30

re scraping. We're quietest on weekends.

posted by seanyboy 27 November | 10:35

Curses! Must try harder! Must post more!

posted by ThePinkSuperhero 27 November | 10:35

DarkForestBot

That's gonna be my sockpuppet name for sure.

posted by DarkForest 27 November | 10:37

allow a few seconds between each hit, I don't think requesting 1657 pages would do the server any harm
re scraping. We're quietest on weekends.

OK, project for next weekend...

posted by DarkForest 27 November | 10:39

DarkForest (I keep trying to call you Deep Forest)

After I signed up, I discovered there are a lot of other Dark Forests out there. I have nothing to do with any of them.

posted by DarkForest 27 November | 11:22

I think I can guess who the top 2 are, although if TPS is number 4, I don't know who number 3 is.

Must...post...more...bunny photos.

posted by muddgirl 27 November | 12:31

Matthewr: that's interesting stuff. I feel like my basic sense of things is borne out: MeCha is something close to what gender-neutral (on the web anyway) might look like.

I'm not sure about how to value "posts without links" though. I mean, I thought that was what the site was created for - to free discussion from having to center around a link.

I do link to stuff, but sometimes I don't. Then, too, I posted today about the movie Dan in Real Life. I could have linked to the movie, or to the other movie I mentioned, or to someone's blog post about running in movies. But I didn't, because it would have been linking for linking's sake - which is one of my pet peeves. Hypertext is like a pocketknife: just because you have it doesn't mean you should always go looking for ways to use it. A link should be doing something, in my opinion. Linking to simple phrases and commonly known names can actually read as condescending.

posted by Miko 27 November | 12:42

I guess the top 3 are jonmc, essexjan and Specklet. I could totally be wrong about that, though.

posted by box 27 November | 12:46

My guesses would be chuckdarwin, essexjan, and jonmc, although maybe not in that order. We're talking weeks 3-39 of this year, right?

Who are the top five, DarkForest?

posted by iconomy 27 November | 12:58

Yea, chuckdarwin had a few very prolific few weeks; I'd imagine he's somewhere in the top 10.

Come on, DarkForest, out with it!

posted by ThePinkSuperhero 27 November | 13:12

Who are the top five, DarkForest?

I wasn't too sure if I ought to name particular people, or how they would feel about it. TPS asked, so I posted her 'ranking'. What would internet etiquette require? If a bunch of the big time users say it's OK, I'll post the top 20.

But I like seeing your guesses on the top 3. Aggregating the ranks of your guesses:
box's 3 guesses add up to 22
iconomy's 3 guesses add up to 12
(obviously, a perfect score would be 6).

posted by DarkForest 27 November | 13:19

It's gotta be essexjan, jonmc, and miko, in that order.

posted by muddgirl 27 November | 13:21

Top five - chuckdarwin, jonmc, essexjan, miko, by the grace of god. Not in any particular order.

I don't see why anyone would care about having their total outed...we can see everyone's posting and commenting totals from looking at their profile. That's just my 2 cents though. I like prolific posters, personally.

posted by iconomy 27 November | 13:27

Frisbee Girl, whiskeychimp and dno!

posted by box 27 November | 13:29

Oh I forgot that TPS is in the top five...doh. Strike BTGOG.

posted by iconomy 27 November | 13:31

OK I really promise not to post again. After this!

chuckdarwin, essexjan, jonmc, TPS, BP.

posted by iconomy 27 November | 13:34

OK, here's the top 20:
rank, posts, user:
20 18 small_ruminant
19 19 Lipstick Thespian
18 20 taz
17 21 shane
16 22 jason's_planet
15 26 mudpuppie
14 27 matteo
13 28 Miko
12 28 Specklet
11 29 Hugh Janus
10 30 desjardins
9 31 jonmc
8 32 By the Grace of God
7 32 SassHat
6 36 miss lynnster
5 43 arse_hat
4 45 ThePinkSuperhero
3 56 BoringPostcards
2 72 chuckdarwin
1 104 essexjan

posted by DarkForest 27 November | 13:36

There's no way I'm in the top 5!

posted by Miko 27 November | 13:37

I'm going to guess you're somewhere between 12 and 14, Miko.

posted by agropyron 27 November | 14:05

Really interesting! jonmc not in the top 5, BoringPostcards #3!

posted by ThePinkSuperhero 27 November | 14:11

Well I, for one, am shocked. And am also wondering how the heck I ever get anything else done...

posted by BoringPostcards 27 November | 14:32

I'M SLIPPING!!!

I am so dumping the girlfriend.

Also shocked at some of the most prolifics.

posted by mudpuppie 27 November | 14:39

whoa! didn't expect to place.
(Wonders if this means I need a life.)

posted by small_ruminant 27 November | 14:53

BP, flying under the radar!

I'ma ctually shocked, shocked that I'm even in the top 20. Ye gods. I thought I'd been being quiet around here.

posted by Miko 27 November | 14:54

I really need to post more things that no one responds to. I'm very good at that.

I also think I may be a thread killer.

posted by iconomy 27 November | 15:14

Oh god. This thread is so dead now.

*wanders away*

posted by mudpuppie 27 November | 15:26

I started collecting threads and analyzing the comments. I won't have them all in till tomorrow, but with about 10% of the threads in, here are the top 3 commenters:

#comments, user:
87 Miko
88 Hugh Janus
130 ethylene

posted by DarkForest 27 November | 15:27

I'm not even in the top 20? OK, something has to go - should I leave my job or my children? Tough call.

On preview - I see that, when you refer to "posts", you mean "threads". There are "threads" and there are "comments" and I would see the term "posts" as being the combined total. Let's argue about this...

posted by dg 27 November | 15:47

'Posts' to refer to the combined total? What's wrong with 'thromments'?

posted by box 27 November | 15:51

OK, posts = threads
Oh wait... doesn't that header up there read "New Post"?
I don't want to argue. I just finished an argument on askme.
comments = comments until further notice...

dg - only 2 threads, unless I did something wrong...

posted by DarkForest 27 November | 16:14

So far, I don't exist. I am awaiting some evidence to the contrary.

posted by danf 27 November | 16:16

How about 'contributions' for combined posts + threads?

(And it's clearly posts, not threads.)

posted by chrismear 27 November | 16:22

Crap! "posts + threads" should have been "posts + comments".

Damn this confusing terminology.

posted by chrismear 27 November | 16:23

danf: you in a 3-way tie for 21st place with iconomy and mischief with 17 posthreads.

posted by DarkForest 27 November | 16:24

Crap, I clicked post halfway through writing a comment. Could a mod plz hope me and delete the above comment? Ta.

posted by matthewr 27 November | 16:28

Okay, I deleted it... unless someone else deleted the real one, and I deleted another one.

Also, who has the most smileyfaces? :)

posted by taz 27 November | 16:31

Also, who has the most smileyfaces? :)

Ooh, I want to know that too. All I can tell you so far is that in 277 threads there are 41 :), 6 ;) and 10 :(

posted by DarkForest 27 November | 16:39

We're a happy bunch! A happy bunch! :)

posted by taz 27 November | 16:46

I think it's "post." A "thread" happens when someone comments on a post. Until then, there's space for a thread, but no thread. Just a post. A post may or may not start a thread.

posted by Miko 27 November | 16:49

I am wondering if there is a way to determine who has the highest percentage of "last" comments in the threads they have said stuff in.

I am sure that I would get the "tread killer" designation, easily, over iconomy.

posted by danf 27 November | 16:51

We're a happy bunch! A happy bunch! :)

I am looking for happiness.

posted by mudpuppie 27 November | 16:57

I am sure that I would get the "tread killer" designation, easily, over iconomy.

Hey, you're not called Spike Strip Dan for nothing, you know.

posted by mudpuppie 27 November | 16:58

I am wondering if there is a way to determine who has the highest percentage of "last" comments in the threads

That would be easy to find out, but I might hesitate to post the results. I wouldn't want to know if it was me...

posted by DarkForest 27 November | 16:59

Hey, you're not called Spike Strip Dan for nothing, you know.

It is merely that my comments are so well-reasoned, and have such a unique combination of concision and comprehensiveness, that people are loathe to comment after one of mine, lest the comparative callowness of their thinking be put on display.

Or sumthin.

posted by danf 27 November | 17:03

Funny, I always thought I was a thread-killer, and iconomy and danf always seem far more humorous than me...

posted by altolinguistic 27 November | 17:14

Lucky number 7! whooo! I really didn't think I posted all that much, I spend a lot more time *reading* here than anything.

Then again, I guarantee that at least 50% of my posts are nothing more than calls to IRC, in which case, there should be some kind of curve. Don't want to inflate that mess, you know?

posted by SassHat 27 November | 17:15

Or sumthin.

Yeah. That 'sumthin' being that I was making fun of your typo. So not only are you a thread killer, you're a punchline killer too! Bah. Bah, I say!

posted by mudpuppie 27 November | 17:17

Um . . . a Post is not a thread unless there are comments under the post. Right? So, I can Post and/or I can participate in a thread, meaning making one or more comments. 'Post' and Comment' seem to be the, *cough* salient designations.

posted by MonkeyButter 27 November | 17:19

So, I ran my script on the archives again (weeks 30-45) and thought of a couple of potentially interesting stats.

Descriptive Stats
In total: 2750 posts by 262 users. On average: 11.69 comments and 0.70 links per post. 1226 (44.58%) of all posts had no links.

Most Posts
essexjan 174
chuckdarwin 124
BoringPostcards 87
ThePinkSuperhero 84
arse_hat 72

'IMPACT'
To measure this, the average number of comments on this user's posts is compared to the overall comment average. It shows whose posts are disproportionately commented on. To avoid it being biased towards people who have made only one or two posts which happened to receive lots of comments, it's weighted by the user's relative number of posts.
Top 5
ThePinkSuperhero
taz
Specklet
chuckdarwin
jonmc
Bottom 5
SassHat
essexjan
matteo
mischief
BoringPostcards

'CHATTINESS'
This is the proportion of a user's posts that contain no links, again weighted by the relative number of posts.
Top 5
ThePinkSuperhero
Lipstick Thespian
occhiblu
lonefrontranger
hadjiboy
Bottom 5
miss lynnster
chuckdarwin
matteo
BoringPostcards
essexjan

I haven't shown the numbers for the last two measures, because although the weights improve the ordering, they make the numbers themselves meaningless. I hasten to add that user impact doesn't measure popularity: for instance, essexjan presumably ranks low because bunny posts get few comments.

posted by matthewr 27 November | 17:26

LOL, I'm very chatty.

posted by ThePinkSuperhero 27 November | 17:29

LT, chatty? Again, I'm shocked, shocked.

Also noted: these results suggest another line from Casablanca: Round up the usual suspects.

And something about beautiful music.

posted by Miko 27 November | 17:41

OH OH OH

OH

I just saw my typo. Actually I thought that your comment was apt and very funny even without the "thread" "tread" confusion.

Now it's doubly so. And it also reminds me that I engage you in a battle of witticisms only at my extreme peril.

posted by danf 27 November | 17:45

THAT'S RIGHT, SUCKA!

posted by mudpuppie 27 November | 17:46

Ya know, pup, Daughter was just accepted to SFSU, so there could be a chance to go mano a mano in person one day.

If you buy the beer (all my extra $$$ going to the above).

posted by danf 27 November | 17:48

Okay. But I spill things.

posted by mudpuppie 27 November | 17:52

(And congrats to the daughter!)

posted by mudpuppie 27 November | 17:53

This is all so cool! Thanks, Bunnies!

posted by deborah 27 November | 22:58

I'm so average I don't appear on any lists!

posted by dg 28 November | 02:49

Woo-hoo! I'm not in the top 20!

posted by plinth 29 November | 16:35

heh- plinth I was thinking the same thing about the newer numbers.

posted by small_ruminant 29 November | 16:38

2nd Most Chattiest? Rock on. Just think what I coulda done had I actually owned my own computer.

what can I say? I miss you guys....

posted by Lipstick Thespian 01 December | 20:36

← Does anyone remember Galaga? || Broadway Strike Update →

HOME || REGISTER || LOGIN