171–180 of 373 posts
> Recommendation of profiles that you may like is a solved technical challenge at Tinder level and at mostly any dating app today. It’s hard to take the rest of the article seriously after reading this!
Earlier quoted context omitted.
It tracks, actually. Men tend to pick a sexual orientation and be disgusted by the other sexual orientation[0]. Women are a lot more hetero- (and homo-) flexible. [0] It works both ways! A guy once told me and my wife that we made a cute couple despite how disgusting heterosexuality really was. And I was like... okay, thanks?
Not completely true. Most men are not disgusted by a lesbian couple. I don’t mean just men that fetishize about it. But many are disgusted or at least discomforted by seeing gay men being romantic. Thought experiment: if a wife in a heterosexual marriage told her husband that she previously had relationships with women, I know many men (including me) who would just shrug and move on and not really think about it. On…
We built a dating app and saw these same metrics. Our "twist" was that anyone could be a matchmaker, pairing other people (in addition to our recommendation engine). An interesting thing happened: lot's of people just like using dating apps in a voyeuristic way, with no intention of dating. Tinder eventually launched this feature, but it fell flat. I still think there's merit to human-based recs over "AI" matchmakers…
Have you done any research on the type of users who use dating apps primarily? The article is really good on the product decisions but light on what kind of life and situation these users have.
Our data was almost identical to the article, showing the same imbalances.
Equal parts fascinating and dystopian. I'm even more convinced now that online dating has reached a local optima, but eventually someone is going to find a solution that is less shallow and predatory and blow it out of the water.
I’m more cynical. Dating apps are easy, and entertaining. Finding a good person to spend your life (or a prolonged period) with is hard. I’m not sure there is an external solution to this in the world we live in - it’s like getting fit. People need to suck it up and put in the work if they want results.
This is such a detailed article but it's giving me weird vibes. For instance there are all these drops to near-zero in the histograms at .28, .46, .56 for no clear reason, and the article doesn't even consider that noteworthy. The "Men Like ratio (y) vs ratio (x)" has an inexplicable wall around .33 which I could only explain with some sort of product limitation maybe? But I really wish it was explained what artifact…
Definitely points to some rounding error, aliasing in the data. It would be fixed by making the buckets larger. No reason for the buckets to be that small.
I worked in this space briefly. This is a really good article in line with the conclusion of the OK Cupid! team which were scrubbed from their web site but are in this book: https://en.wikipedia.org/wiki/Dataclysm If you're a guy and you want to depress yourself, find a female friend who is signed to a dating app and see all the likes and messages she gets. When I was in this space, an attractive woman was getting on…
The thing interesting about that blog was — if I remember it correctly — they posted results of initiations by men and women. That suggested there was kind of an optimal strategy for men and women, in that there was kind of mismatching going on in terms of who reached out to who.
IIRC, they were suggesting that it was good for women to reach out to very attractive men, because men in general were not getting contacted, including attractive men, but were responsive to being contacted, and a woman could take advantage of that, even if they were less attractive themselves. Men in contrast would do better to reach out to women who were slightly less attractive, because most of mens' attention collectively was focused on a very small group of very attractive women, and the women who were less attractive, even slightly so, were getting contacted much less often.
Typing this out I think the distributions were very different from in the current article, but also I think they were analyzing different things. In that they were looking at actual initiations, and in this they're looking at likes. Also this article breaks things down a bit more in terms of goals in using the app, which plays into things as well in a big way.
The complexity of it all increases quite a bit very rapidly.
Earlier quoted context omitted.
Another theory: when you swipe and don't get a match, that could be considered a rejection and women are worse at handling rejection (probably due to never having to learn to deal with it). Men, on the other hand, have to learn to accept rejection so little is felt when almost all swipes don't match.
Most people (all sexes and genders) are bad at handling rejection, period. It’s why online dating is popular in general—a rejection over a digital medium is a lot less intimidating and less confrontational than the same in person.
The data here falls in line with the infamous OKcupid study (which got cancelled and taken down because men and women are identical, donchaknow?) The takeaway is that humans best date by meeting people in person through mutual acquaintances. Without the forced direct social interaction, women are only interested in the top 10% of guys, and guys are just aimlessly running at anything that moves regardless of their act…
It does imply a potentially interesting algorithm though: Show all of the women to all of the men. From this you get numbers on which women are the most highly rated. Then, show women a) men who are at the same percentile as them or below and b) the smaller number of men who have already liked them. The result is that 60th percentile women end up primarily liking 60th percentile men and so on, because they're the hig…
showing a 60 percentile woman, men that around 60 percentile is a sure way to drive women away from your app
simplest CRUD.
issue is either we are ugly or dark patterns make dating sites unusable.
Earlier quoted context omitted.
Ugly by most standards is simply not being desired by anyone. You don't have to be an acid victim like two-face to be ugly. I'd say someone is definitely ugly if you swipe on a few hundred average looking people and get zero matches back. That's ugly.
> if you swipe on a few hundred average looking people and get zero matches back. That's ugly. What's ugly is using an anemometer to measure a distance. I mean, using number of matches in a dating app, to measure uglyness. Dating apps are products designed by psychologist and built by engineers to generate frustration and make people pay, not to serve as a measurement stick of the average person's attractiveness. Dat…