9 ms·

OkCupid’s Unblushing Analyst of Attraction

118 points by boh 12y ago

keerthiko 12y ago
I have a lot of respect for the people behind OkCupid. They tread a lot of grey areas that many online entities are scared to. On the whole, I also think usage of willfully submitted and gathered user data for things other than advertising is amazing. Too many companies have invested too many resources into data analytics purely for increasing revenue from advertising. I think if companies can find ways to gain better social understanding and from that provide better value to end users (rather than to advertisers as in the advertising model) in a way that gets them revenue, it will be amazing. I am all for social media companies doing research from their user data, in the hope that we can move on from the ad-driven web.
- Theodores 12y ago
  > I have a lot of respect for the people behind OkCupid. Those people being match.com - the same people that operate the 'tinder' app. Much like how the company that sells the eco-friendly smoothies is probably Coca Cola. Ultimately match.com want to own online dating. Since there is always going to be a niche for free dating sites (as well as the subscription model), they have to have the best 'free'/freemium dating site going. The marketing/PR consists of 'insights' rather than TV adverts or social networking thingies. The site is just one part of their strategy to own online dating. People on the site might not pay, however those adverts probably bring in a lot of revenue - see Plenty of Fish story. Returning to the eco-friendly smoothies - there are no saints at the helm of the company determined to make a better world, they are just doing what is needed to bring in the revenue for the parent company and making sure nobody else is in that space. Another analogy - big software company making sure their software is reasonably easy to pirate so people don't use any other software, free or paid for. Clearly OKCupid was more valuable to match.com than anyone else hence they bought it. However the idea is no more respect worthy than the operations of the eco-friendly smoothie company owned by a fizzy drinks giant mega corporation.
  - keerthiko 12y ago
    I see what you're saying. Sure, by some definitions the people behind match.com are behind OkCupid, but I'm pretty sure they're quite hands-off from okc's operations and general demeanor. Of course, they are probably involved in their profit strategies as you say, and that's hard to avoid. I still have respect for whoever came up with the eco-friendly smoothies and was able to pose enough of a threat to Coca Cola, but instead of getting crushed by them is able to harness the giant's resources to make more eco-friendly smoothies. It's a lot better than folding under the evil. There are no saints at the helm of any for-profit organizations, or they'd get destroyed. I'm talking within that (very realistic and tolerable) space, they do some creative things [as a past okcupid user and as someone interested in their data] that provide interesting value to the customer that are hard to find elsewhere.
kevinthew 12y ago
Might be cool stuff they're doing but it doesn't make it any less unethical no matter how much they try to whitewash it. This is slippery slope stuff
- tormeh 12y ago
  They're not slippery-sloping anything. They're saying your data is theirs, period.
- Mikeb85 12y ago
  Using aggregate data (ie. not any individual's personal data) about interactions to study societal behaviour is unethical? On a website that you willingly sign up for, no doubt with disclaimers about how they're allowed to use your data? As far as slippery slopes go, Facebook crossed that line quite a while ago...
  - burgers 12y ago
    When you publicly disclose the results, without any way for the data to be peer reviewed? Absolutely.
cbhl 12y ago
I really enjoyed reading OkTrends posts, but having near radio silence for three years (apart from one post last July) followed by a full-on PR blitz for a new book (comes out on Tuesday) makes me a little sad. This piece almost makes it sound as if Rudder has been blogging based on OkCupid results this whole time... and if you go to the OkTrends site, you see huge inline placement for Rudder's new book.
- Semaphor 12y ago
  > but having near radio silence for three years Coincided with their acquisition by match.com. Even though they said it didn't have anything to do with that (just like removing their critique about sites like match.com was voluntary). Well, I hope they get back to writing those posts.
- beloch 12y ago
  OKTrends went dormant when OKCupid was sold to the company that operates match.com. I don't know what Rudder's involvement with OKCupid was after that, but he was no longer one of the owners. He might have still worked there, but kept his blog silent under pressure from management.
  - nether 12y ago
    Pressure from management? Sam Yagan is still the site's CEO. Also the pressure must have failed because he's releasing a whole book about his findings now, not just blog posts. You people are ridiculous.
    - Natsu 12y ago
      Most of us have not forgotten the "why you should never pay for a dating site" analysis that vanished, nor will we any time soon. EDIT: Might as well link to it while I am at it - http://machinesentience.com/why_you_should_never_pay_for_online_dating.html http://machinesentience.com/why_you_should_never_pay_for_onl... Also, I know the explanation they gave. I don't give a damn whether anyone was pressured into taking it down and they never bothered to actually explain what was wrong with the original analysis, identify the 'public sources' used, or to do anything but try to bury the story with a cheap excuse and hope we'd all forget.
norswap 12y ago
The blog was dormant for 3 years, apparently they just posted a new post... Of course I unsubscribed last month :D
templeos2 12y ago
Know what's fucked-up? I used to be a long-haired handsome kid. Not that pretty boyed. Now? Men should wear buzcuts and facial hair. If taking an avatar photo for OSDEV? Don't seduce the camera, homo! It's dudes, dumb fuck.
kachnuv_ocasek 12y ago
Off-topic, but why am I getting told to log in on the NYT website?
burgers 12y ago
The spookiest thing is not the data or the experiments, but the off the cuff conclusions. For example: > As a group, for instance, Latino men rated Latinas as 13 percent more attractive than the average for the site, while they rated African-American women 25 percent less attractive. That is an insane generalization going on. What metric are we using to determine this? Is it possible that people who tend to participate in the rating of the looks of potential mates are more inclined to align with race? Not purely Latino men. Which is an enormous initial generalization to be making at the outset. > Witness the actions of 35-year-old heterosexual men on OkCupid. These men typically search for women between the ages of 24 and 40, Mr. Rudder reports, yet in practice they rarely contact anyone over 29. Again on this one, was age the only possible metric that caused the under 30 to be contacted more by 35 year old men? There isn't anything else that might be different about an under 30 profile that causes more communication to occur? This is the actual scary stuff to be publicly releasing as real science. Just as the general public is ill informed about the experiments going on, they are also not aware of what metrics are used to determine these results. In my experience, many of these metrics are not as concrete as the appear and full of pushing the data to fit a narrative etc. Combine that with the fact that most of this data is proprietary and private with no way to be peer reviewed. Dangerous stuff.
- fiatmoney 12y ago
  Why is it "dangerous"?
  - beachstartup 12y ago
    because it doesn't subscribe to his fantasyland world where all people are equal in the sexual marketplace.
    - burgers 12y ago
      This has nothing to do with my "fantasyland" or wanting everyone to be equal. Its about making racial and other generalizations without publishing the full data with it. Racial and gender generalizations are dangerous things for a society. If you are going to go down the road of pointing out these things, you need to do it with extra caution and transparency.
philip1209 12y ago
I'm always confused by the phrase "African-American." OkCupid is international. Why does somebody with African ethnicity have to be classified as "African-American"?
- iamdave 12y ago
  It's a question I've been asking, as a black man for a very, very, very long time. Couldn't count on two hands, even a third if I had one the number of different answers I've gotten, and not a single one of them-no matter the rationality or validity behind the history of it, has yet to answer why we still use it. That said, "Person of Color" seems to be the new trendy thing and that baffles me even more (well, to be succinctly snarky about it: other people's reactions to that phrase are what baffle me. That's an entirely different conundrum not fit for HackerNews, though.)
  - vacri 12y ago
    I remember seeing an article talking about a black British model, born and raised in the UK, in a French fashion shoot, described as African-American. Her own country would refer to her as British, not African-British... Here in Australia we're getting infected with this compound term, to reassert that heritage doesn't mean non-citizen. It's really annoying, because I grew up in a suburb of Greeks and Italians, who were also Australians... and now (from some corners) it's Greek-Australian and Italian-Australian. It wasn't perfect - the stereotypical Australian is still seen as a WASP - but this compound noun system isn't the answer to fixing the problem.
    - iamdave 12y ago
      Agreed. John McWhorter brought up a lot of decent and salient points about the words we use to describe individuals, and how simply substituting one for another doesn't really solve the problem (http://time.com/2369/richard-sherman-thug-n-word/ http://time.com/2369/richard-sherman-thug-n-word/). That can be said about a lot of social ailments in swiftly moving, developed civilizations, if I must offer a caveat. There's a lot of bluster that goes on among certain "progressive" crowds that ultimately sums the solutions up to policing of rhetoric; and while I have to admit they're somewhat onto something with the notions that "language informs impressions", a lot of their approaches seem deeply misguided if only because there's a vacuum of nuance. I personally don't have the answers to how to remediate the baggage that comes with cultural nomenclature... I do however believe rather sternly, that demanding a change of vernacular is not a silver bullet.
paulhauggis 12y ago
I'm still not sure why this article is equating preference with racism. If I'm attracted to white women and I'm white, it has nothing to do with racism.
- Chinjut 12y ago
  What does it have to do with?
  - mpweiher 12y ago
    Preference. See Wikipedia. "Racism consists of both prejudice and discrimination based in social perceptions of biological differences between peoples" ... "In sociology and psychology, some definitions only include consciously malignant forms of discrimination.[4][5] While some consider any assumption that a person's behavior is tied to their racial categorization is inherently racist, regardless of whether the action is intentionally harmful or pejorative, because stereotyping necessarily subordinates individual identity to group identity" So if you just prefer one group, that's not racist, as far as I can tell. If you make claims about a group's behavior based on race, that's racist. http://en.wikipedia.org/wiki/Racism http://en.wikipedia.org/wiki/Racism
    - corey 12y ago
      "because stereotyping necessarily subordinates individual identity to group identity" No it doesn't. Many stereotypes are useful when you have very little information about a person. You're only subordinating individual identity to group identity if you refuse to update your beliefs about a person after receiving new evidence about him.
    - burgers 12y ago
      So I guess the KKK isn't racist. Interesting... > "We're not racists," Chambers told CNN Monday. "We just want to be with white people. If that's a crime, then I don't know. It's all right to be black and Latino and proud, but you can't be white and proud. I don't understand it." http://inamerica.blogs.cnn.com/2012/06/12/a-kinder-gentler-ku-klux-klan-we-do-not-hate-anyone-imperial-wizard-says/ http://inamerica.blogs.cnn.com/2012/06/12/a-kinder-gentler-k...
deciplex 12y ago
This is really annoying (emphasis mine): >Mr. Rudder is particularly interested in the divide between the mates people claim they want and their actual online pursuits. Witness the actions of 35-year-old heterosexual men on OkCupid. These men typically search for women between the ages of 24 and 40, Mr. Rudder reports, yet in practice they rarely contact anyone over 29. >“I see this as a statement of what men imagine they’re supposed to desire,” he writes in the book, “versus what they actually do.” For a guy talking up all the awesome science he thinks he can do, this is rather unscientific. I'm sure he realizes there are other explanations for this trend (older women are more likely to be divorced or have kids, for example), but by promoting his own narrative he muddies the waters. For what it's worth I don't think his interpretation of the data is correct in this case. A lot of other stuff on the OKCupid blog, e.g. claims of racism because black women receive fewer first messages than women of other races, fell into the same trap. The blog was very interesting to read, to be sure, but I couldn't take it seriously because there was so much pseudoscience and just-so stories, even if it was backed up with mountains of data and nifty charts.
- colmvp 12y ago
  > A lot of other stuff on the OKCupid blog, e.g. claims of racism because black women receive fewer first messages than women of other races, fell into the same trap. I don't see how that was faulty. It's pretty clear that Black women are involved in interracial marriage at a significantly lower rate than women of other races, or even black men. Then if you look at response rate by race (a previous OKTrends post), black women generally have a lower response rate from men of all races compared to women of other races. Data supports the idea that black women are generally not favored.

118 points by boh 12y ago

I have a lot of respect for the people behind OkCupid. They tread a lot of grey areas that many online entities are scared to. On the whole, I also think usage of willfully submitted and gathered user data for things other than advertising is amazing. Too many companies have invested too many resources into data analytics purely for increasing revenue from advertising. I think if companies can find ways to gain better social understanding and from that provide better value to end users (rather than to advertisers as in the advertising model) in a way that gets them revenue, it will be amazing. I am all for social media companies doing research from their user data, in the hope that we can move on from the ad-driven web.

Theodores 12y ago
> I have a lot of respect for the people behind OkCupid. Those people being match.com - the same people that operate the 'tinder' app. Much like how the company that sells the eco-friendly smoothies is probably Coca Cola. Ultimately match.com want to own online dating. Since there is always going to be a niche for free dating sites (as well as the subscription model), they have to have the best 'free'/freemium dating site going. The marketing/PR consists of 'insights' rather than TV adverts or social networking thingies. The site is just one part of their strategy to own online dating. People on the site might not pay, however those adverts probably bring in a lot of revenue - see Plenty of Fish story. Returning to the eco-friendly smoothies - there are no saints at the helm of the company determined to make a better world, they are just doing what is needed to bring in the revenue for the parent company and making sure nobody else is in that space. Another analogy - big software company making sure their software is reasonably easy to pirate so people don't use any other software, free or paid for. Clearly OKCupid was more valuable to match.com than anyone else hence they bought it. However the idea is no more respect worthy than the operations of the eco-friendly smoothie company owned by a fizzy drinks giant mega corporation.
- keerthiko 12y ago
  I see what you're saying. Sure, by some definitions the people behind match.com are behind OkCupid, but I'm pretty sure they're quite hands-off from okc's operations and general demeanor. Of course, they are probably involved in their profit strategies as you say, and that's hard to avoid. I still have respect for whoever came up with the eco-friendly smoothies and was able to pose enough of a threat to Coca Cola, but instead of getting crushed by them is able to harness the giant's resources to make more eco-friendly smoothies. It's a lot better than folding under the evil. There are no saints at the helm of any for-profit organizations, or they'd get destroyed. I'm talking within that (very realistic and tolerable) space, they do some creative things [as a past okcupid user and as someone interested in their data] that provide interesting value to the customer that are hard to find elsewhere.

kevinthew 12y ago

Might be cool stuff they're doing but it doesn't make it any less unethical no matter how much they try to whitewash it. This is slippery slope stuff

tormeh 12y ago
They're not slippery-sloping anything. They're saying your data is theirs, period.
Mikeb85 12y ago
Using aggregate data (ie. not any individual's personal data) about interactions to study societal behaviour is unethical? On a website that you willingly sign up for, no doubt with disclaimers about how they're allowed to use your data? As far as slippery slopes go, Facebook crossed that line quite a while ago...
- burgers 12y ago
  When you publicly disclose the results, without any way for the data to be peer reviewed? Absolutely.

cbhl 12y ago

I really enjoyed reading OkTrends posts, but having near radio silence for three years (apart from one post last July) followed by a full-on PR blitz for a new book (comes out on Tuesday) makes me a little sad. This piece almost makes it sound as if Rudder has been blogging based on OkCupid results this whole time... and if you go to the OkTrends site, you see huge inline placement for Rudder's new book.

Semaphor 12y ago
> but having near radio silence for three years Coincided with their acquisition by match.com. Even though they said it didn't have anything to do with that (just like removing their critique about sites like match.com was voluntary). Well, I hope they get back to writing those posts.
beloch 12y ago
OKTrends went dormant when OKCupid was sold to the company that operates match.com. I don't know what Rudder's involvement with OKCupid was after that, but he was no longer one of the owners. He might have still worked there, but kept his blog silent under pressure from management.
- nether 12y ago
  Pressure from management? Sam Yagan is still the site's CEO. Also the pressure must have failed because he's releasing a whole book about his findings now, not just blog posts. You people are ridiculous.
  - Natsu 12y ago
    Most of us have not forgotten the "why you should never pay for a dating site" analysis that vanished, nor will we any time soon. EDIT: Might as well link to it while I am at it - http://machinesentience.com/why_you_should_never_pay_for_online_dating.html http://machinesentience.com/why_you_should_never_pay_for_onl... Also, I know the explanation they gave. I don't give a damn whether anyone was pressured into taking it down and they never bothered to actually explain what was wrong with the original analysis, identify the 'public sources' used, or to do anything but try to bury the story with a cheap excuse and hope we'd all forget.

norswap 12y ago

The blog was dormant for 3 years, apparently they just posted a new post... Of course I unsubscribed last month :D

templeos2 12y ago

Know what's fucked-up? I used to be a long-haired handsome kid. Not that pretty boyed. Now? Men should wear buzcuts and facial hair. If taking an avatar photo for OSDEV? Don't seduce the camera, homo! It's dudes, dumb fuck.

kachnuv_ocasek 12y ago

Off-topic, but why am I getting told to log in on the NYT website?

burgers 12y ago

The spookiest thing is not the data or the experiments, but the off the cuff conclusions. For example: > As a group, for instance, Latino men rated Latinas as 13 percent more attractive than the average for the site, while they rated African-American women 25 percent less attractive. That is an insane generalization going on. What metric are we using to determine this? Is it possible that people who tend to participate in the rating of the looks of potential mates are more inclined to align with race? Not purely Latino men. Which is an enormous initial generalization to be making at the outset. > Witness the actions of 35-year-old heterosexual men on OkCupid. These men typically search for women between the ages of 24 and 40, Mr. Rudder reports, yet in practice they rarely contact anyone over 29. Again on this one, was age the only possible metric that caused the under 30 to be contacted more by 35 year old men? There isn't anything else that might be different about an under 30 profile that causes more communication to occur? This is the actual scary stuff to be publicly releasing as real science. Just as the general public is ill informed about the experiments going on, they are also not aware of what metrics are used to determine these results. In my experience, many of these metrics are not as concrete as the appear and full of pushing the data to fit a narrative etc. Combine that with the fact that most of this data is proprietary and private with no way to be peer reviewed. Dangerous stuff.

fiatmoney 12y ago
Why is it "dangerous"?
- beachstartup 12y ago
  because it doesn't subscribe to his fantasyland world where all people are equal in the sexual marketplace.
  - burgers 12y ago
    This has nothing to do with my "fantasyland" or wanting everyone to be equal. Its about making racial and other generalizations without publishing the full data with it. Racial and gender generalizations are dangerous things for a society. If you are going to go down the road of pointing out these things, you need to do it with extra caution and transparency.

philip1209 12y ago

I'm always confused by the phrase "African-American." OkCupid is international. Why does somebody with African ethnicity have to be classified as "African-American"?

iamdave 12y ago
It's a question I've been asking, as a black man for a very, very, very long time. Couldn't count on two hands, even a third if I had one the number of different answers I've gotten, and not a single one of them-no matter the rationality or validity behind the history of it, has yet to answer why we still use it. That said, "Person of Color" seems to be the new trendy thing and that baffles me even more (well, to be succinctly snarky about it: other people's reactions to that phrase are what baffle me. That's an entirely different conundrum not fit for HackerNews, though.)
- vacri 12y ago
  I remember seeing an article talking about a black British model, born and raised in the UK, in a French fashion shoot, described as African-American. Her own country would refer to her as British, not African-British... Here in Australia we're getting infected with this compound term, to reassert that heritage doesn't mean non-citizen. It's really annoying, because I grew up in a suburb of Greeks and Italians, who were also Australians... and now (from some corners) it's Greek-Australian and Italian-Australian. It wasn't perfect - the stereotypical Australian is still seen as a WASP - but this compound noun system isn't the answer to fixing the problem.
  - iamdave 12y ago
    Agreed. John McWhorter brought up a lot of decent and salient points about the words we use to describe individuals, and how simply substituting one for another doesn't really solve the problem (http://time.com/2369/richard-sherman-thug-n-word/ http://time.com/2369/richard-sherman-thug-n-word/). That can be said about a lot of social ailments in swiftly moving, developed civilizations, if I must offer a caveat. There's a lot of bluster that goes on among certain "progressive" crowds that ultimately sums the solutions up to policing of rhetoric; and while I have to admit they're somewhat onto something with the notions that "language informs impressions", a lot of their approaches seem deeply misguided if only because there's a vacuum of nuance. I personally don't have the answers to how to remediate the baggage that comes with cultural nomenclature... I do however believe rather sternly, that demanding a change of vernacular is not a silver bullet.

paulhauggis 12y ago

I'm still not sure why this article is equating preference with racism. If I'm attracted to white women and I'm white, it has nothing to do with racism.

Chinjut 12y ago
What does it have to do with?
- mpweiher 12y ago
  Preference. See Wikipedia. "Racism consists of both prejudice and discrimination based in social perceptions of biological differences between peoples" ... "In sociology and psychology, some definitions only include consciously malignant forms of discrimination.[4][5] While some consider any assumption that a person's behavior is tied to their racial categorization is inherently racist, regardless of whether the action is intentionally harmful or pejorative, because stereotyping necessarily subordinates individual identity to group identity" So if you just prefer one group, that's not racist, as far as I can tell. If you make claims about a group's behavior based on race, that's racist. http://en.wikipedia.org/wiki/Racism http://en.wikipedia.org/wiki/Racism
  - corey 12y ago
    "because stereotyping necessarily subordinates individual identity to group identity" No it doesn't. Many stereotypes are useful when you have very little information about a person. You're only subordinating individual identity to group identity if you refuse to update your beliefs about a person after receiving new evidence about him.
  - burgers 12y ago
    So I guess the KKK isn't racist. Interesting... > "We're not racists," Chambers told CNN Monday. "We just want to be with white people. If that's a crime, then I don't know. It's all right to be black and Latino and proud, but you can't be white and proud. I don't understand it." http://inamerica.blogs.cnn.com/2012/06/12/a-kinder-gentler-ku-klux-klan-we-do-not-hate-anyone-imperial-wizard-says/ http://inamerica.blogs.cnn.com/2012/06/12/a-kinder-gentler-k...

deciplex 12y ago

This is really annoying (emphasis mine): >Mr. Rudder is particularly interested in the divide between the mates people claim they want and their actual online pursuits. Witness the actions of 35-year-old heterosexual men on OkCupid. These men typically search for women between the ages of 24 and 40, Mr. Rudder reports, yet in practice they rarely contact anyone over 29. >“I see this as a statement of what men imagine they’re supposed to desire,” he writes in the book, “versus what they actually do.” For a guy talking up all the awesome science he thinks he can do, this is rather unscientific. I'm sure he realizes there are other explanations for this trend (older women are more likely to be divorced or have kids, for example), but by promoting his own narrative he muddies the waters. For what it's worth I don't think his interpretation of the data is correct in this case. A lot of other stuff on the OKCupid blog, e.g. claims of racism because black women receive fewer first messages than women of other races, fell into the same trap. The blog was very interesting to read, to be sure, but I couldn't take it seriously because there was so much pseudoscience and just-so stories, even if it was backed up with mountains of data and nifty charts.

colmvp 12y ago
> A lot of other stuff on the OKCupid blog, e.g. claims of racism because black women receive fewer first messages than women of other races, fell into the same trap. I don't see how that was faulty. It's pretty clear that Black women are involved in interracial marriage at a significantly lower rate than women of other races, or even black men. Then if you look at response rate by race (a previous OKTrends post), black women generally have a lower response rate from men of all races compared to women of other races. Data supports the idea that black women are generally not favored.