5 ms·

I once went on a date with someone who did research at OKCupid who told me that they were doing NLP-style analysis of peoples' messages that they sent to each o

by chaps 6mo ago

I once went on a date with someone who did research at OKCupid who told me that they were doing NLP-style analysis of peoples' messages that they sent to each other. Still not really sure what to think of the date itself, but it was a fucked up admission.

m463 6mo ago
makes me wonder if the person you went on a date with cherry-picked you due to your data. (anyone who would post on hacker news is obviously a good catch!)
- toast0 6mo ago
  > anyone who would post on hacker news is obviously a good catch! "the odds are good, but the goods are odd" may apply here
- chaps 6mo ago
  You're funny. I think the "only thing" that would make me cherry-pickable from their data is that I used an autoclicker to give everyone a 5 star... I have mixed feelings about doing that, but I got a couple (surprisingly nice) dates out of it that never went anywhere.
- scottyah 6mo ago
  If only they had the long term data too. It might make for easier discussions on the first date, but maybe there's more to opposites attracting/different roles in a relationship.
probably_wrong 6mo ago
If you remember the old OkCupid blog they used to post interesting articles about online dating. I know their article about whether you should smile on your profile picture was eventually debunked [1], but it was nonetheless nice to have objective, data-based, non-pua advice on how to be successful in online dating. [1] https://blog.photofeeler.com/okcupid-is-wrong-about-smiling-eye-contact-mens-dating-profile-photos/ https://blog.photofeeler.com/okcupid-is-wrong-about-smiling-...
- jacquesm 6mo ago
  You mistook a marketing effort for science.
  - loverboy69 6mo ago
    There was an actual effort at data science going on here before the marketing team took it over in the latter years. See the published book Dataclysm by one of the founders for more of the good stuff.
scottyah 6mo ago
I did like that they shared a lot of hard data with insightful analysis. At the time, there were a lot of narratives about what women wanted and it was refreshing to see them post what was actually working. I remember being skeptical about anything being private online at the time, but I guess that perspective wasn't as pervasive.
crazygringo 6mo ago
They did tons of data analysis across all aspects of profiles, and had a popular blog where they published the results. They were heavily involved in researching what factors more reliably led to not just better matches, but better relationships -- when you disabled your account, they'd ask if it was because you'd met someone through OkC and ask you to pick who, if you were willing to share. I don't think there was anything fucked up about it, as long as it was all anonymized and at scale. Trying to understand what messaging strategies worked better or worse could be a major part of figuring out how to improve matches. Like, one obvious factor could be to match people who send lots of long messages with lots of questions with each other, while a separate set matches people who's messaging style is one sentence at a time. I'm not saying that would necessarily work well, but it's not crazy to research if NLP analysis of messages can produce additional potential compatibility signals. The whole point of OkC back then was to try to develop as many data-based signals as possible to improve matches.
- chaps 6mo ago
  You realize that you're responding in a thread about OkCupid deceiving users and sharing data with 3rd parties right? 34. In response to this request, Humor Rainbow gave the Data Recipient access to nearly three million OkCupid user photos. Humor Rainbow’s President and Chief Technology Officer were directly involved in facilitating the data transfer. 35. In addition to user photos, Humor Rainbow shared other personal data with the Data Recipient, including each user’s demographic and location information.
  - crazygringo 6mo ago
    You realize I was responding to a comment that wasn't about that? A company running NLP on its own user data doesn't have anything to do with third parties.
    - chaps 6mo ago
      > I don't think there was anything fucked up about it, as long as it was all anonymized and at scale I'm confused why "as long as" carries so much weight here considering the article that started this discussion. You seem to trust that they stopped their privacy fuckups with third parties. I don't know where your trust comes from.
loverboy69 6mo ago
As a former employee, we did do NLP on messages but exclusively to find and suppress spammy/scammy/offensive messages. That system is dead now tho.
smcin 6mo ago
No it totally wasn't a fucked up admission, it was actually a useful and pro-user measure (all this good stuff was before the 2011 acquisition). Christian Rudder's OKTrends blog (and Sam Yagan's presentation at their acquisition celebration) even spelled out the reason why: some women on OKC (or, more rarely, men) would acquire the "Replies rarely"/red color on their profile, for almost never (<10%) replying to initial messages, which was generally considered to be undesirable behavior, even in the negative (there is value in a negative message: "Thanks but I'm not interested due to age/location/other factor". And also OKC could then measure whether users' stated preferences mismatched preferences inferred from which set of users they message e.g. people who say they're looking for 30-55 for LTR but tend to message people 21-35 for short-term). And before anyone points out that younger more attractive female profiles would get more initial messages than males (up to 200:1 more), OKC used to allow you to set filters on the other user's age/distance/other criteria, so you could automatically filter those out. Also, factor in the usual caveats that many users on dating sites tend to lie about their age/weight/height/location/status/etc. Anyway, to avoid getting labeled the dread "Replies rarely", some (mostly female) users got in the habit of sending one-liner responses that were ambiguous/non-committal/cryptic/negging. And then not responding further (but without unmatching, which only took a single click). This was making their profile look less undesirable but generating pointless message traffic and reducing the overall utility of the platform at actually attempting to match people (for compatibility, not just initial attraction). Hence, OKC tried to actually measure initial exchanges to figure out which ones led to genuine back-and-forth conversations of 3+ messages (which is an ok proxy for inferring a match, certainly a better proxy than just counting initial messages/likes/votes on photos). Yagan jokingly referred to this as "Every Monday morning, we ask 'How many three-ways did we set up over the weekend?'"). (PS, Rudder and Yagan both stressed that users' names/identities/ identifying characteristics were kept out of the analysis.) After the 2011 IAC acquisition, most of this platform quality control (and looking for constructive insights) went out the window pretty quickly and the three cofounders moved to OKCupid Labs. But it was good for the brief while it lasted. By 2013 a chainsaw had been taken to most of OKC's unique features, esp. for free users.
- chaps 6mo ago
  Hey, my point is that it's fucked up that I went on a date with someone who admitted their job was to read other people's messages. If you don't think that's fucked up, then we simply have a difference in opinion. I don't know what the rest of your post is about.

by chaps 6mo ago

m463 6mo ago

makes me wonder if the person you went on a date with cherry-picked you due to your data. (anyone who would post on hacker news is obviously a good catch!)

toast0 6mo ago
> anyone who would post on hacker news is obviously a good catch! "the odds are good, but the goods are odd" may apply here
chaps 6mo ago
You're funny. I think the "only thing" that would make me cherry-pickable from their data is that I used an autoclicker to give everyone a 5 star... I have mixed feelings about doing that, but I got a couple (surprisingly nice) dates out of it that never went anywhere.
scottyah 6mo ago
If only they had the long term data too. It might make for easier discussions on the first date, but maybe there's more to opposites attracting/different roles in a relationship.

probably_wrong 6mo ago

If you remember the old OkCupid blog they used to post interesting articles about online dating. I know their article about whether you should smile on your profile picture was eventually debunked [1], but it was nonetheless nice to have objective, data-based, non-pua advice on how to be successful in online dating. [1] https://blog.photofeeler.com/okcupid-is-wrong-about-smiling-eye-contact-mens-dating-profile-photos/ https://blog.photofeeler.com/okcupid-is-wrong-about-smiling-...

jacquesm 6mo ago
You mistook a marketing effort for science.
- loverboy69 6mo ago
  There was an actual effort at data science going on here before the marketing team took it over in the latter years. See the published book Dataclysm by one of the founders for more of the good stuff.

scottyah 6mo ago

I did like that they shared a lot of hard data with insightful analysis. At the time, there were a lot of narratives about what women wanted and it was refreshing to see them post what was actually working. I remember being skeptical about anything being private online at the time, but I guess that perspective wasn't as pervasive.

crazygringo 6mo ago

They did tons of data analysis across all aspects of profiles, and had a popular blog where they published the results. They were heavily involved in researching what factors more reliably led to not just better matches, but better relationships -- when you disabled your account, they'd ask if it was because you'd met someone through OkC and ask you to pick who, if you were willing to share. I don't think there was anything fucked up about it, as long as it was all anonymized and at scale. Trying to understand what messaging strategies worked better or worse could be a major part of figuring out how to improve matches. Like, one obvious factor could be to match people who send lots of long messages with lots of questions with each other, while a separate set matches people who's messaging style is one sentence at a time. I'm not saying that would necessarily work well, but it's not crazy to research if NLP analysis of messages can produce additional potential compatibility signals. The whole point of OkC back then was to try to develop as many data-based signals as possible to improve matches.

chaps 6mo ago
You realize that you're responding in a thread about OkCupid deceiving users and sharing data with 3rd parties right? 34. In response to this request, Humor Rainbow gave the Data Recipient access to nearly three million OkCupid user photos. Humor Rainbow’s President and Chief Technology Officer were directly involved in facilitating the data transfer. 35. In addition to user photos, Humor Rainbow shared other personal data with the Data Recipient, including each user’s demographic and location information.
- crazygringo 6mo ago
  You realize I was responding to a comment that wasn't about that? A company running NLP on its own user data doesn't have anything to do with third parties.
  - chaps 6mo ago
    > I don't think there was anything fucked up about it, as long as it was all anonymized and at scale I'm confused why "as long as" carries so much weight here considering the article that started this discussion. You seem to trust that they stopped their privacy fuckups with third parties. I don't know where your trust comes from.

loverboy69 6mo ago

As a former employee, we did do NLP on messages but exclusively to find and suppress spammy/scammy/offensive messages. That system is dead now tho.

smcin 6mo ago

No it totally wasn't a fucked up admission, it was actually a useful and pro-user measure (all this good stuff was before the 2011 acquisition). Christian Rudder's OKTrends blog (and Sam Yagan's presentation at their acquisition celebration) even spelled out the reason why: some women on OKC (or, more rarely, men) would acquire the "Replies rarely"/red color on their profile, for almost never (<10%) replying to initial messages, which was generally considered to be undesirable behavior, even in the negative (there is value in a negative message: "Thanks but I'm not interested due to age/location/other factor". And also OKC could then measure whether users' stated preferences mismatched preferences inferred from which set of users they message e.g. people who say they're looking for 30-55 for LTR but tend to message people 21-35 for short-term). And before anyone points out that younger more attractive female profiles would get more initial messages than males (up to 200:1 more), OKC used to allow you to set filters on the other user's age/distance/other criteria, so you could automatically filter those out. Also, factor in the usual caveats that many users on dating sites tend to lie about their age/weight/height/location/status/etc. Anyway, to avoid getting labeled the dread "Replies rarely", some (mostly female) users got in the habit of sending one-liner responses that were ambiguous/non-committal/cryptic/negging. And then not responding further (but without unmatching, which only took a single click). This was making their profile look less undesirable but generating pointless message traffic and reducing the overall utility of the platform at actually attempting to match people (for compatibility, not just initial attraction). Hence, OKC tried to actually measure initial exchanges to figure out which ones led to genuine back-and-forth conversations of 3+ messages (which is an ok proxy for inferring a match, certainly a better proxy than just counting initial messages/likes/votes on photos). Yagan jokingly referred to this as "Every Monday morning, we ask 'How many three-ways did we set up over the weekend?'"). (PS, Rudder and Yagan both stressed that users' names/identities/ identifying characteristics were kept out of the analysis.) After the 2011 IAC acquisition, most of this platform quality control (and looking for constructive insights) went out the window pretty quickly and the three cofounders moved to OKCupid Labs. But it was good for the brief while it lasted. By 2013 a chainsaw had been taken to most of OKC's unique features, esp. for free users.

chaps 6mo ago
Hey, my point is that it's fucked up that I went on a date with someone who admitted their job was to read other people's messages. If you don't think that's fucked up, then we simply have a difference in opinion. I don't know what the rest of your post is about.