Skip

Shocking news: personality doesn't matter when dating online
July 28, 2014 10:09 AM   Subscribe

OkCupid: we experiment on human beings! But by comparing Love Is Blind Day to a normal Tuesday, we learned some very interesting things. In those 7 hours without photos...people responded to first messages 44% more often, conversations went deeper...in short, OkCupid worked better. When the photos were restored at 4PM, 2,200 people were in the middle of conversations that had started “blind”. Those conversations melted away. The goodness was gone, in fact worse than gone. It was like we’d turned on the bright lights at the bar at midnight.
posted by shivohum (96 comments total) 43 users marked this as a favorite

 
So what they're saying is that the site was doing better at what OkCupid thinks its purpose should be. It assumes that physical attraction is shallow, and love should be blind. That's a very popular ideal in this culture, but it doesn't really describe how the world works, or how we want it to work if we're being honest about it.

Surely the fact that people changed their behavior when given more information indicates that the site with photos was doing a better job of achieving what people really wanted in terms of seeking mates. They wanted to find someone personally compatible, to be sure, but they also wanted to be physically attracted to that person.
posted by Naberius at 10:15 AM on July 28, 2014 [14 favorites]


OK Trends is back?!? Christmas in July!
posted by Skwirl at 10:16 AM on July 28, 2014 [23 favorites]


Agreed. I also chuckled at, "Contact details (phone numbers, emails) were exchanged more quickly." Yeah, no kidding. People wanted to see pics.
posted by cribcage at 10:16 AM on July 28, 2014 [23 favorites]


I was a little annoyed that the title here is a gross exaggeration of the quoted results.

RTFA, it turns out that "personality doesn't matter when dating online" is only a slight exaggeration of the complete results. Damn.
posted by roystgnr at 10:16 AM on July 28, 2014 [3 favorites]


Probably better to link to the static url, if OkTrends is back.
posted by the man of twists and turns at 10:16 AM on July 28, 2014


MetaFilter: obviously a really cool person to hang out and talk to and clutch driftwood with
posted by theodolite at 10:21 AM on July 28, 2014 [2 favorites]


The last time I made an okcupid page I went 100% sarcastic, making my profile around a cursed, immortal ghost. It was fun until the site kept insisting I use a person for my profile picture and kept deleting my ghost selfies.

Like it was actually fun.
posted by hellojed at 10:23 AM on July 28, 2014 [17 favorites]


Okay, when I first saw object d'schmoop's ad on OKC, it was his picture that made me stop and take a closer look. (He is ADORABLE - he looks like he's half Muppet.)

But it was the text itself that made me decide "okay, THIS guy gets a message."
posted by EmpressCallipygos at 10:29 AM on July 28, 2014


If I was still single I'd just eliminate the anguish and use tindr. I know a handful of couples who met that way. It's no fun when you see someone and then know everything about them (via their written profile); it's much better to have a mild-to-moderate physical attraction and then uncover the mystery from there. That's how it typically happens IRL.
posted by St. Peepsburg at 10:29 AM on July 28, 2014 [5 favorites]


It assumes that physical attraction is shallow, and love should be blind. That's a very popular ideal in this culture, but it doesn't really describe how the world works, or how we want it to work if we're being honest about it.

Agreed. I also chuckled at, "Contact details (phone numbers, emails) were exchanged more quickly." Yeah, no kidding. People wanted to see pics.


This. I grew up in one of the PC capitals of the world and was constantly told that looks don't matter. That is such utter horseshit I can't even tell you.
posted by Melismata at 10:33 AM on July 28, 2014 [11 favorites]


>Basically, people are exactly as shallow as their technology allows them to be.

Thus spake McLuhan.
posted by Skwirl at 10:35 AM on July 28, 2014 [6 favorites]


Looks are a big part of it, sure, and will generally be the first-order filter for dating, but this study (Yay! OKTrends is back!) ignores two big things:

1. OKC actually puts a big personality filter before the pics, in that their questionaires will work to show you people who are already half-way to compatible.

2. A profile pic tells a lot more information than merely whether someone is hot or not. It says how a person presents themselves, what they think is important, what their body language is like, etc.

So yeah, we're shallow, but it's not quite as bad as all that.
posted by Navelgazer at 10:39 AM on July 28, 2014 [12 favorites]


If a frog had wings, it wouldn't bump its ass a-hoppin'.
posted by OmieWise at 10:40 AM on July 28, 2014 [1 favorite]


I don't know, OkCupid's matching algorithm isn't magic but I find it to be an extremely helpful benchmark. 95%+ usually makes me think, "wow, this person seems really, freakishly cool!". 80%+ are all potentially viable dates. Less than 70% or so tend to strike me as people I deeply wouldn't get along with; and I usually set enough dealbreakers (fucking obvious stuff like don't be a self-avowed racist) in answering questions that it's pretty likely I really wouldn't. The primary problem is that, yeah, actual attraction is ineffable, and it often seems like my top matches are more besties material or personality clones of myself than people I would want to get jiggy with.
posted by threeants at 10:42 AM on July 28, 2014 [11 favorites]


Yeah I agree with Navelgazer - the "looks" component of attractiveness is at least as much about looking for someone who is a compatible type of person.

For example, the conventionally attractive, clean-cut, polo-wearing dude is not going to draw my attention as much as an average-looking bearded fellow with a flannel shirt. I can't see myself getting along with the former as well as I would with the latter.
posted by mai at 10:44 AM on July 28, 2014 [2 favorites]


Hey, there's going to be a book! Could be interesting.
posted by Chrysostom at 10:48 AM on July 28, 2014


Heh, I liked the split looks/personality rating, and I'd regularly give divergent answers. Also give me back the bell curves on the personality tab.
posted by vibratory manner of working at 10:49 AM on July 28, 2014


an average-looking bearded fellow with a flannel shirt

Ahem.

OkCupid's matching algorithm isn't magic but I find it to be an extremely helpful benchmark. 95%+ usually makes me think, "wow, this person seems really, freakishly cool!". 80%+ are all potentially viable dates. Less than 70% or so tend to strike me as people I deeply wouldn't get along with; and I usually set enough dealbreakers in answering questions that it's pretty likely I really wouldn't.

They tested this, and found that telling people (who were bad matches) that they were good matches worked fine, and the reverse holds true as well.
posted by Dip Flash at 10:51 AM on July 28, 2014


Yeah I agree with Navelgazer - the "looks" component of attractiveness is at least as much about looking for someone who is a compatible type of person.

Given that Mr. bowtiesarecool insists that one of the reasons he responded to my message was my choice of hat in my primary photo, my anecdata supports this theory. Your choice of background/attire/activity/pose in your photos can say something about who you are and what you're about, and that matters in addition to plain physical features.
posted by bowtiesarecool at 10:53 AM on July 28, 2014


also every time I've started a new OkCupid account it's always all,

You are:

more thrifty
more sloppy
more laid-back
less into exercise
less ambitious
less interested in love
less interested in sex

which makes me think both that their algorithm is highly accurate and that there is no hope for me in this world
posted by threeants at 10:54 AM on July 28, 2014 [23 favorites]


When I met my associate on OKC, he had not one picture where I could properly see his face. After we started emailing, he sent me a picture of his bicep, where he had drawn an anchor (in response to my asking if he had any tattoos) but not a single picture of his face. When we finally met IRL, he was wearing aviators and I told him to please take off your sunglasses, upon which I finally saw his face. Flustered, I could only stammer out "Well, you are very attractive" because he's beautiful. It was like an fabulous and unexpected gift, because I had no idea what he really looked like.
posted by alltomorrowsparties at 10:54 AM on July 28, 2014 [12 favorites]


I met my husband online. Even though he's a total cutie, I wouldn't have paid attention to him on the street because he wasn't the type of guy I was "into." (Woe to limiting ourselves with "types.") Yes, we were matched, but it was his personality that totally caught my eye at first. I'm SO glad I didn't see his photo right away or I might have not responded to his message for precisely the reasons Navalgazer outlined.

Also, um, how come people aren't upset that OKCupid was essentially...running experiments on them? Didn't Facebook do that with a resulting outcry? (Not on either, so just curious.)
posted by barchan at 10:56 AM on July 28, 2014 [2 favorites]


They tested this, and found that telling people (who were bad matches) that they were good matches worked fine, and the reverse holds true as well.

I think the average user does not answer many questions, and thus has a pretty ambiguous sense of what is actually a good match. In my experience people who match me highly and people who match me very little are incredibly distinct, and I feel extremely confident that I could pick them apart in a blinded experiment, but obviously I can't prove that.
posted by threeants at 10:56 AM on July 28, 2014 [1 favorite]


which makes me think both that their algorithm is highly accurate and that there is no hope for me in this world

I wouldn't put too much faith into their personality descriptors. I was told I was both more literary and less well-read.
posted by lunch at 10:58 AM on July 28, 2014 [2 favorites]


Having gone on plenty of dates with 90%+ matches and being bored out of my mind, I no longer use the match percentage to gauge whether someone is going to be cool or interesting.

All it means is they meet a lot of my personal minimum standards for human decency: They're progressive, they're pro-choice, they don't think taxes and government are evil. I think of the match percentage as more of a "not-horrible percentage" because that's all it really is. Which is really nice and I think a huge advantage over off-line dating - meet someone at a bar, you don't know if they're a gun fanatic or think women are obligated to shave their legs or what have you.

But just because someone is liberal and non-religious, or prefers music festivals over sci fi conventions doesn't mean I'm automatically gonna be soul mates with them. The match percentage is useful for screening people, but that's it.
posted by windbox at 11:02 AM on July 28, 2014 [11 favorites]


I was an unwilling participant in OKC's most avant-garde experiment to date. I received a message from an attractive woman with a witty profile. We shared many interests, including certain things I vaguely recall idly googling the previous week. Our messages grew lengthly. She had an encyclopedic knowledge and always signed off on a salacious note, such as “you won't believe what happened next.”

I had to know more, so we arranged a meeting at Starbucks. I was stood up, of course. I figured she was out of my league, but the next week I received a $25 check from OkTrends and a letter informing me, “Your efforts towards Total Brand Synergy have been invaluable.”
posted by gorbweaver at 11:04 AM on July 28, 2014 [28 favorites]


You Won't Believe What Appealed To Me About Your Profile

Ten Mindblowing Cultural Products You Like That I Also Like

You Came Up On My QuickMatch. The Personality Traits We Share Will Shock You
posted by threeants at 11:06 AM on July 28, 2014 [16 favorites]


Also, um, how come people aren't upset that OKCupid was essentially...running experiments on them?

I dunno, who do you care more about -- your friends or some random people you've never met?
posted by effbot at 11:09 AM on July 28, 2014 [2 favorites]


As many people know, OKC has very poor options for sex (F/M only), gender (no options) and sexual orientation (gay/bi/straight). As a non-cisgender queer attracted to similar non-cisgender queers, unless the person happens to put descriptors (and *spell them correctly* so they can be text-string searched) in their essays, pix are all I have to go on. But visuals are also data!

My searches would also be a lot easier with a simple data field: hair length. Most of the people I'm into have short or no head hair but no or very little facial hair.

OKC's analytics also target me as a 30-something straight chick - I'm at a loss as to what drives that marketing utter fail.
posted by Dreidl at 11:09 AM on July 28, 2014 [6 favorites]


OKCupid also sells the fuck out of your information.
posted by entropicamericana at 11:09 AM on July 28, 2014 [3 favorites]


how come people aren't upset that OKCupid was essentially...running experiments on them?

The blog addresses this. They basically brush off the concern ("you're always being experimented on, so chill"), which might be unsatisfying but it's a response. But more, it doesn't sound like anything secretive was done. Somebody who actually uses OKCupid can correct me if I'm wrong, but my understanding of the article is that, to launch their new app, they held a celebratory "Love Is Blind Day" probably with banners, emails, etc, in addition to the missing photos. It doesn't sound like anybody was tricked or misled. "We experimented on people" is just a sensationalistic way to get clicks for the unremarkable data they collected.
posted by cribcage at 11:12 AM on July 28, 2014 [3 favorites]


It doesn't sound like anybody was tricked or misled.

Ah, cribcage, I missed that it was so blatantly advertised. Thanks.
posted by barchan at 11:18 AM on July 28, 2014


Also, um, how come people aren't upset that OKCupid was essentially...running experiments on them? Didn't Facebook do that with a resulting outcry? (Not on either, so just curious.)

Facebook set out to manipulate people's moods. This study isn't really a manipulation of internal states.

(many ethics review boards require that mood manipulation studies have a post experimental manipulation to restore those negatively effected (affected?) - yay candy!)
posted by srboisvert at 11:19 AM on July 28, 2014 [2 favorites]


I've begun to realize that whether I'm attracted to someone or not is 100% impossible to predict from an OKC profile. So I keep going on dates with people who seem great/fun/attractive and then it's just a big ol pile of no chemistry. Sigh.
posted by showbiz_liz at 11:22 AM on July 28, 2014 [5 favorites]


I think personality matters with in the subset of people you're attracted to. For me, attractive with a great personality would win out every time over super freaking attractive with an eh personality. On the other hand a person I'm not attracted to to begin could have the world's greatest most-compatible personality for me but there'd still be no hope because the physical aspect wasn't there.

Also, photos aren't the end-all, be-all for gauging physical attractiveness. I think voice, mannerisms, smell and all that factor in in a big way too.
posted by Jess the Mess at 11:24 AM on July 28, 2014 [1 favorite]


Is it possible, on OKCupid, to search for people who don't show themselves in profile photos? I'm attached, but I think those would be the ones I'd want to meet.
posted by amtho at 11:25 AM on July 28, 2014


Weirdly, when I briefly had an OkCupid profile, I got a number of inquiries (from queer women - queer women and gender non-conforming people being the desired demographic) although I had not posted photos. I didn't have any - really, I don't generally have my photo taken - and I figured I'd just write the damn thing while I felt inspired and add pictures later, assuming that the profile would sink to the bottom of the pool and no one would even read it. But lo, they did! None of them seemed wildly out of my league looks-wise, but I suppose that's the advantage of marking the "fat person" thingy and being in your late thirties - my assumed looks were probably pretty close to my actual looks.

Honestly, it's voice that does it for me - there are people who are immediately physically attractive, but even if someone's appearance does nothing for me by itself, the right speaking voice and use of language turns everything around. Also glasses. It's really difficult for me to be attracted to people who don't wear glasses.
posted by Frowner at 11:27 AM on July 28, 2014 [1 favorite]


It doesn't sound like anybody was tricked or misled.

@cribcage, isn't this the definition of misleading someone?
To test this, we took pairs of bad matches (actual 30% match) and told them they were exceptionally good for each other (displaying a 90% match.)
I realize there's a footnote explaining that users were informed after, but that's basically "let's ask for forgiveness not permission."
posted by kyp at 11:28 AM on July 28, 2014 [1 favorite]


I'm newly single and OK Cupid looks like a bit of fun if nothing else, but I am paralyzed with fear of choosing the wrong username! What to do?

On topic: I am SRSLY hot.
posted by Mister_A at 11:36 AM on July 28, 2014


Here's Tim Carmody's take on why he thinks people weren't as upset by the OKC manipulations, compared to facebook's.
posted by Mchelly at 11:38 AM on July 28, 2014 [2 favorites]


Obvious answer: It's way more socially acceptable to use Facebook, still, than to use a dating website, so the quantity of the uproar is naturally going to be a lot less.

This doesn't really do much, anyway, but confirm what I already knew. I am not attractive. Therefore, none of my other traits really matter. They might as well not exist.
posted by dekathelon at 11:39 AM on July 28, 2014 [3 favorites]


OKC is pretty open about their "experimentation" and most of that is low to no-input (i.e. just observing the aggregate data of what their users are already doing) so it's a bit different from Facebook.
posted by Navelgazer at 11:39 AM on July 28, 2014 [1 favorite]


> It was like we’d turned on the bright lights at the bar at midnight.

"My job is to take away the beer goggles just when the party's getting good." William McChesney Martin Jr. OKCupid
posted by jfuller at 11:42 AM on July 28, 2014 [1 favorite]


For me the difference is this:
"The Facebook study had an added wrinkle in that academics were involved in designing the study and writing it up. This raised all sorts of factual and ethical issues about university institutional review boards and the responsibility of the journal's editors and publishers that don't seem to be relevant here." (from the Tim Carmody link just above).

I disagree that they're not relevant here. Facebook/OKCupid are free to manipulate data (or their users) however they want-- it's the academics involved that make things sketchy for me. I expect corporations to act only in their own self-interest, but researchers I expect to abide by ethical guidelines.
posted by matcha action at 11:43 AM on July 28, 2014


I think another reason people are not upset by OKC's experiments is their relationship to the service. People relate to OKC as a platform for running their own experiments (dates), and there is this expectation that everyone involved is trying to find the best way to identify a successful match.

On the other hand, facebook is a medium for your so-called private life, and when that gets experimented on, it feels much more invasive and exploitative. People don't want intruders in their home, even if that home doesn't belong to them.
posted by I-Write-Essays at 11:47 AM on July 28, 2014 [6 favorites]


I am paralyzed with fear of choosing the wrong username! What to do?

Honestly, I wouldn't sweat it. When I was on OKC and similar sites, I always chose long phrases (like "thesearesomewordsiknow" or "twoturntablesandamicrophone") as my username--partly because I have an irrational hatred of how usernames with trailing numbers look--and I don't think anyone ever once made a comment about it.
posted by psoas at 11:48 AM on July 28, 2014 [1 favorite]


I also hate enumerated usernames. My method for finding unique usernames is to search my bookshelves for obscure proper nouns.
posted by I-Write-Essays at 11:51 AM on July 28, 2014 [1 favorite]


I just made up a name. Since it's unique, nobody else has it on any site, so I can use it anywhere.
posted by Mitrovarr at 11:53 AM on July 28, 2014 [1 favorite]


When I first started using a dating system back in the 1990s, it was by telephone, so it was all blind dating and the only physical characteristic you had to go on was the person's voice. I found I connected with more people and went on more dates, but most of the dates went nowhere as we weren't attracted to each other. I'll take fewer dates with a higher probability of being attracted to the person, thank you. It's way more efficient.

I am paralyzed with fear of choosing the wrong username! What to do?

Whatever you do don't call yourself therapistboy. I damn near reported someone with that user name until I realized the words didn't break down as I had first assumed. I used to use Orange Swan on dating sites until the more enterprising bachelors started piecing together who I was elsewhere online, and finding out much more about me than I was comfortable revealing so early in our acquaintance. Now I use a name that I only use for dating sites.
posted by orange swan at 11:54 AM on July 28, 2014 [6 favorites]


My OKC nick is a variation on the one I use here. But my profile's blank ATM as I haven't had time to date anyway and I wanted to break it all down and start up from scratch.
posted by ursus_comiter at 11:56 AM on July 28, 2014 [1 favorite]


I think people aren't as upset about OKC because everyone expects that at least half the people on the site are lying anyway. No one is taking it as Very Very Serious and True. It gives you a decent idea of whether you want to sink 45 minutes into coffee with someone. It doesn't present itself as being the funnel by which all of your family and friends will communicate. Big difference.
posted by stoneweaver at 11:58 AM on July 28, 2014 [3 favorites]


If they tell 30% compatible people they're 90% compatible, 17% have a "real" conversation. If they tell 90% compatible people they're 90% compatible, 20% have a conversation.

The lack of significant difference between 20% and 17% would make me conclude that OK Cupid's matching is completely unsuccessful. I'm curious why anyone would think otherwise. All that math, for 3 more conversations in 100? That's not much value added. 85% of your conversations were basically guaranteed, and the last 15% were possible the result of value-added by their filtering, and possibly just statistical noise.

It's especially troubling that it's non-linear: 60% matches that were told they were 90% matches were actually slightly less likely (16% compared to 17% and 20%) to have a conversation after a message.

"The mere myth of compatibility works just as well as the truth. " Woah, if you own OKC stock, sell!
posted by anotherpanacea at 12:00 PM on July 28, 2014 [2 favorites]


how come people aren't upset that OKCupid was essentially...running experiments on them?

In the case of Facebook, I think most of us want as unmediated an experience as possible. (It's probably the #1 reason I prefer Twitter; I get all the tweets of everybody I follow, not a sampling picked by the platform.) Facebook's "top stories" meddling is, at best, grudgingly tolerated, but the further it strays from an honest attempt at picking the stories most relevant to the user, the more people are likely to get upset. We have chosen to follow various people; we don't want Facebook deciding what we see of theirs, and any experimenting they do with that is unlikely to be to our benefit.

By contrast, the main selling point of OkCupid is that it analyzes the hell out of its data to better figure out whose profiles to show you. I think it's understood that in order to make those algorithms better, it has to do a certain amount of experimenting. We want them to keep doing it and refine their service even more.
posted by Shmuel510 at 12:00 PM on July 28, 2014 [4 favorites]


Data point: my crushing lack of self confidence leads me away from women I deem too attractive. The most beautiful women I've dated were from faceless ads, or old-school print ads.
posted by MrMoonPie at 12:02 PM on July 28, 2014 [2 favorites]


I consider this experiment successful because it made me log back into okcupid after a couple of months off. Hope they keep writing their blogs about data, because I've found those far more interesting than most people I've met on the site, match percentages be damned.
posted by antonymous at 12:05 PM on July 28, 2014 [1 favorite]


anotherpanacea: The lack of significant difference between 20% and 17% would make me conclude that OK Cupid's matching is completely unsuccessful.

Really? I'd just take it as a sign that having a single conversation is a really bad metric. Deep incompatibility would take a while to turn up, particularly if people were told they were actually very compatible, so a measure of couples that actually met and dated for a significant time would be a much better measure. Of course, that would require a much larger sample set and be even more ethically questionable.
posted by Mitrovarr at 12:05 PM on July 28, 2014 [1 favorite]


Mitrovarr: I'd just take it as a sign that having a single conversation is a really bad metric.

Given that the overall variance is between 10% (for super-incompatible) and 20% (for super compatible) response rate, they'd need to run this experiment on a pretty large n in order to justify their business model. You're certainly right that a 4-message conversation does not a successful marriage make. But that adds even more uncertainty, and makes the data even more suspect. It's the joke about the drunkard, looking for his keys (lost far away) by the lamppost because that's where the light is.

There's also risk of regional variances or certain personality opposites that are "more compatible." Maybe average looking people prefer more personality differences than hot people. Maybe people with college degrees demand more similitude than people without.

So basically you should HOPE that their filters are constantly being tinkered with and that you're constantly being shown experimental data. Because this is the great question of human nature and it's going to be hard.

But if you figure that, you'll end up responding less in general (treating everyone as incompatible) and they'll have even less data!
posted by anotherpanacea at 12:10 PM on July 28, 2014 [1 favorite]


justify their business model.

No they wouldn't. Their business model is making money off advertising and subscriptions. The justification of their business model is that people are using them in enough of a critical mass to provide that money. The algorithm research stuff, or the fact that it does succeed sometimes in creating genuine connections, is just the bullshit on top of the pie.
posted by dekathelon at 12:12 PM on July 28, 2014 [1 favorite]


The algorithm research stuff, or the fact that it does succeed sometimes in creating genuine connections, is just the bullshit on top of the pie.

Except that if that were widely believed, their response rate would plummet, according to their own data, by approximately half. And that's going to affect advertising and subscriptions.
posted by anotherpanacea at 12:14 PM on July 28, 2014


Huh.

Continue Testing
posted by Slackermagee at 12:21 PM on July 28, 2014


The best user names, in my experience, are short evocative sentences that appear to make a conclusive claim about something, but are ultimately more or less meaningless.
posted by You Can't Tip a Buick at 12:24 PM on July 28, 2014 [31 favorites]


anotherpanacea: Given that the overall variance is between 10% (for super-incompatible) and 20% (for super compatible) response rate, they'd need to run this experiment on a pretty large n in order to justify their business model. You're certainly right that a 4-message conversation does not a successful marriage make. But that adds even more uncertainty, and makes the data even more suspect. It's the joke about the drunkard, looking for his keys (lost far away) by the lamppost because that's where the light is.

Ultimately, the only conclusion I'd really draw from that experiment is that people care about the reported match percentage. Which is not really surprising. The very small differences between actual high matches and bad matches reported as high seems like it could easily just be statistical error. And I really think their metric doesn't work, even in theory.

However, I think this is a tricky experiment to run because it feels like setting up people with extremely poor compatibility could have negative real-world consequences. In particular, you could set up people with severe political/cultural incompatibility that don't catch it until later, which could set people up for actual violence. At the very least, you are likely to waste the time of everyone involved, like setting up an evolutionary biologist with a creationist, or something.
posted by Mitrovarr at 12:24 PM on July 28, 2014


The best user names, in my experience, are short evocative sentences that appear to make a conclusive claim about something, but are ultimately more or less meaningless.

personally I find short noun phrases particularly sexy
posted by threeants at 12:30 PM on July 28, 2014 [3 favorites]


user names that begin and end in vowels are the most attractive, if you ask me.
posted by entropicamericana at 12:32 PM on July 28, 2014 [3 favorites]


personally I find short noun phrases particularly sexy

Perhaps you'd like to meet my friend "SackOfToes."
posted by Edgewise at 12:41 PM on July 28, 2014 [10 favorites]


Semi-obscure lit-crit jargon with a nice patter-y rhythm and an internal near-rhyme is just about perfect, I think.
posted by Navelgazer at 12:43 PM on July 28, 2014 [1 favorite]


You know, all the old received wisdom about looks not mattering wasn't meant to be an empirical statement about how people actually go about their usual daily business. It was meant to be wise advice from a more experienced, worldlier POV. The point was supposed to be that, while people definitely do show a strong tendency to overemphasize appearances when looking to start relationships, the reality that most people's experience in life eventually bears out is that appearances don't really matter in a more fundamental way (or specifically, relying on appearances to make judgments about who you seek out for a relationship doesn't reliably predict future relationship happiness because other personal characteristics are more useful predictors of that). So I think, depending on how you parse the claim, it's either only trivially true or completely misses the point.

Also, I think OKCupid has always been pretty upfront and public about the fact they run behavioral experiments on their user-base, so I suppose in this case, users knew what they were getting into. Doesn't seem like they did anything with the potential to harm anyone too badly here, unless I'm missing something.
posted by saulgoodman at 12:46 PM on July 28, 2014 [2 favorites]


Also, I think OKCupid has always been pretty upfront and public about the fact they run behavioral experiments on their user-base, so I suppose in this case, users knew what they were getting into.

I dunno, I really didn't like it when they set me up with a single mother who turned out to be made of cloth and wire.
posted by prize bull octorok at 12:54 PM on July 28, 2014 [45 favorites]


I thought she was nice, prize bull octorok, till the electric shocks began.
posted by bswinburn at 1:01 PM on July 28, 2014 [2 favorites]


All I know is that Beatles references make terrible user names. Christ, it's grandpa music.
posted by Elementary Penguin at 1:02 PM on July 28, 2014 [4 favorites]


Also, I think OKCupid has always been pretty upfront and public about the fact they run behavioral experiments on their user-base

This was the main reason I joined in the alpha phase of the site... they experimented and then *shared the results*, aka OKTrends Blog. Their results were useful to me in improving my profile and how I approached other users. When they were bought out by match.com, the blog stopped being updated. Booo.
posted by Dreidl at 1:02 PM on July 28, 2014


it's much better to have a mild-to-moderate physical attraction and then uncover the mystery from there. That's how it typically happens IRL.

That seems to underestimate the contextual information you receive when meeting someone in person. For instance, I met my fiancée at church and was introduced to her by friends of mine, both of which provided quite a bit of contextual information.
posted by Jahaza at 1:07 PM on July 28, 2014 [1 favorite]


Also, for women, "the mystery" tends to be "the mystery of whether this person will assault me or worse," so, y'know, it's not so enticing.
posted by dekathelon at 1:24 PM on July 28, 2014 [4 favorites]


Also, um, how come people aren't upset that OKCupid was essentially...running experiments on them? Didn't Facebook do that with a resulting outcry? (Not on either, so just curious.)

Well, i don't buy the "academics were involved" argument for the outrage outside of MeFi. 99% of what i saw was "how dare they experiment on us! despite all my rage i'm still just a rat in a cage!".

I honestly think it's because most people see facebook as "evil", it's in that trifecta along with google and apple of corporations that are cool to dislike now.

OKC is kind of benevolent, and also sort of shamed/looked down upon by people who think online dating of that variety is "stupid" or for oldsters or whatever since they either decry it entirely or are young and use tinder if anything.

Also, lots of buzzfeed type sites sold the outrage of the facebook one perfectly. Give it a few days for this to go viral and we'll see how outraged people get. It hasn't even popped up on my newsfeed/timeline on any social media sites yet.

But seriously, i think a lot of the facebook outrage from laypeople who didn't get the academic angle was essentially because they were told they should be outraged. It was pretty "thing bad!" from what i saw. And from the other-than-academics angle, i really don't see how this is any less outragey at all.
posted by emptythought at 1:45 PM on July 28, 2014


I honestly think it's because most people see facebook as "evil", it's in that trifecta along with google and apple of corporations that are cool to dislike now.

For me, it was more that I've spent many years fighting off chronic depression and after some initial resistance, had finally let myself buy into the idea that maybe FB could represent a new way to maintain my social life despite the demands of a challenging career and family responsibilities as a parent, and the thought that FB might be deliberately fucking with my feed to try to make me sad after finally winning me over to the idea that the new normal for social existence in the US might not be unbearably oppressive and dehumanizing just really pissed me off in ways that I'm pretty sure are completely justified.
posted by saulgoodman at 1:53 PM on July 28, 2014 [3 favorites]


I was told I was both more literary and less well-read.

That is a real thing, though.
posted by michaelh at 2:11 PM on July 28, 2014 [2 favorites]


Um, is the article supposed to end after the line ". “Love Is Blind Day” on OkCupid—January 15, 2013."? Because I got nothing.
posted by jenfullmoon at 2:35 PM on July 28, 2014


There are more data-nerds on OKCupid, and you're already trusting the algorithm to match you. People are less heavily invested in OKC for *all* their online social interaction since it has a particular purpose.

On Facebook you're trusting the algorithm to show you the "right" updates from friends and family but most people probably don't think about that and just assume they're seeing everything that matters.
posted by subdee at 2:35 PM on July 28, 2014


The critical difference between Facebook and OkCupid, I think, is that Facebook was tampering with existing relationships, whereas OkCupid was tampering with potential relationships. There's a ton of psychology research showing that people place much stronger weight on avoiding loss than on creating gain, which is why the Facebook manipulation seems worse: Facebook is mostly about maintaining existing relationships, so the manipulation threatens to take away something people already have. In contrast, OkCupid is about creating new relationships, and so the manipulation only threatens to prevent gain rather than create loss.
posted by Pyry at 2:43 PM on July 28, 2014 [11 favorites]


Is it possible, on OKCupid, to search for people who don't show themselves in profile photos?

I've wondered this as well.

When I run searches, I've never had a profile come up in the results that was photo-less. So, I'm assuming they don't show up.

As far as I can tell, the only way to view a photo-less profile is if you already know the username -- say, if a photo-less user messages you first -- and you access the profile directly.

I'm surprised that people choose to go completely photo-less, rather than at least slapping up some hair-covering-face-plus-sunglasses (basically, unrecognizable) photo to allow themselves to show up in the results. That's what I did back in the day (and oh, the rage I got from guys DEMANDING that I add more photos to my profile. It was a pretty useful screening mechanism for entitled a-holes, actually.)

You also need to have a photo up yourself in order to see other people's photos (other than the rollover previews).
posted by nacho fries at 2:54 PM on July 28, 2014


Typing with my OKC mod hat on - on the desktop version, historically there was a user settings checkbox (Browse Matches - Who I'm Looking For) that limited/didn't limit searches to profiles with photos. Default was profiles with photos only. I'm not sure if that's still a user option, or that no-pic profiles will only see no-pic profiles. I don't think the iOS or other mobile versions have this option at all.

As a user, I have pics and I see both pic and non-pic profiles. Non-pic users show up in my Visitors list as well. Quickmatch only shows users with pics.
posted by Dreidl at 3:25 PM on July 28, 2014


Ah, that's right -- I recall there used to be an option to only show users with photos, but I just took a look-see on OKC, and I'm no longer seeing it as an option anywhere. I'm using the non-mobile version of the site on my laptop.

When I remove my profile photo, and do a search, I'm still only shown matches that have photos.
posted by nacho fries at 3:47 PM on July 28, 2014


Nine times out of ten the opening comment is on my picture. And the guys with the single kinda blurry shot pulled out so you can't see his features clearly always asks me for more pictures, especiaally full body. I've given up on people actually reading my profile or caring what I say.
posted by FunkyHelix at 4:06 PM on July 28, 2014


The most important thing in a username is frank and unambiguous disclosure of one's sexual orientation.
posted by straight at 4:19 PM on July 28, 2014 [8 favorites]


I don't have an online dating profile. But if I did, and I didn't have a picture posted and someone asked me to describe myself I would joke:

I look like Slartibartfast must have looked like when he was 45.
posted by juiceCake at 4:52 PM on July 28, 2014


juiceCake: look like Slartibartfast must have looked like when he was 45.

Ha! Slartibartfast1969. I knew that was you.
posted by Sing Fool Sing at 5:23 PM on July 28, 2014 [1 favorite]


when choosing a username it is best to eschew vowels and consonants altogether
posted by 5_13_23_42_69_666 at 5:28 PM on July 28, 2014 [3 favorites]


Whatever you do don't call yourself therapistboy. I damn near reported someone with that user name until I realized the words didn't break down as I had first assumed.

Maybe he was just a fan of this SNL skit.
posted by obliterati at 6:26 PM on July 28, 2014


Wow they totally fudged the coloring on that little heat map at the end - there are squares colored differently that have the same value. Also no scale bar. Revise and resubmit.
posted by en forme de poire at 6:39 PM on July 28, 2014 [1 favorite]


"It assumes that physical attraction is shallow, and love should be blind. That's a very popular ideal in this culture, but it doesn't really describe how the world works, or how we want it to work if we're being honest about it."

Speak for yourself and only yourself and try to avoid speaking for other people about what you presume are their secret, inner thoughts. There's a psychological term for that.
posted by Ivan Fyodorovich at 9:46 PM on July 28, 2014 [1 favorite]


Wow they totally fudged the coloring on that little heat map at the end - there are squares colored differently that have the same value. Also no scale bar. Revise and resubmit.

Depressingly (hilariously? weirdly?), I can bet you exactly why that is - it's made in Excel using the automatic conditional formatting, and I suspect they went by row (so values are colored relative to other values in the same row).

My life is basically Excel conditional formatting right now.
posted by quadrilaterals at 2:00 AM on July 29, 2014 [2 favorites]


Please keep the username tips coming because I obviously blew it the first time. Thanks.
posted by DynamiteToast at 7:38 AM on July 29, 2014


For a while I had a Chinese idiom for my username, but that just resulted in me explaining it over and over and also having to admit to interested Mandarin speakers that I actually don't know any of the language anymore.

Now my username is just tacky_as_fuck because I no longer give a shit and am basically just conceptually hatefucking OKCupid at this point.
posted by showbiz_liz at 8:10 AM on July 29, 2014 [5 favorites]


Early Byzantine saint names are good.
posted by Chrysostom at 10:45 AM on July 29, 2014 [3 favorites]


Puns pretty much always work.
posted by saulgoodman at 4:06 PM on July 30, 2014


omg just got that
posted by psoas at 6:04 AM on July 31, 2014 [2 favorites]


« Older Distillers vs. Bottlers   |   Hail the Traveler Newer »


This thread has been archived and is closed to new comments



Post