Introducing Says Who?, the Metafilter identification game July 16, 2010 11:03 AM   Subscribe

Think you know your mefites? Try your hand at Says Who?, the Metafilter identification game.

I spent the last couple days putting this together, based on an old idea someone hacked together years ago (and which I cannot seem to track down).

It's pretty simple: you're presented with a random comment from the blue, the green, or the grey, and a list of five users, and your job is to make your best guess as to which of them is the actual author.

All the listed mefites are, or were, highly prolific on the subsite in question, having made at least on the order of a thousand comments or so (so the names will tend to be familiar). If you have no idea on any given option, that's what the "(I couldn't even guess)" option is for, so don't be afraid to skip the ones where you don't have even a sliver of a reason to guess.

The game keeps track of all the answers, so there's fun Infodump-style analytical possibilities in the long run once a lot of folks have played and put data into the system.

If you play while logged in, it'll also track your results for you over time, so you can see how you've done in terms of correct and incorrect guesses and number of skips. You can also play if you're not logged in, without scorekeeping. If you really want to play and really don't want your guesses associated with your userid, logging out first is the way to go.

pb has done his best to make my code perform really clean and quick, but there's always the possibility that I buggered something up so if you see anything weird going on, please let me know.

Misc. notes:

- You are very unlikely to see the same comment twice, given how large the pool of comments to choose from is.

- I experimented with larger pools of users, but even with the relatively small number included in this version (about 200 or so per subsite, with some overlap between for folks very talkative on multiple subsites) the game is hard. If you're not seeing your username pop up, it's nothing personal; most people aren't in here.

- I may find an elegant way to include people who are newer-but-quite-active in the candidates in the long run, based on some sort of comments-per-unit time, but for now the selection process favors older users since it's just looking at a raw threshold of a large minimum number of comments.
posted by cortex to MetaFilter-Related at 11:03 AM (317 comments total) 34 users marked this as a favorite

This is really, really, really, REALLY awesome. :D


Although I suck at it.
posted by zarq at 11:06 AM on July 16, 2010


Thanks - not like I had any work to do on a Friday afternoon anyway.

(not true but I don't care)

Though I was originally imagining it as this Guess Who? which was, admittedly, a lot more fun.
posted by MCMikeNamara at 11:07 AM on July 16, 2010 [1 favorite]


This is awesome.

Impossible, but still awesome.
posted by Think_Long at 11:08 AM on July 16, 2010


Ok that is awesome. And a little scary that I got the first two out of three right.
posted by restless_nomad at 11:08 AM on July 16, 2010


God I'm such a noob.

You guessed languagehat; it was quonsar.


But not so much that I don't see why that's funny. At least to me.
posted by MCMikeNamara at 11:10 AM on July 16, 2010 [5 favorites]


I like this game!! the only two I've gotten right so far were 2 of my fav mefites: Marisa Stole the Precious and husband-to-us-all Brandon Blatcher
posted by supermedusa at 11:14 AM on July 16, 2010


I'm quitting at 2 for 2 with 2 skipped
posted by shothotbot at 11:15 AM on July 16, 2010


: >

posted by ??????


Oh come on. Maybe you should have some minimum length filter.
posted by desjardins at 11:18 AM on July 16, 2010 [10 favorites]


I gave up with 2 out of 16 when it presented me with a post consisting of the word "Doh!"

Impossible.
posted by crunchland at 11:18 AM on July 16, 2010


I'm doing ever so slightly worse than blind chance.

I think about 50% of the fun is waiting to see if your own name comes up.
posted by le morte de bea arthur at 11:18 AM on July 16, 2010 [2 favorites]


On MetaChat once, a user was accidentally logged in as his SO, but he typed a comment anyway, and everyone instantly knew that it was him and not her. Funny how the way we write defines us.
posted by Melismata at 11:18 AM on July 16, 2010


I am not good at this game.
posted by infinitywaltz at 11:19 AM on July 16, 2010


You're 1 for 6, with 0 skipped.

And that one was when I stopped trying to figure it out and just guessed randomly. This is hard!
posted by nickmark at 11:19 AM on July 16, 2010


Am I not in there because lewistate didn't list me as credible?

It doesn't like nd¢'s username.

Also, once or twice no quote was delivered, total blank, maybe put a minimum characgter lenght on the sample?
posted by cjorgensen at 11:20 AM on July 16, 2010


You're 18 for 71, with 0 skipped.

YOU KNOW I ACTUALLY DO HAVE WORK TO DO TODAY....


*pries self away*
posted by zarq at 11:20 AM on July 16, 2010


Nice game! I'm already thinking up ways to cheat at it using the Infodump.
posted by FishBike at 11:21 AM on July 16, 2010 [4 favorites]


This is pretty awesome, but I suck at it. The only one I got right so far had the user's name included as part of the post.

It would be neat if somehow word could get back to a user if someone correctly guessed one of their posts. That would be a fun way of learning just how predictable are posts are.

Also, maybe an advanced version could use some sort of language analyzer to group together users with similar posting styles. I realize this is way more work than you probably want to put into it, but I'm just throwing out ideas.
posted by bondcliff at 11:21 AM on July 16, 2010


meh
posted by ??????


[] Someone who needs to stop doing that.
posted by Artw at 11:22 AM on July 16, 2010 [14 favorites]


Ooh, ooh, I'm one of the choices!

Wish there was a way to skip all the Ask comments, though, since I never read that subsite.
posted by Eideteker at 11:22 AM on July 16, 2010


Little bug I found: ND¢'s username displayed at just ND in both the list of possible users and in the final answer blurb.
posted by chiababe at 11:22 AM on July 16, 2010


Much like today's media, I am more interested in the thought than the person saying it.
posted by Dagobert at 11:22 AM on July 16, 2010 [2 favorites]


Any time there's a comment featuring fire-shooting dildoes, Pikachu vagina, or people touching dick, I appear to be really good at pointing out the culprit.

I'm a little worried about that, honestly.
posted by fairytale of los angeles at 11:24 AM on July 16, 2010


The only one I have gotten right so far is the one that started out with a quoted section, i.e.:

"zarq, blah blah blah ..."

I correctly identified the MeFite responding to that statement as zarq.
posted by yhbc at 11:25 AM on July 16, 2010


I totally got the konolia one. She was talking about picking up Ralph's socks.
posted by availablelight at 11:26 AM on July 16, 2010 [2 favorites]


Also, maybe a minimum length.
posted by Eideteker at 11:26 AM on July 16, 2010


I went 9 fpor 25. Better than I thought I'd do.
posted by slogger at 11:26 AM on July 16, 2010


Oh come on. Maybe you should have some minimum length filter.

Heh. I may implement one, yeah. Maximum length or excerpting long comments is also a possibility. Both of those are a bit more complicated than the current "hey, database, gimme a comment from user x" method, though, so they're on the TODO list for a possible 2.0 release.

It doesn't like nd¢'s username.

Yeah, there may be some character set dumbness in how I'm configuring output. Not my forte, pb is taking a poke at it.

Also, maybe an advanced version could use some sort of language analyzer to group together users with similar posting styles. I realize this is way more work than you probably want to put into it, but I'm just throwing out ideas.

This is actually the sort of idea I started with (prompted by that dumb I Write Like... web toy the other day, oddly enough) before I realized that I needed to do more groundwork on both writing analysis and vector search space implementations before I was going to make any progress. And then I remembered this idea and realized that that I could actually knock out in a day or two, and here we are.
posted by cortex (staff) at 11:27 AM on July 16, 2010 [1 favorite]


: >

posted by ??????

Oh come on. Maybe you should have some minimum length filter.
posted by desjardins at 1:18 PM on July 16 [1 favorite +] [!]

That's actually one of the easier ones I can think of.
*remembers amberglow*
posted by Ufez Jones at 11:27 AM on July 16, 2010 [11 favorites]


Wish there was a way to skip all the Ask comments, though, since I never read that subsite.

Conversely, I wish there were a way to limit it to only Ask comments (aside from manually skipping every blue/grey comment).
posted by Jaltcoh at 11:27 AM on July 16, 2010


I'm only good at it when the person names himself either directly or indirectly.
Ironmouth, what do you think about blah blah . . .

Well, blah blah is absurd on its blah face . .
Or:
Well, the ArtW way of rolling and smoking old comic books is . . .
I'm well under 50% without those indicators.
posted by Skot at 11:28 AM on July 16, 2010


Cool game, though. I'm extremely bad at it -- worse than randomly guessing (which is odd, since that's pretty much how I play it).
posted by Jaltcoh at 11:28 AM on July 16, 2010


From the site:

This is great!

Oh come on.
posted by Hiker at 11:28 AM on July 16, 2010 [1 favorite]


[few comments removed - if you can't give advice without insulting other commenters, please do not comment. thank you.]
posted by ??????


phrontist
thinkingwoman
LobsterMitten
jessamyn
electroboy



HRM, this is going to be a hard one.
posted by deezil at 11:29 AM on July 16, 2010 [31 favorites]


Holy crap, I am in there! I was reading one of the comments I'd made, thinking, "Wow, this is just like my life, I could have written this, I wonder who's like me?" then I look down, and there I am, as an answer. So one of my right answers is a cheat.

Also, I think you'll do fairly OK if you guess Astro Zombie every time.

It is kinda k=funny to me that the only one I've been certain on so far was a delmoi quote.
posted by cjorgensen at 11:29 AM on July 16, 2010 [1 favorite]


small nitpick: You should consider moving the results from the last answer up above the next question, which, if it's a long quote, makes the results down below the scroll.

And also, trumpets should sound and quarters should come out the bottom of my monitor when I actually get one right.
posted by crunchland at 11:30 AM on July 16, 2010 [2 favorites]


Also, I'm praying that before cortex implements any length filters that someone gets nailed with the Treaty of Westphalia.
posted by Skot at 11:30 AM on July 16, 2010 [1 favorite]


This just goes to show how interchangeable we all are.
posted by Eideteker at 11:30 AM on July 16, 2010


I got an image of a Hillary Duff album.
posted by xorry at 11:30 AM on July 16, 2010


Wish there was a way to skip all the Ask comments, though, since I never read that subsite.

Definitely on the 2.0 TODO list, yeah. More likely in the form of "limit to this subsite" than "exclude this subsite", just for simplicity's sake.
posted by cortex (staff) at 11:31 AM on July 16, 2010


You think so?
You guessed dersins, that was correct!


Um. Wow. That was odd.

3 for 7 so far
posted by Night_owl at 11:31 AM on July 16, 2010


After 50-something clicks, I finally got one I recognized as coming from myself, and HOLY FUCK I can be an asshole sometimes.
posted by Cool Papa Bell at 11:31 AM on July 16, 2010 [5 favorites]


"You guessed Skygazer; it was Navelgazer."

That's MeFi in a nutshell.
posted by Eideteker at 11:36 AM on July 16, 2010 [28 favorites]


Though I was originally imagining it as this Guess Who? which was, admittedly, a lot more fun.

Are you...wearing a hat? No?
Do you have orange hair? Yes?
Are you Frans?
posted by sallybrown at 11:36 AM on July 16, 2010 [4 favorites]


MCMikeNamara: "Though I was originally imagining it as this Guess Who? which was, admittedly, a lot more fun."

Does your person wear a fedora?
posted by griphus at 11:37 AM on July 16, 2010


This game is very hard! The only answer I recognized instantly was an Ethereal Bligh.
posted by ThePinkSuperhero at 11:37 AM on July 16, 2010 [1 favorite]


I suck at this. I did manage to identify the one I wrote, though. And no, it was not about crafting.
posted by orange swan at 11:37 AM on July 16, 2010


Does your person wear a fedora?

No. Does your person have a mustache?
posted by sallybrown at 11:39 AM on July 16, 2010


Look, guys, it's like this: rock music is evil. Demonic. Awful stuff. Please, Jesus-dudes, leave it to the pros.

i <3 BitterOldPunk
posted by nadawi at 11:39 AM on July 16, 2010 [1 favorite]


You're 10 for 30, with 4 skipped.

Man, this is fucking HARD. But kinda fun.
posted by Rhomboid at 11:40 AM on July 16, 2010


To be perfectly honest, this makes me feel like not a member of this community.
posted by rlk at 11:42 AM on July 16, 2010 [4 favorites]


After three questions, I was "2 for 3, with 0 skipped."

I should have stopped there.
posted by DevilsAdvocate at 11:42 AM on July 16, 2010


I wonder if it should also be limited to comments that got at least one favorite. That way, you know you'd be getting comments that actually said something of significance (or were funny). They're more likely to have style.
posted by jacquilynne at 11:42 AM on July 16, 2010 [5 favorites]


i'm really skilled at spotting posts by pyramite termite
posted by Dr-Baa at 11:43 AM on July 16, 2010 [1 favorite]


that's pyramid- man, I can be such a tool sometimes
posted by Dr-Baa at 11:44 AM on July 16, 2010


Are deleted comments included? I got this:

test
posted by ??????


which I guessed right as mathowie, but isn't where it says it is
posted by rollick at 11:44 AM on July 16, 2010


jacquilynne has a good point there.
posted by Artw at 11:44 AM on July 16, 2010


".

posted by ??????"


Oh, fuck you.
posted by Eideteker at 11:45 AM on July 16, 2010 [17 favorites]


I suck at this.
posted by marxchivist at 11:45 AM on July 16, 2010


I just had a post come up that was attributed to aaron, who hasn't posted here in 7 years. Some of the folks here were practically still in diapers back then.
posted by crunchland at 11:45 AM on July 16, 2010


And after an unsuccessful guess...

You guessed grapefruitmoon; it was Marisa Stole the Precious Thing.

Awkwaaaaaaaaard.
posted by GuyZero at 11:47 AM on July 16, 2010 [1 favorite]


No wait, nevermind. Just forget I was ever here.
posted by GuyZero at 11:49 AM on July 16, 2010 [2 favorites]


Sorry if someone already mentioned this, but this really should exclude any comment containing $USERNAME posted by $USERNAME. I got a few right just from people quoting things asked of them.
posted by floam at 11:49 AM on July 16, 2010


Are deleted comments included? I got this:

Ooh, good catch. Yeah, I need to throw a little exclusion into the comment-fetch query I guess.

Some of the folks here were practically still in diapers back then.

Yeah, that's one of the side effects of the simple comment-count criterion: folks who were once quite active but don't come around much or at all anymore are probably overrepresented in terms of recognizability. Another thing for the TODO list.

The nice thing about how this is laid out behind the scenes is that altering the lists of users on the fly is dirt simple, so when I find something that seems like a better method for selection I can swap it in without gumming up the works at all; people will just see (if they specifically notice at all) a somewhat different collection of names popping up.
posted by cortex (staff) at 11:50 AM on July 16, 2010


Sorry if someone already mentioned this, but this really should exclude any comment containing $USERNAME posted by $USERNAME. I got a few right just from people quoting things asked of them.

I'd like to think that this is the game's way of apologizing to you for the times it serves you up a "meh" or a "." or such.
posted by cortex (staff) at 11:51 AM on July 16, 2010 [5 favorites]


It would be interesting to see which users are frequently mistaken for one another.
posted by Solon and Thanks at 11:51 AM on July 16, 2010


You're 33 for 100, with 1 skipped.

FAIL. Better than I would have expected though, given that I just guessed usernames that have been active since I've been reading.

Protip: If the comment has a :) and the end, it's probably zarq. And if it's long and well written, it's probably grumblebee or Miko.
posted by pecknpah at 11:52 AM on July 16, 2010


*at. Grr...
posted by pecknpah at 11:53 AM on July 16, 2010


I'm going to spend $5 on a sockpuppet named "(I couldn't even guess)" and ROOL THE SKOOL!
posted by robocop is bleeding at 11:53 AM on July 16, 2010 [4 favorites]


Any lengthy too-rational-for-this-site comment: Miko
posted by saveyoursanity at 11:53 AM on July 16, 2010


suggestion: could you have the system do a search and not include comments which have the name of one of the posters you are pulling from? This would eliminate the problem that Skot talks about.
posted by Night_owl at 11:54 AM on July 16, 2010


Out of curiosity, what was the basis for choosing the users? I know they had to have at least 1,000-ish comments on the given subsite, but is it just based on who had the most, or did you cherry-pick users who have distinctive styles, or what?
posted by Jaltcoh at 11:55 AM on July 16, 2010


This just goes to show how interchangeable we all are.

I thought it was interesting how, on quite a few questions, I was able to pretty readily narrow the answer down to 2 people. I expected to do a lot more entirely blind guessing. So it seems there are some kind of limits on the pool of interchangeability.

On preview - seconding Solon and Thanks. It would be interesting to see if there are patterns in terms of who users think sound alike.
posted by EvaDestruction at 11:55 AM on July 16, 2010


Very cool, even though I suck at it. Could there be a character minimum on the quotes? I really think that anything under 5 words is unlikely to be guessable.
posted by QIbHom at 11:57 AM on July 16, 2010


I'm Batman!
posted by ??????


sgt.serenity
cgc373
fixedgear
OmieWise
Abiezer
(I couldn't even guess)


I just found out Bruce Wayne's metafilter identity is. I actually remember reading some of these comments, but I still get the answers wrong.
posted by lilkeith07 at 11:57 AM on July 16, 2010


Protip: If the comment has a :) and the end, it's probably zarq.

I overuse the smiley?
posted by zarq at 11:57 AM on July 16, 2010


zarq, I got two of them (yours) correct because of the smilies. Heh. No judgment.(
posted by heyho at 11:59 AM on July 16, 2010


folks who were once quite active but don't come around much or at all anymore are probably overrepresented in terms of recognizability.

Since it knows the login name of the person playing, maybe it could only pick comments posted after the date the player joined Metafilter?

Also, an "Images / No Images" option would be kind of cool. I'd hate for a picture of an elephant shitting an American flag to pop up.

Wait, no, actually that'd be kind of awesome.
posted by bondcliff at 12:00 PM on July 16, 2010


All right, 24 for 100, I have things I need to do today.
posted by Night_owl at 12:01 PM on July 16, 2010


Zarq: Nope, it's just how I recognize when you wrote something. It doesn't get on my nerves or anything.
posted by pecknpah at 12:02 PM on July 16, 2010


I've skipped through 90 just to see if I show up. I think I'm not in there.

I am curious how identifiable MeFites (myself included) are from their comments. I tend to suspect that everyone except me can tell who everyone else is.
posted by Metroid Baby at 12:02 PM on July 16, 2010


Incredibly fun! 9/21. The jokesters were the easiest for me - after a while you can spot the individual humor styles. Secondly, the researchers: which one of these people was likely to look this factoid up?

I'm getting increasingly self-conscious about my verbosity. Why can't I say anything in just a few lines? Safe bet: if it's longer than 2 paragraphs, it might be me.
posted by Miko at 12:02 PM on July 16, 2010


I know they had to have at least 1,000-ish comments on the given subsite, but is it just based on who had the most

At the moment, yeah, that's the sole criterion. The idea of handpicking the list struck me as both onerous and not as much fun in terms of surprises—I don't want this to be a list of People Who Cortex Thinks Of Offhand, I like that there's people in here who I don't necessarily pay personal attention to, etc.

I'm thinking about ways to refine the selection heuristic a bit, as mentioned above esp. regarding high-volume-but-newer users, but I probably won't do any hand-picking unless there are specific exceptional cases to add into the mix, but I'm not sure how i feel about that idea yet in any case.

It would be interesting to see which users are frequently mistaken for one another.

That and a dozen other things, yeah. I think there's a lot of potentially fun data in here.

I thought it was interesting how, on quite a few questions, I was able to pretty readily narrow the answer down to 2 people.

This is almost more interesting to me than the main game play. On my TODO list is the notion of having an occasional "and why?" field pop up, or have it always there as an optional thing, so you can detail your process of elimination a bit. But it seemed a little clunky so I didn't bother yet.
posted by cortex (staff) at 12:03 PM on July 16, 2010


I finally got one right because I remember the thread (and also I'm pretty sure I favorited sidhedevil saying it at the time).

Ideal winnable game: only comments that the player has favorited! not a real pony, just a lament
posted by catlet at 12:04 PM on July 16, 2010 [1 favorite]


You're 33 for 100, with 1 skipped.

I feel like a failed dorky stalker.
posted by elizardbits at 12:04 PM on July 16, 2010 [1 favorite]


Nailed a few. Great stuff.
posted by fixedgear at 12:05 PM on July 16, 2010


You're 61 for 204, with 0 skipped. Yeesh.

It's been weird seeing comments that contain images (and no text!). There was a shot of the MetaTalk logo in blue, on a blue background that I got wrong.

Also, I got one of my own damned comments wrong. :P
posted by zarq at 12:06 PM on July 16, 2010 [1 favorite]


This is super interesting. I often am able to tell some user's comments by their writing style, syntax, lexical choices (and content, views, etc.) within a thread (without seeing the tagline)...but this is really hard without context or date/time stamps. And by hard, I mean, FUN!
posted by iamkimiam at 12:06 PM on July 16, 2010


zarq, I got two of them (yours) correct because of the smilies. Heh. No judgment.(

That's hilarious!
posted by zarq at 12:08 PM on July 16, 2010


Er, 'I often am' → 'I am often' ...damn you, adverbial freedom!
posted by iamkimiam at 12:08 PM on July 16, 2010


Yay, I'm in there! I haven't gotten a comment of mine yet, though.

It's also amusing that both crunchland and Dave Faris are in the database.
posted by yhbc at 12:09 PM on July 16, 2010


The game keeps track of all the answers [...]

Could you add a feature that gives us a way to see our own answer history? A sort of scorecard, so to speak? Because although I, too, suck at this game, I think there are a small number of users I've identified correctly more than once, and would love to review my previous answers to see if that impression is correct.
posted by FishBike at 12:09 PM on July 16, 2010


Haha, I just realized...the big tipoff to my comments is excessive use of 'and/or' and parenthesis (I like them a lot...feels like I'm sharing a secret about what I just wrote). Oh, and pedantry.
posted by iamkimiam at 12:10 PM on July 16, 2010 [1 favorite]


If you can keep your Australians and Canadians straight, it becomes pretty easy.
posted by desjardins at 12:10 PM on July 16, 2010


This is awesome and I'm terrible at it (yo-yoing somewhere between 25% and 33% correct). Since toys seem to be on your mind cortex, how about resurrecting MarkovFilter?
posted by carsonb at 12:12 PM on July 16, 2010


Oh, and pedantry.

I'm sorry, but this isn't particularly selective on MetaFilter.
posted by FishBike at 12:12 PM on July 16, 2010 [2 favorites]


Earlier today, I said that seeing Metatalk thread titles out of context was funny. This is about 23 times more true for some comments.

I've found that though I'm not a good guesser, I'm really good at the process of elimination sometimes. It is also really funny to imagine some of the AskMe questions being answered by some of the incorrect choices. For example, [NAME REDACTED BUT IT'S A GUY I'VE MET] answering a question about a lady's delicate bikini wax reaction will make me smile for a while.
posted by MCMikeNamara at 12:12 PM on July 16, 2010


Could you add a feature that gives us a way to see our own answer history? A sort of scorecard, so to speak? Because although I, too, suck at this game, I think there are a small number of users I've identified correctly more than once, and would love to review my previous answers to see if that impression is correct.

Yeah, I think that'd be fun. I'm trying to think about what a good balance of info would be with analysis stuff—I don't really have a personal beef with a "scoreboard" in a context like this where it's literally a silly game, but it's a touchy enough topic potentially that I'm definitely going to proceed in an experiment-privately-first, make-public-second-if-at-all fashion with the output.
posted by cortex (staff) at 12:14 PM on July 16, 2010


Since toys seem to be on your mind cortex, how about resurrecting MarkovFilter?

Making it work securely unfortunately remains a Hard Problem for esoteric reasons, but I'll try to rope pb into another look at how we might approach it, yeah.
posted by cortex (staff) at 12:16 PM on July 16, 2010 [1 favorite]


Oh good everyone else is terrible at this too. Unbelievably, I guessed quonsar for one and it turned out to be dios. It involved bending over and grabbing your ankles! How could I be wrong? So then I gave up.
posted by furiousthought at 12:17 PM on July 16, 2010


It's pretty scary how much personal information I have brained away about the lot of you. Well, let's see.... the comment's about NYC so that can't be jonson or Justinian or jack_mo... it's not in a moderator tone so it's not jessamyn either... must be jonmc!
posted by carsonb at 12:18 PM on July 16, 2010 [1 favorite]


You're 70 for 227, with 0 skipped.

You guessed caddis; that was correct!

(see original comment in context)
.



HA!
posted by zarq at 12:18 PM on July 16, 2010


Oh, and pedantry.

I'm sorry, but this isn't particularly selective on MetaFilter
"

Case in point. ;)
posted by iamkimiam at 12:21 PM on July 16, 2010


So am I understanding correctly that there will be the possibility of an Infodump on which people are guessed correctly most often?

Because there are one or two are just so obvious to me every time they've come up (this is a huge surprise to me actually) that I can't wait to know if my #1 is similar to everyone else's.
posted by MCMikeNamara at 12:21 PM on July 16, 2010

I read this site all the time, and can frequently identify a poster well before I get to the bottom of the comment. If I think that I can tell who the person is, I click down to the bottom, check who it is, and click back up. It's not hard.

Like, when a comment starts like this:

"This lexical slapstick usually repeats three or four times.
Yes, but where is the sturdy repitition..? Many times soldiers battle language when feild grabs Lex in Con. Why??"

...I generally know who it is, and I just skip right over it. Just use your brain to identify the person who's writing. It's more fun that way, like a game.
Says Who?

Pretty amazing to see that one come up in the game.
posted by kyleg at 12:21 PM on July 16, 2010


Another possible approach to selecting users (though one that would be a little trickier to manage logistically) would be to grab candidates from the whole userbase proportional to their contributions. So the highest volume people would still be most common, but folks who had fewer comments could still show up sometimes.

Which I think would be really nice, if I can find a way to make it work effectively. I'll put it on the experiment list and see if (a) I can make it work slickly and (b) if it does or doesn't make the game harder to play.
posted by cortex (staff) at 12:22 PM on July 16, 2010 [1 favorite]


I'm trying to think about what a good balance of info would be with analysis stuff—I don't really have a personal beef with a "scoreboard" in a context like this where it's literally a silly game, but it's a touchy enough topic potentially that I'm definitely going to proceed in an experiment-privately-first, make-public-second-if-at-all fashion with the output.

Well, to be clear, the scorecard style list of previous answers that I was imagining would be only for the logged-in user's own answers. I wasn't suggesting something where people could see other people's answers.

That said, a leaderboard of people who are doing the best at this game might cool, especially if it was opt-in with something like a "submit high score" button. I'm wondering if that would encourage cheating, though? It does on pretty much every other online game in the history of time.

Stats about the most easily identifiable posters might be kind of neat as well. Although that really depends on the reason why they are easily identifiable. I'm more than a little alarmed by the number of these I've gotten right on the basis of something like "hey look, a smug, dismissive comment, that's probably from user X" and sure enough it is.
posted by FishBike at 12:24 PM on July 16, 2010


I have 4 correct answers out of 33.
there is a price for the lanterne rouge, right?

posted by Dumsnill at 12:25 PM on July 16, 2010


You're 0 for 16, with 1 skipped.

Apparently, I am noticeably worse at this game than a fair die.
posted by Johnny Assay at 12:26 PM on July 16, 2010


33* for 109, and no I don't feel like doing any work today.

*Of which 2 were my own comments; is that cheating, or it is a SUPER FANCY BONUS?
posted by scody at 12:29 PM on July 16, 2010


I think if you've written enough comments to get recognized, recognizing your own is actually still an accomplishment worth recognizing.
posted by MCMikeNamara at 12:31 PM on July 16, 2010


At least I got my own right.

(And man, it's a true statement.)

I'm actually shocked to see how often my name comes up as a possibility. Only one of them has actually been me so far, but I still feel all shiny.

How does the script (I'm guessing it's a script) come up with possible users? Sometimes it's way obvious which ones it's NOT and other times... I feel like it's an evil genius.
posted by grapefruitmoon at 12:32 PM on July 16, 2010


I think if you've written enough comments to get recognized, recognizing your own is actually still an accomplishment worth recognizing.

I'm one of those crazy people who reads the MeTa archives for fun, and there's nothing quite so surreal as reading something engaging and totally right-on, saying 'huh', and then being surprised that it was I who wrote it.
posted by carsonb at 12:34 PM on July 16, 2010 [3 favorites]


I finally got one in which mathowie was a possible answer; the question was about bicycles. I couldn't not choose him, and he was, of course, the correct answer.
posted by yhbc at 12:35 PM on July 16, 2010


Longer comments are easier to identify. Terseness isn't great for analysis.
posted by adipocere at 12:36 PM on July 16, 2010


I got 6 out of 20, skipping, um, quite a lot. Huh. I'm much better at the Guess the Markovfilter game that we used to play on Metachat. That probably says something about the scrambledness of my brain. Still, this is fun!
posted by mygothlaundry at 12:37 PM on July 16, 2010


How does the script (I'm guessing it's a script) come up with possible users? Sometimes it's way obvious which ones it's NOT and other times... I feel like it's an evil genius.

It doesn't do anything clever; it just picks four other random users from the candidate list. The weird juxtapositions, of both the Well Duh and the All Of The Above? varieties, are just happy accidents. Which I think is kind of fun, and is why I'm keeping track of not just the correct answer and the player's guess but the other candidates as well for each round of play.

I had thoughts about, instead of choosing the others randomly, going into the thread that the selected comment is from and selecting other users who also commented there, but I haven't tried it yet. Worth looking at; I don't know if it would make the game easier or harder, but I wouldn't be surprised if it was mostly harder since you'd have a lot more of those "okay, comment about Apple, five guys who constantly talk about Apple...uh..." moments.
posted by cortex (staff) at 12:38 PM on July 16, 2010


So awful at this.
One thing I noticed is that while I have no idea what characterizes certain user's comments, the setup seems to favor guessing based on whichever user of the options available has more total comments. Is that just me seeing things that aren't there, or is that how it actually pans out, cortex?
posted by juv3nal at 12:38 PM on July 16, 2010


Much like the rule of the genus edition of Trivial Pursuit was: When in doubt, say "Rod Steiger"...

The first rule of Says Who? is: When offered pb as a possible answer, jump on it. (Doesn't work every time if cortex is also offered up, but it's a general rule I'm having nothing but success with.)
posted by heyho at 12:42 PM on July 16, 2010


I started out well over .500 with about as many skipped ones as guesses, but the longer I played, the worse I did. Then I discovered how to cheat, and quit.
posted by Kirth Gerson at 12:44 PM on July 16, 2010


Then I discovered how to cheat, and quit.
Wait, you mean something besides googling/searching up the comment in question?
posted by juv3nal at 12:45 PM on July 16, 2010


Also, I encountered zero
.
comments.
posted by Kirth Gerson at 12:46 PM on July 16, 2010


One thing I noticed is that while I have no idea what characterizes certain user's comments, the setup seems to favor guessing based on whichever user of the options available has more total comments. Is that just me seeing things that aren't there, or is that how it actually pans out, cortex?

I'm pretty sure you're just seeing things; the game should be picking among the couple hundred candidates for a given subsite equally, and picking the other four candidates equally as well.

But I'll definitely take a look at the data as it piles up (and, man, people have already played over eleven thousand rounds of this so far in the first hour and a half!) and see if there's any discrepancy there; if there's any sort of unexpected bias come out of the intended equal-chances picks, that'll make it obvious.

Then I discovered how to cheat, and quit.

Needless to say, any trivia game based on publicly-available data is fundamentally broken. Ergo no cash prizes, and the possibility of stern looks from the guy capable of closely analyzing the data should it ever become an issue of e.g. public braggery on the part of the cheater. Heh.
posted by cortex (staff) at 12:46 PM on July 16, 2010


The cheat is way simpler than that.
posted by Kirth Gerson at 12:46 PM on July 16, 2010


People cheat when they play solo games and no one else can see the score? Really?? Nooooo. Not really, right?
posted by heyho at 12:49 PM on July 16, 2010 [1 favorite]


The cheat is way simpler than that.

I imagine I know precisely what you're talking about, and I considered trying to route around it, but see above re: fundamentally broken trivia games, so, eh. Again: no valuable cash prizes.
posted by cortex (staff) at 12:51 PM on July 16, 2010


People cheat when they play solo games and no one else can see the score? Really?? Nooooo. Not really, right?

No, of course no one ever does that, not really.
posted by Kirth Gerson at 12:52 PM on July 16, 2010


Freaks.
posted by heyho at 12:54 PM on July 16, 2010


I'm pretty sure you're just seeing things; the game should be picking among the couple hundred candidates for a given subsite equally, and picking the other four candidates equally as well.

I was about to say that juv3nal is right, but then I read this more closely. Are you saying it picks a random author (from the list of candidates, obviously) first, then goes and picks a random one of their comments? Because that would indeed ensure that the comments presented are equally likely to be from any of the candidates.

Whereas for some reason I was thinking it picked a random comment drawn from the pool of all comments by the list of candidate authors, in which case comments from more prolific posters are more likely to be selected. In which case, guessesing based on who has commented the most would be a good strategy.
posted by FishBike at 12:58 PM on July 16, 2010


Fun and frustrating. By dint of skipping lots of them, I'm now at
You're 11 for 22, with 21 skipped.

Which seems like a good place to stop for now, since my percentage can only go down. (Closest thing to a gimme so far: "You guessed Steven Den Beste; that was correct!")

But seriously, do something about minimum length. It's just dumb having things like "thanks for the post and links" pop up.
posted by languagehat at 1:01 PM on July 16, 2010


Yeah, the back button does make cheating kinda easy. But it also kind of defeats the purpose.
posted by jacquilynne at 1:01 PM on July 16, 2010


(Note: I say "in which case" too much. And I have no idea what "guessesing" is.)
posted by FishBike at 1:02 PM on July 16, 2010


Oooh! I'm two answers!

I'm also spending too much time on this.
posted by grapefruitmoon at 1:02 PM on July 16, 2010


Are you saying it picks a random author (from the list of candidates, obviously) first, then goes and picks a random one of their comments?

Just so, yes.

Whereas for some reason I was thinking it picked a random comment drawn from the pool of all comments by the list of candidate authors, in which case comments from more prolific posters are more likely to be selected. In which case, guessesing based on who has commented the most would be a good strategy.

Yeah, the "just guess the name you know" strategy is the biggest problem with that sort of technique, and part of why I decided not to go with a proportional method originally.

Yeah, the back button does make cheating kinda easy. But it also kind of defeats the purpose.

And it leaves a bright blinding I Am Cheating trail, since the game records every move and two successive entries from the same person on the same comment with the same collection of candidates is astronomically improbable and trivial to search for.
posted by cortex (staff) at 1:05 PM on July 16, 2010


I came to two conclusions while getting 25 out of 85:

1) Who are you people?

2) You dorks are nerds!
posted by ND¢ at 1:08 PM on July 16, 2010


Actually, The Whelk is the result of a failed attempt to clone me, wendell, resulting in the creation of my non-evil twin.

Hey, I managed to get this one.
posted by DevilsAdvocate at 1:09 PM on July 16, 2010


Regarding cheating in online games, a friend of mine has been a games designer for, well, I guess it's decades now. We had a discussion about doing a lot of calculations on the client side vs. the server side, and how most games designers were doing as much as they could server side to prevent cheating with hacked clients and the like.

My friend's opinion was that you can do a lot on the client side and just do some very basic tests for cheating on the server side. The reason being, when people cheat, they are virtually never subtle about it. People don't run hacked clients to gain a 3% advantage when they can get a 100% advantage. Except that the 100% advantage is generally easy to spot.
posted by FishBike at 1:11 PM on July 16, 2010


Well, seeing as how I am a long-haired Metafilter moderator and live in Portland and like to play music and do websites and stuff, I will have to agree with you.
posted by ND¢ at 1:14 PM on July 16, 2010 [8 favorites]


That is going to totally fuck someone up when it comes up in the rotation.
posted by ND¢ at 1:14 PM on July 16, 2010 [5 favorites]


(regarding MarkovFilter):

Making it work securely unfortunately remains a Hard Problem for esoteric reasons, but I'll try to rope pb into another look at how we might approach it, yeah.

I knew it was taken down for security reasons, but what was the actual security problem that it presented?
posted by FishBike at 1:15 PM on July 16, 2010


You could use it to produce replicants.
posted by Artw at 1:16 PM on July 16, 2010


I'm going to stop at 30 out of 100. 33% is probably about the best I've done throughout.

But seriously, do something about minimum length. It's just dumb having things like "thanks for the post and links" pop up.


I agree, though one of the ones I got right was "me, too." Posted by grouse.
posted by oneirodynia at 1:17 PM on July 16, 2010


Holy shit, I think I'm the Comment Whisperer. So far I'm 5 for 5 with none skipper, and ad least two have been epic bullshit (ie, a single link to Torus on wikipedia).

Sure, it COULD be that I've been reading the site since 2000 and see a lot of the regular posters as a kind of extended family that I creepily watch from the bushes. Or, maybe, I'm really lucky. I AM only 5 in.

No, no, I don't think so. I'm going with option C. The comments, they call to me. Whispers in the dark, sliding into my brain through some unknown channel. One night soon, I will sit up suddenly, slightly tilt my head, and go to them. I will be welcome. I will belong. As the gibbering wails of a thousand flameouts wash over me and drag me down, I will finally be able to hear the Final Comment.

Cortex will speak to me, and my sanity will scatter to the farthest reaches of chatfilter.
posted by Stunt at 1:17 PM on July 16, 2010 [1 favorite]


Clearly, I should stop playing this game and start making more comments on the subsites so that I can be a special snowflake and have people try to recognize me. Of course, I'd almost have to double my number of lifetime posts which could completely change the tone and character of my typical posts. Perhaps I can spend a little time writing a Random greekphilosophy Comment Generator based on all my previous comments, and then set it to work posting sufficiently me-esque comments until I reach the appropriate threshold to be included in the game.
posted by greekphilosophy at 1:21 PM on July 16, 2010


16 for 59.

I easily recognized mine, because gah.

It'd be kinda neat to know who the most-recognized mefite is, although I suppose that could potentially result in toxic site-wide popularity contest syndrome.
posted by Sys Rq at 1:23 PM on July 16, 2010


YAAAAAAAAAAAAAAY!

You guessed ThePinkSuperhero; that was correct!
posted by Powerful Religious Baby at 1:23 PM on July 16, 2010 [1 favorite]


I got my own comment when it came up, but I have to admit I wasn't 100% sure it was mine...
posted by DevilsAdvocate at 1:23 PM on July 16, 2010


I knew it was taken down for security reasons, but what was the actual security problem that it presented?

Dark voodoo involving Perl on a 64-bit windows box passing some kinds of data to and from SQL Server, or something. pb could explain; he has, to me, several times, and it just doesn't stick in my brain because it's well above my paygrade.

But seriously, do something about minimum length. It's just dumb having things like "thanks for the post and links" pop up.

I worry a little about this cutting out some of the really wonderfully identifiable-but-pithy stuff, though in practice it may be the case that that's a worthy tradeoff for avoiding a lot of inscrutable comments. It's definitely on the list for possible improvements.

One nice thing is I can actually do some analysis on how comment length correlates to correctness; I wouldn't be surprised at all to see a significant positive correlation there. More interestingly, I could look at how that correlation depends on the commenter (who is still recognizable at sub-twitter lengths?), on the player (who can read into minimalism well?), or the combination of the two (which users would make really excellent partners on a game of The 10,000 Tater Pyramid), etc.
posted by cortex (staff) at 1:24 PM on July 16, 2010


I went through seventeen, and the only two I was able to identify were Steve_At_Linwood and, um, me.
posted by Faint of Butt at 1:24 PM on July 16, 2010


3 answers! One of them even brought back schmoopy memories.

The odds that I will be doing this continuously for the next half hour while Thing 1 watches Cars and Thing 2 naps on the couch are approximately 100%
posted by grapefruitmoon at 1:24 PM on July 16, 2010


This is me.
posted by ?????


I guessed correctly!
posted by Navelgazer at 1:27 PM on July 16, 2010


One guessing technique on the shorter ones that seems to paying off, though who knows if it's really statistically meaningful: lack of capitalization in a short comment appears to correlate with lack of capitalization in username.
posted by yarrow at 1:49 PM on July 16, 2010


Does getting one of your comments wrong mean you spend too much time here? Yes. It does. Not that I'm likely to cut back. Damn it.
posted by y6y6y6 at 1:49 PM on July 16, 2010


This is GREAT and I am TERRIBLE at it.
posted by Mikey-San at 1:55 PM on July 16, 2010


One guessing technique on the shorter ones that seems to paying off, though who knows if it's really statistically meaningful: lack of capitalization in a short comment appears to correlate with lack of capitalization in username.

I noticed this, as well, at least in one direction. Comments without proper capitalization don't ever seem to belong to someone with a capitalized username. Comments with proper capitalization, though, don't necessarily correspond to capitalized usernames, because non-caps usernames are just incredibly common and not necessarily meaningful.
posted by jacquilynne at 1:57 PM on July 16, 2010


Cortex: Are there people who haven't been cheating and are really, really good at this game? Has anyone done over 100 and answered better than 50% correct?
posted by pecknpah at 1:58 PM on July 16, 2010


I am not cool enough to be in the database, am I.
posted by shakespeherian at 2:00 PM on July 16, 2010



You and me both, shakes!
posted by Mister_A at 2:01 PM on July 16, 2010


Protip: If the comment has a :) and the end, it's probably zarq. (pecknpah)

Except when it's MiguelCardoso.
posted by ocherdraco at 2:03 PM on July 16, 2010


Cortex: Are there people who haven't been cheating and are really, really good at this game? Has anyone done over 100 and answered better than 50% correct?

Haven't even begun to do analytics; I may start taking a peek this afternoon but likely it'll wait till next week after I can recover form Epic Birthday Meetup Weekend.

I am not cool enough to be in the database, am I.

Nonsense, you made the cut for both the mefi and meta databases. It's just that so did a couple hundred other people.

You and me both, shakes!

See above. Neither of you have answered enough questions on the green to make that cut, though, but then neither have I (though I'm close).
posted by cortex (staff) at 2:04 PM on July 16, 2010


Well, blah blah is absurd on its blah face . .

blah is usually absurd on its blah face, yes.
posted by Ironmouth at 2:05 PM on July 16, 2010


cortex you are a liar and you did not invite me to your birthday party.
posted by shakespeherian at 2:06 PM on July 16, 2010


one of the ones I got right was "me, too." Posted by grouse.

Ouch.
posted by grouse at 2:08 PM on July 16, 2010


Shit sandwich.

posted by ?????

You guessed scarabic; that was correct!
posted by ODiV at 2:09 PM on July 16, 2010


Actually, The Whelk is the result of a failed attempt to clone me, wendell, resulting in the creation of my non-evil twin.

Hey, I managed to get this one.


Thi is a strange feeling. I don't know how to describe it. Internets Vu?

Is it wrong that I'm playing this game mostly to see if I come up once?

It feels wrong.

I just want to see how special my snowflakes are .. .. ah who am I kidding, just pick the quote that kinda looks like English but doesn't scan at all ..and is about robot vampies in spaaaaace
posted by The Whelk at 2:14 PM on July 16, 2010


shakespheherian, I just got one of yours wrong, so...
posted by carsonb at 2:14 PM on July 16, 2010


I ended up with 8 out of 12 correct and stopped in a panic as I realized that I've been spending farrrrrrr too much time on Metafilter.
posted by jeanmari at 2:15 PM on July 16, 2010


Metroid Baby: "I tend to suspect that everyone except me can tell who everyone else is."

No way; this site is a blizzard of special snowflakes, each with their own styles. I've lurked on here for at least 5 years before I joined and still got only 8 for 20 with 1 skipped. And one of those had the name in it. You're not alone.
posted by Hardcore Poser at 2:15 PM on July 16, 2010


You guessed Ethereal Bligh; that was correct!

That was a toughie.
posted by shakespeherian at 2:17 PM on July 16, 2010


You and me both, shakes!

Not even. "Mister_A" was the answer to the only comment with less than 7 words that I absolutely certain I knew.
posted by EvaDestruction at 2:21 PM on July 16, 2010 [1 favorite]


It's alarming how much casual information I've picked up about you people.
posted by The Whelk at 2:23 PM on July 16, 2010


Nice to know the rest of the country has caught up to me using IRC to meet under genitally-minded men when I was 16.
posted by ??????

smackfu
wendell
Flunkie
jb
The Whelk
(I couldn't even guess)


It's like a mirror that shows you as you truly are.
posted by The Whelk at 2:24 PM on July 16, 2010 [1 favorite]


MetaFiller
posted by ??????

not_on_display
Ironmouth
dios
caddis
Justinian
(I couldn't even guess)


now its a zen koan.
posted by The Whelk at 2:31 PM on July 16, 2010


I gave up after 0 for 2. Too much text.
posted by yeti at 2:32 PM on July 16, 2010


It's really easy to spot a Miko. I have twice confused Astro Zombie and Miguel Cardoso...
posted by Mister_A at 2:34 PM on July 16, 2010


> To be perfectly honest, this makes me feel like not a member of this community.

It would be nice to expand the pool somehow, so it could do something like was mentioned above where it confines the answers to only people who have joined after the person taking the quiz (or even cooler...people only from threads you've participated in).

This criticism is valid, but the counter argument is that if participation is minimal it will be reflected. If you're not on enough to get a high enough comment count to be counted then chances are no one would guess you even if you were included.

Also, I don't know a lot of those people at all!
posted by cjorgensen at 2:35 PM on July 16, 2010


My trouble is that, while having been around for a while I often recognize the names, I don't always associate individual users with a particular posting style. There are a handful of exceptions to that, but especially for folks who aren't around much anymore, it mostly tends to be a matter of "oh, I remember that username! I wonder if they could have written that? Well, I don't recognize these other folks so well, so I'll guess that one."

And then my wrong answer count goes up.
posted by nickmark at 2:36 PM on July 16, 2010




Cheat sheet!
posted by Rhaomi at 2:43 PM on July 16, 2010


I got 33 out of 100, and I was feeling pretty good about it, then I skipped through another 100 just to see if my own name would come up as one of the choices. It never did, even though I meet the criteria for inclusion. I saw other names, who are less active than I am, show up multiple times. I guess I'm not one of the cool kids.
posted by amyms at 2:58 PM on July 16, 2010


You're in the askme list, amyms, but the random number generator is a harsh mistress and so there's about even odds that you'd see your name or not in about 200 trials.
posted by cortex (staff) at 3:03 PM on July 16, 2010


amyms, you are definitely in there. You showed up as an option about six times before I gave up. I have poor turgid dahlia spotting skills.
posted by catlet at 3:07 PM on July 16, 2010


Ahh thanks, cortex. I'm in a "woe is me" mode lately and I'm latching onto to weird things to add fuel to my pity party. Sorry for inflicting my whinyness on you.
posted by amyms at 3:11 PM on July 16, 2010


My god, it's been five years since I did this last.
posted by Wolfdog at 3:19 PM on July 16, 2010


[few comments removed - if you can't give advice without insulting other commenters, please do not comment. thank you.]
posted by Wolfdog at 3:28 PM on July 16, 2010 [2 favorites]


There, that should really annoy somebody when it eventually turns up.
posted by Wolfdog at 3:29 PM on July 16, 2010 [4 favorites]


amyms, I saw you too. I guessed you once, but I was wrong.
posted by Miko at 3:38 PM on July 16, 2010 [1 favorite]


cortex: "based on an old idea someone hacked together years ago (and which I cannot seem to track down)."

Perhaps you mean my silly toy Who Wrote This MetaFilter Comment?, which somehow still works and was based on this thread.
posted by Plutor at 3:39 PM on July 16, 2010 [2 favorites]


Lots of fun! How many questions are there? I got to 150 and stopped.
posted by garnetgirl at 3:43 PM on July 16, 2010


Oh, if you quit, it will remember your game! I was worried I'd have to start over. I'm up to 25 of 64.

A little disappointed that I haven't come across a highly recognizable ericb so far - long quote from a news source.
posted by Miko at 3:46 PM on July 16, 2010


How many questions are there? I got to 150 and stopped.

Somewhere on the order of a million or so. It's not pre-selected, it's randomly selected comments from the several hundred total high-volume candidates, so there's just oodles of replay value.

And, you know, more comments every day. Crowdsourced DLC!
posted by cortex (staff) at 3:49 PM on July 16, 2010


Perhaps you mean my silly toy Who Wrote This MetaFilter Comment?, which somehow still works and was based on this thread.

Yes! Ha, for some reason I assumed it was in the heady pre-Projects days.
posted by cortex (staff) at 3:50 PM on July 16, 2010


yeti: "I gave up after 0 for 2. Too much text"

B-but this is Metafilter - a text site - but this could be sarcasm - or a meta-Meta comment - must suss yeti's intentions - but NOT ENOUGH TEXT TO FIGURE IT OUT - *head asplodes*
posted by Hardcore Poser at 3:56 PM on July 16, 2010


MetaFilter: toxic site-wide popularity contest syndrome.
posted by fleacircus at 4:44 PM on July 16, 2010


I'm super good at guessing mefites I've met in person!

Everyone else blends together.
posted by internet fraud detective squad, station number 9 at 5:00 PM on July 16, 2010


After playing with that for like, an hour, I got much better at seeing patterns within the high-volume posters and my guessing skillz were more accurate.

That means it was an hour well spent, right? RIGHT?!
posted by grapefruitmoon at 5:03 PM on July 16, 2010


Wait how come this game remembers my score when I opened it on a different machine?!?!?!
posted by shakespeherian at 5:09 PM on July 16, 2010


DARK MAGIC
posted by The Whelk at 5:09 PM on July 16, 2010


and I think it's tied into your mefi cookie log-in stuff
posted by The Whelk at 5:09 PM on July 16, 2010


MAGIC COOKIES OM NOM NOM
posted by shakespeherian at 5:13 PM on July 16, 2010 [1 favorite]


You might need to widen your browser, schyler523; I had to, to get the columns to show up side by side. Otherwise, the YT comments chill out underneath the MF comments. Rogue float:left, I'm guessing.
posted by ??????

puke & cry
cortex
The Whelk
carsonb
The God Complex
(I couldn't even guess)


This one is almost too easy.
posted by The Whelk at 5:17 PM on July 16, 2010


Wait how come this game remembers my score when I opened it on a different machine?!?!?!

It's tracking per user, based on your login cookie. So you can play from multiple machines and it'll track correctly, and spouses/roomies/etc can keep their scores separate by being diligent about logging out and in per usual.
posted by cortex (staff) at 5:30 PM on July 16, 2010


So, I run that old grumblebee script that displays users' avatars next to their comments, which is helpful in a way in remembering separate users' personalities (placing a "face" to the personality, if you will.) An unintended consequence of this, however, is that Miko and grapefruitmoon have run together in my head quite a bit, I now realize.
posted by Navelgazer at 5:37 PM on July 16, 2010


12 for 36 before I gave up.

One was easy, though - it was Rumple calling me out for criticism or something.
posted by UbuRoivas at 5:58 PM on July 16, 2010


Well, I answered 100 and got 31 right, so I guess that's better than chance. I hate to say this, but I got several of them right by making some gender assumptions (women were less likely to provide a lengthy answer about HDTVs, men were less likely to go into detail about how much they loved their puppy).
posted by desjardins at 6:01 PM on July 16, 2010


desjardins: I've been doing the same thing, TBH.
posted by Navelgazer at 6:04 PM on July 16, 2010


Am I the only one who can identify a fair number of people by their punctuation abilities/quirks? (e.g., matteo almost never puts a period at the end of the final sentence of his comments, but will punctuate normally otherwise.)

*crickets*

All righty then. I'll just be over here hanging out with my Chicago Manual of Style and a glass of wine. Friday night, and the editor is letting her hair down!
posted by scody at 6:14 PM on July 16, 2010 [3 favorites]


scody, I'd like to hope that I'm recognizable by my constant typos and screwing up of it's/its, you're/your and their/they're, at least.
posted by Navelgazer at 6:38 PM on July 16, 2010


this. Is. AWESOME.

Huh.

But the better I do at it, the worse I feel about how I've spent my life. Hm.
posted by EatTheWeak at 7:15 PM on July 16, 2010


I'm a little concerned by how many of these I got right.
posted by thivaia at 7:23 PM on July 16, 2010


Previously.
posted by Jaltcoh at 7:34 PM on July 16, 2010


Just got this one:

"I'm hoping someone will pay for a large list of other users, with links to specific defining quotes those users have made."

Which made me laugh a little bit.
posted by Navelgazer at 7:44 PM on July 16, 2010


You're 26 for 100, with 1 skipped.

Apparently, I am good at picking out BitterOldPunk and the mods.
posted by Abbril at 7:55 PM on July 16, 2010


I'm both impressed by the cleverness of the game, and amazed at how excluded I feel it is, as one of the thought I was involved but realized that I am one of the many thousands of under 1,000 comment-having, Non-200.

I'll go ahead and just own that.

And say that I think it is pretty neatly done.
posted by anitanita at 8:01 PM on July 16, 2010


Also, interesting data point: I joined the site a little over four years ago, so there are a number of users on this game who were entirely or almost entirely before my time. The only e that I get with any consistency (but who I get almost every time) is Steven C. den Beste.
posted by Navelgazer at 8:03 PM on July 16, 2010


den Beste is easy. He's like a mister_crash_davis (original version) or dios. On any of those three, I don't need to even look down to see if they're in the list of choices; they are.
posted by yhbc at 8:10 PM on July 16, 2010


Am I the only one who can identify a fair number of people by their punctuation abilities/quirks?

I got a couple sgt. serenity ones, an andrew cooke, and a pyramid termite this way.
posted by LionIndex at 8:15 PM on July 16, 2010


hye now i don';t ave any particular puncuation to notivce k?
posted by The Whelk at 8:18 PM on July 16, 2010 [1 favorite]


Finally, after 199 guesses, I got one that I actually wrote. Luckily for me, when I'm not actually making a new comment, I spend all my time reading and re-reading all of the posts I've ever written here, so I was able to recognize it.

Ah well. Back to the archives.
posted by crunchland at 8:32 PM on July 16, 2010


crunchland: I'm in the five-hundreds and haven't gotten one of mine yet, I've seen my name come up in the choices a number of times. I have gotten quite a few of yours, if that helps.
posted by Navelgazer at 8:44 PM on July 16, 2010


With everyone practicing like this, it'll make the Sniffing-Out-Sock-Puppets midterm exam just that much more competitive. Heads up, people!

I can't help feeling like we're being groomed for some nefarious purpose...
posted by heyho at 8:44 PM on July 16, 2010


And with apologies to Mister_A and GuyZero, your names are similar enough and your styles/habits not so dissimilar that I'm getting y'all mixed up constantly here.
posted by Navelgazer at 8:58 PM on July 16, 2010


All I got was a broken image link and I was able to correctly guess quonsar!
posted by Abiezer at 9:00 PM on July 16, 2010 [1 favorite]


cortex: can we reset our stats? Start over, as it were?
posted by yhbc at 9:07 PM on July 16, 2010


28 out of 66 with 34 skipped. I claim 42%, since a lot of those skipped comments were just too short. Still, I thought I knew y'all better than that.

Sometimes the quotes were a little surreal. Cortex, are you sure you didn't slip a few markovfilter comments in there? Either context means everything, or there's a lot of posting-while-baked around here.
posted by Quietgal at 9:12 PM on July 16, 2010


It'll be embarrassing when I can't recognise my own comments.
posted by rodgerd at 10:00 PM on July 16, 2010


Aha! At #736, I finally got one of my own:

"."
posted by Navelgazer at 10:02 PM on July 16, 2010


yhbc said: can we reset our stats? Start over, as it were?

I'd like this option too, since I wasted 100 of my guesses by skipping over them in my searching-for-my-own-name/boo-hoo-woe-is-me whinefest earlier [insert embarrassed face here].
posted by amyms at 11:36 PM on July 16, 2010 [1 favorite]


35 for 102 with 13 skippped.

POSTED BY KLANGKLANGSTON

Future generations will thank me.
posted by klangklangston at 12:12 AM on July 17, 2010


Could we get a hint, maybe the date and time and favorites?

STILL KLANG HERE.
posted by klangklangston at 12:14 AM on July 17, 2010


Really brilliant and entrancing. What a great idea.

Thought I would be able to do a lot better. Increasingly ominous the way my name keeps showing up in the choices. (23 of 52 w/no skips, so far.)
posted by jamjam at 12:40 AM on July 17, 2010


I just got:

.

Perhaps it should filter out things that don't contain any words.
posted by Mikey-San at 3:14 AM on July 17, 2010


I AM KLANGKLANGSTON!
posted by Kirth Gerson at 4:15 AM on July 17, 2010 [2 favorites]


You're 16 for 66, with 76 skipped.

Should I feel ashamed of this score? Because it looks totally underwhelming, and possibly below chance, and yet I still feel like I made a couple of good recognitions. This game is hard enough that this shitty achievement might still be above average. Very few people reveal a distinctive personality in their comments, most comments were short, generic, and fragmentary, and there are only a handful of people on MeFi that are truly popular. I didn't start skipping answers for a while, and I'd mostly just pick the one name I sort of recognized, biasing my answers towards the super-prolific commentors.
posted by dgaicun at 6:32 AM on July 17, 2010


I find the short ones are a lot easier--they're usually not-well-thought-out comments so they have more personality than the shit that people actually think about and make into paragraphs
posted by internet fraud detective squad, station number 9 at 7:05 AM on July 17, 2010


jackmo always comes up when you guess him - super.

havent seen migs or quonsar in there yet, maybe they are too distinctive ?
posted by sgt.serenity at 7:06 AM on July 17, 2010


I got one by quonsar.
posted by the littlest brussels sprout at 7:19 AM on July 17, 2010


There's not really any super clean way to offer per-user score resets at the moment, though I'll think about it. I don't want to just blow random data out of the game's memory, since collecting data is part of the interesting thing as far as I'm concerned.

What I can totally see doing at some point is doing a like Grand Reset, so I can swap out all the data up to that point and give everybody a fresh start. Maybe do that quarterly, sync it up with any major changes to who the game works so that the data will be relatively well partitioned by game setup too.
posted by cortex (staff) at 7:25 AM on July 17, 2010


orange swan: I suck at this. I did manage to identify the one I wrote, though. And no, it was not about crafting

I only identified 2 out of 20 and one of them was yours. It wasn't about crafting either.

Cortex, this is a very fun toy; thanks to you and PB!

But I think you should release the list of the 200 members per subsite. This will save me a huge amount of time, since I can stop playing this game just to see if it includes me and start paying attention to the actual answers.

Why yes, yes I do want a pony with my name on it.

Actually, I demand that the next pony you grant be named DarlingBri, regardless of what or whom it's for.
posted by DarlingBri at 8:41 AM on July 17, 2010


I could probably put together a list, yeah, but for the moment let's just do it this way:

For each of the three subsites, you are on the current list if you have (or, well, had at some point yesterday morning) at least the following number of comments on that subsite:

Mefi: 2000
Askme: 1500
Meta: 1000

So if you want to know but don't want to just blow through a bunch of rounds hoping the RNG will be kind and prove it to you, check your profile page against each of those numbers.

Note that eventually this will probably change if I can find a workable way to get more usernames in the mix without mucking up the game or making it significantly harder in the process. But for the near future, that's the deal.

I'll rerun the list generation routine occasionally too, maybe once a week or something. Not a whole lot of people are going to cross those thresholds in a given week, but if you're at 997 metatalk comments right now and you're feeling left out three comments later, I don't mind a gentle poke to rerun it sooner.
posted by cortex (staff) at 8:52 AM on July 17, 2010


cortexPosterAdmin: Askme: 1500

YAY! I finally got included in a MetaFilter thing. I knew brute force would eventually win! It's OK if it changes; for one brief and shining moment, I was there.

I'm going to die happy now. Although probably not today.
posted by DarlingBri at 8:57 AM on July 17, 2010


Darling Bri: I got you a few times. You're definitely on it.
posted by Navelgazer at 9:02 AM on July 17, 2010


Mefi: 2000
Askme: 1500
Meta: 1000

So if you want to know but don't want to just blow through a bunch of rounds hoping the RNG will be kind and prove it to you, check your profile page against each of those numbers.


So basically what you're saying is that I need to spam the holy hell out of these three sites? Right?

I will need some beers first, I think.
posted by elizardbits at 9:13 AM on July 17, 2010 [1 favorite]


OPERATION PANPOSTIA
posted by The Whelk at 9:15 AM on July 17, 2010


25 out of 53 with 201 skips. This has totally confirmed that I don't really pay attention to who posted stuff. Also I managed to get the same comment twice and got my answer wrong the second time

carsonb writes "I'm one of those crazy people who reads the MeTa archives for fun, and there's nothing quite so surreal as reading something engaging and totally right-on, saying 'huh', and then being surprised that it was I who wrote it."

I love that when it happens to me.
posted by Mitheral at 9:17 AM on July 17, 2010


I don't know how much development you want to do, but for me the game would have much more staying power if I also got points for eliminating possibilities. It's frustrating to narrow it down to two users and still have it be all or nothing.

In addition the resulting data could be used to determine people's "opposite", that is the poster who could never ever be confused with them. I've got my money on a Miko/quansar matchup, but would be cool to see others.
posted by Meta Filter at 10:03 AM on July 17, 2010


I don't know how much development you want to do, but for me the game would have much more staying power if I also got points for eliminating possibilities. It's frustrating to narrow it down to two users and still have it be all or nothing.

I hear you, and the idea of doing a like "ranked by likelihood" version is one thing I've been chewing on; the biggest problem there is that it'd make gameplay a bit slower and clunkier since you'd be spending more time fiddling with each round. But it could be a fun variant to try at some point.

In the mean time, take solace in the fact that if you are doing an effective job of eliminating possibilities, it is helping your score in the long run. Those standardized-test-taking skills aren't going to waste.

In addition the resulting data could be used to determine people's "opposite", that is the poster who could never ever be confused with them. I've got my money on a Miko/quansar matchup, but would be cool to see others.

Yeah, there are a bunch of things like this I want to play with, and I'll start dabbling with it sooner rather than later. The amount of data required to get meaningful results out of a matrix hundreds of users on a side is big, though; assume for the sake of clean numbers (and this is significantly lowballing it) that there were a hundred usernames in the game; that'd present a 100*100 matrix of e.g. actual/guessed user rates, and so we'd need to run 10,000 trials just to have an average of one trial for each possible ordered pair. We're actually looking at some thing more on the order of three distinct 200*200 matrices and we're only crawling up toward 40K rounds played so far, so it'll be a while before there's real data to chew on on any of those more rarified pair-wise ideas.

Per-user basic score stuff (both for players in a "how am I doing in detail?" context and for guessable users in a "how identifiable is user x?" sense) will be easier to put together on the short term, though.
posted by cortex (staff) at 10:31 AM on July 17, 2010


Also, from a glance through a very, very rudimentary dump of score data so far, it looks like a number of people have gone on epic skipping runs, probably of the "man, where am I / where is user x" sort, something I didn't anticipate but probably should have. I'll have to think about how to look for and maybe ignore some of that data in any analysis I do of actual gameplay, though that's not a huge problem given that the non-skip info is probably the most useful in any case.
posted by cortex (staff) at 10:42 AM on July 17, 2010


POSTED BY KLANGKLANGSTON

Future generations will thank me.


Believe me, klang, you are hard to miss even without the name tag. I think I got three of your comments, and nailed each of them.
posted by scody at 11:34 AM on July 17, 2010


You're 33 for 100, with 0 skipped.

Twice as good as random. Take that, random!
posted by team lowkey at 11:34 AM on July 17, 2010


I said to myself this afternoon, KLANGKLANGSTON!, dude! why are you spending so much time on this game when you have other things you need to be doing?!
posted by Secret Life of Gravy at 12:41 PM on July 17, 2010 [1 favorite]


My epic skipping run was "anything with a green background", although I later changed my strategy to "if the quote is from AskMe AND iconomy is an option, select her; otherwise, skip it". This has improved my score considerably.
posted by yhbc at 12:45 PM on July 17, 2010


Right now I'm 98 for 300 - nearly a 1/3 success rate - but with 198 skipped.
posted by yhbc at 1:01 PM on July 17, 2010


iconomy is female? I did not know that. No wonder I'm doing so terribly. Currently 35 for 122, with 18 skipped.
posted by desjardins at 1:05 PM on July 17, 2010


This is oddly addictive. Only including comments that have favourites might make an interesting change, my apologies is someone already suggested this earlier in the thread.
posted by rjs at 1:24 PM on July 17, 2010


iconomy is female?

Boy if I had a million bucks for every time that has been posted...I'd have enough to buy a house now.
posted by Secret Life of Gravy at 1:37 PM on July 17, 2010


desjardins: it helps to remember that ever third answer is "iconomy." For some weird reason, the game randomly chooses her comments above all others.
posted by Navelgazer at 1:37 PM on July 17, 2010


So far, I've stopped at 9/25 with 4 skips.
posted by The Great Big Mulp at 1:45 PM on July 17, 2010


It's weird how usernames appear gendered when they're not; a long time ago I used to use the name Scorpio (with a suffix) on other sites, and I was invariably perceived as male. I've never had anyone perceive "desjardins" as male even though it's not gendered. For some reason "iconomy" reads as a male username to me.
posted by desjardins at 2:13 PM on July 17, 2010


Gaming the system.
You guessed ??????; it was ??????.
*back button*
You guessed ??????; that was correct!
You're 100 for 100, with 0 skipped.
posted by unliteral at 2:17 PM on July 17, 2010


Protip: fshgrl is always the right answer on the green.

I'm not going to embarrass myself with my score thus far other than I HAVE BEEN DOING THIS A LOT. And I'm embarrassingly easy to spot for doing SHIT LIKE THIS with the all caps.
posted by grapefruitmoon at 2:18 PM on July 17, 2010


Also: There have been a lot, and I mean A LOT, of comments from the original longboat coming up for me. Ah, memories.

Also also: I'm surprised Todd Lokken hasn't come up in any context yet.

- Todd Lokken
posted by grapefruitmoon at 2:20 PM on July 17, 2010


Can you just hook up a motion detecting light? These things are very common in the suburbs -- if he's trying to take a dump a bright light will shine down and illuminate his actions?

Kinda wish I checked the context for this one before I clicked away. But probably that would ruin it.
posted by jeoc at 3:14 PM on July 17, 2010


it looks like a number of people have gone on epic skipping runs

For what it's worth, refreshing the page counts as a skip.
posted by crunchland at 3:33 PM on July 17, 2010


Yeah, if you do a refresh it'd likely be resending the form data. I may build a small cleanup tool at some point to look for semi-duplicate rows that'd result from that.
posted by cortex (staff) at 3:45 PM on July 17, 2010


Back for my third visit it says I'm "75 for 201, with 8 skipped" which looks pretty good compared to reports above (or pretty bad if it says I spend too much time here). turgid dahlia will be pleased that I spotted one of his because he used the word 'poo'.
posted by Abiezer at 3:51 PM on July 17, 2010


Heh - I just got an "All right, I'm closing this thread" comment. None of the mods were listed as answers. It was Astro Zombie, not mischief.


It should have been mischief.
posted by Kirth Gerson at 4:04 PM on July 17, 2010


I've never had anyone perceive "desjardins" as male even though it's not gendered.

Because of a race-bike-parts vendor by that name, I read desjardins as male.
posted by Kirth Gerson at 4:08 PM on July 17, 2010


Hm, I knew about the bank, but not the bike vendor. There's also a (male) pro hockey player named Desjardins.
posted by desjardins at 4:38 PM on July 17, 2010


Another quickie analysis note: the two most consistently recognized people so far are pb (with a stunning 67%) and jessamyn (with 45%). There next three at 40% are Pope Guilty, Astro Zombie, and The Whelk. From there there's a few dozen folks in the 30s (I'm at 35%), and then a solid crop of 20s, and on down through the tens and oughts as well.

I don't know exactly what to make of this distribution yet.

This is based on number of correct identifications of a comment vs. total number of rounds in which that person's comment was the answer, so either a wrong guess (player guessed not-them) or a skip (player declined to guess) will count against the total. Any given person on this will have shown up in somewhere between 50 and 200 rounds of play so far; one interesting note is that the bottom end of the recognizability list seems to have folks who showed up in fewer rounds on average, who are hence likely folks only in one of the lists rather than two or three. Multi-subsite visibility presents itself as a strong indicator of recognizability, go figure.

The flip side that i haven't touched yet is how often any given user is incorrectly identified as the author of another comment. I'm guessing that will have it's own set of odd effects where being generally known on the site will correlate to a higher level of false positives.
posted by cortex (staff) at 4:39 PM on July 17, 2010


pb is easy, he posts almost exclusively on one topic in one subsite. jessamyn has a certain style and tone, and uses "yeah" a lot. Astro Zombie - some indefinable tone and style. I think The Whelk's brain gets ahead of his fingers because he seems to make certain typos, plus he's generally not too serious and is kind of quirky. I couldn't pick out a Pope Guilty comment (sorry dude).
posted by desjardins at 5:11 PM on July 17, 2010


Mefi: 2000
Askme: 1500
Meta: 1000


I haven't hit any of those thresholds yet, but I suspect I'll surpass the one for MeTa first. I was sort of worried I might be the only such user when that happens, but from a quick query of the Infodump I see there are already about 23 other MeTa-only users for this thing.
posted by FishBike at 5:22 PM on July 17, 2010


I guessed the first 4/5, which is sort of scary.
posted by mecran01 at 6:08 PM on July 17, 2010


I'm pretty sure I sussed out a post by Jonson correctly, every time.
posted by crunchland at 7:11 PM on July 17, 2010


I used to think you were a man, desjardins, but strictly because of my man-crush on Eric Desjardins.
posted by Mister_A at 7:15 PM on July 17, 2010


Dudes I'll BRB I have to go make 1300 comments on AskMe.
posted by Mister_A at 7:18 PM on July 17, 2010 [2 favorites]


I just got crunchland and dave faris as options; it seemed like a trap.
posted by Mister_A at 7:20 PM on July 17, 2010 [1 favorite]


The last time I put on ice skates I spent more time on my ass than upright.
posted by desjardins at 7:21 PM on July 17, 2010


Hah! That's phase 2 in learning to skate. Phase 3 involves actual skating--hang in there!
posted by Mister_A at 7:26 PM on July 17, 2010


"Believe me, klang, you are hard to miss even without the name tag. I think I got three of your comments, and nailed each of them."

It was easier when half of my comments began with some variation of YOU ARE STUPID AND WRONG.

I've just been here so long that I figure people can browse my history if they want to learn why their opinions are STUPID AND WRONG instead of me having to tell them every time.
posted by klangklangston at 7:44 PM on July 17, 2010


heh. pb is definitely the easiest to pick out. Look for a gray post with "Okay, I did some [TECH TECH TECH TECH TECH] so everything should be working fine now As for jessamyn, of all of the mods, she's the most likely to use the plural "we" and speak as a mod even if otherwise commenting (as opposed to cortex and mathowie, who will often jump into threads as users just to comment) and she has a highly distinctive tone as well.

Astro Zombie's got both a shitload of comments and as good a name recognition as the mods, so there's likely a trend towards picking him above all other names whenever anything sounds like him - we'll know when the "false positive" data is mined. (Correct me if I'm wrong, but it seems like the code picks out fake usernames by picking randomly from the comment set, hence users with more comments get chosen as false answers more often.)

The Whelk has a fairly distinctive tone, and has been super active recently, so if one can place the date of the comment, it makes him that much more of a suspect. Pope Guilty also has a very distinctive tone (humorous and aggressive) combined with (sorry dude) anti-religious axe-grinding, which makes him easy to spot as well.
posted by Navelgazer at 7:50 PM on July 17, 2010


Hmph. 105 out of 304, with 0 skipped, is better than 33% accuracy, but I missed some people I thought I should recognize, and for some reason, I have a near-perfect record on recognizing Kattulus. (Note to Kattulus: if you decide to post any embarrassing AskMe questions as anonymous, make sure that I'm out of town or swamped with work first.)
posted by maudlin at 7:52 PM on July 17, 2010


I got busted on an anon question once. I didn't think my style was that recognizable.
posted by desjardins at 7:54 PM on July 17, 2010


Damn, just guessed myself for a comment TomMelee left. Tom, I also <3>
POSTED BY KLANG!
posted by klangklangston at 8:01 PM on July 17, 2010


Whups. Dunno what I did there. "I also <3>
I think I maybe tried to close the <3>

—Todd Lokken

posted by klangklangston at 8:02 PM on July 17, 2010


HEART THE GODDAMN TRIBE CALLED QUEST'S "LEFT MY WALLET IN EL SEGUNDO", TOM. THE MODS ARE TRYING TO SILENCE MY APPRECIATION FOR OUR APPRECIATION OF NATIVE TONGUES RAP!
posted by klangklangston at 8:04 PM on July 17, 2010


YOU ARE STUPID AND WRONG.
posted by crunchland at 8:12 PM on July 17, 2010


OH NO! And crunchland has over 1000 comments here, so it could show up at any time!

I said to myself, crunchland about something I, crunchland, also said.

take that!
posted by klangklangston at 8:15 PM on July 17, 2010


I got busted on an anon question once. I didn't think my style was that recognizable

That happened to me once too. It freaked me out a bit at the time. Turns out I used a phrase in the question that I'd used in a bunch of comments. Between that and my posting history, the person was able to figure it out.

I can spot your comments more easily than others. Same with theora55. I don't know why that is.
posted by zarq at 8:17 PM on July 17, 2010


jeoc writes "Kinda wish I checked the context for this one before I clicked away. But probably that would ruin it"

It's here jeoc.
posted by Mitheral at 9:51 PM on July 17, 2010


I need a MetaTalk only version; but I'm doing better than I thought I would and running at about 33% correct.

And I'm certain there are people I've (mentally) outed in their anon questions. I never thought to meMail them and tell them that his/her slip is showing. Should that be done?
posted by deborah at 10:21 PM on July 17, 2010


The AskMes are throwing me off, I think, because I hardly ever use that part of the site (although there are some people I'm aware know a lot about particular subjects, so that helps). The people who are most recognizable to me thus far seem to be Artw, Klang, Empress Ican'tspellthesecondpartofhername and Miss Lynnster, but I think in most cases I've been fortunate enough to get selections that are really, really obvious. (I would include jessamyn, but a couple of her selections either include her name in the italicized material at the top of her post or are directly concerned with mod questions; if these are trends, they surely have a lot to do with why she tends to be so recognizable.) Pyramid termite is by far the easiest person to spot, provided you've been reading the site long enough to associate him with his, er, distinctive punctuation/capitalization habits.
posted by kittens for breakfast at 10:58 PM on July 17, 2010


I'm so, so, so, so, so, so, so, so, so, so, so, so, so, so, so, so, so, so, so, so, so, so, so, so, so, so, so, so, so, so, so, so, so, so, so, so, so, so, so, so, so, so, so, so, so, so, so, so, so, so, so, so, so, so, so, so, so, so, so, so, so, so, so, so, so, so, so, so, so, so, so, so, so, so, so, so, so, so, so, so, so, so, so, so, so, so, so, so, so, so, so, so, so, so, so, so, so, so, so, so, so, so, glad I'm awful at this.
posted by Joseph Gurl at 11:19 PM on July 17, 2010


Fawlty Towers.. can get Pythonesque, especially since John Cleese is the main character!

It seems to jostle with Only Fools and Horses for #1 on all the "best comedy" polls here in the UK.. but I've certainly never found a foreigner who even heard of OFaH (probably because it wouldn't make any sense).
posted by ??????

- dirtynumbangelboy
- wackybrit
- Goofyy
- amyms
- damn dirty ape
- (I couldn't even guess)


Turns out, I could even guess.
posted by team lowkey at 1:22 AM on July 18, 2010


"Says Who?" is a great name, by the bye, but I just have to throw this one out there for your reconsideration: Eponysteria.
posted by carsonb at 4:20 AM on July 18, 2010


(I couldn't even guess)
(I couldn't even guess)
(I couldn't even guess)
posted by the littlest brussels sprout at 6:15 AM on July 18, 2010


old grumblebee script that displays users' avatars next to their comments

I kind of like that idea. Where would I find this?
posted by Miko at 8:46 AM on July 18, 2010




*is briefly entertained by the idea of a wooden shack called Old Grumblebee's Scripts-N-Stuff*
posted by The Whelk at 8:59 AM on July 18, 2010 [4 favorites]


I'd shop there... If grumblebee wore overalls and totaled up purchases on an abacus while puffing on a corncob pipe.
posted by the littlest brussels sprout at 9:03 AM on July 18, 2010 [1 favorite]


I am profoundly inept at this game. Poo.
posted by turgid dahlia at 6:07 PM on July 18, 2010


Yes, I got one! It was Burhanistan, and he said "boobs". I guess some things never really leave you.
posted by turgid dahlia at 6:13 PM on July 18, 2010


It took me 99 tries before I saw my own name. This game sucks.
posted by mr_crash_davis mark II: Jazz Odyssey at 8:26 PM on July 18, 2010


I got 99 tries but a mr_crash_davis mark II: Jazz Odyssey aint one.
posted by The Whelk at 8:30 PM on July 18, 2010


I can only tell which posts are Mutant's. I always know because it's like three paragraphs long and usually about finance.
posted by anniecat at 8:01 AM on July 19, 2010


21/50
posted by The Straightener at 10:16 AM on July 19, 2010


I am confused. How does this game end? How do I win? Am I better off with a pass or a wrong answer? 7 fer 21
posted by JohnnyGunn at 10:49 AM on July 19, 2010


Wow, I suck something fierce at this; I guessed wrong and it was me.

In fairness, it was a movie quote, but still.
posted by quin at 1:17 PM on July 19, 2010 [1 favorite]


> How do I win?

Congratulations, you have already won! Contact cortex for your prize.
posted by languagehat at 2:21 PM on July 19, 2010


You guessed dios; it was Ambrosia Voyeur.

How long will you taunt me with my late night lack of skillz, "Says Who" ? HOW LONG!
posted by honest knave at 2:55 PM on July 19, 2010


5/10, which is better than I did in most of primary school. Take that Mrs. Tolson! (who was a lovely lady when I met her and I was all growed up).
posted by arcticseal at 4:57 PM on July 19, 2010


My dreadful score was partially redeemed only by the following:
Yeah, languagehat would be proud.

And indeed I was.

But only if "hamhandedly" included the hyphen: "ham-handedly".

Well done, sir! (Though I would not have used the British placement of the period, mind you.)

Otherwise he may just be irked.

Aargh... might be irked! Might!! You wanna see irked?
*blasts away with shotgun*
posted by ??????
Even then I was half-convinced it was a trick question.
posted by Doktor Zed at 6:51 PM on July 19, 2010


Kinda late to this, but I had to say that this is a lot of fun. Currently 3 for 5, with two skipped. The only one I got right that I didn't randomly guess at was one of Miko's from AskMe, which I barely visit, but it just sounded like her to me.
posted by Marisa Stole the Precious Thing at 8:36 PM on July 19, 2010


Got a ParisParamus quote right, but it was in a political thread so that was a gimme.
posted by Civil_Disobedient at 3:55 AM on July 21, 2010


Oh, you should probably weight comments by # of favorites, cortex. The higher the threshold, the better the chances that RandomUser will recognize them.
posted by Civil_Disobedient at 4:03 AM on July 21, 2010


It's depressing that I spend so much damn time reading this site and yet have failed to enter its inner life, as evidenced by my utter lack of clue re: this game.
posted by flotson at 6:10 PM on July 21, 2010




« Older Let's hear it for New York   |   Event Organizers should be able to edit IRL events Newer »

You are not logged in, either login or create an account to post comments