Dating apps have filter bubble issue
It only took I skimmed through it for three days before it showed up. I froze, pointing my thumb at the cross. I flipped through his photos and clues, looking at what he had changed since the first time I saw him.
The first photo was the same: he is holding a climbing rope somewhere far away, curly hair sticking out from under a baseball cap. His simple pleasures continued to be “mountain roads, forests and carefree mornings.” He added a photo of him standing shirtless at the foot of a cliff.
The twist of the knife was a note from Hij at the top: “Most Compatible: We think you two should meet.”
I had Hinge’s version of the 90s-era Microsoft Word paperclip assistant, Clippy, squeaking at me in my head, “You seem to like camping and concerts, would you like to be in touch with that other person next to you who loves camping” . and concerts?
The app couldn’t have known that two of its users were taxiing down the dating runway but never took off – a classic. situationalityas kids call it these days. The algorithm just saw a 31-year-old local resident working outdoors in biotech and connected the dots with a 30-year-old outdoors worker a few miles away working in science media. And just like Microsoft users in 1997, I hated it. I wanted to crush a computer program that thought it knew what I wanted, whether it was a letter or my perfect match.
Dating apps promise to show you all the romantic options in your city, but behind the scenes, the algorithm cultivates a very specific, limited, at least somewhat different dating landscape for each user.
First big dating site was Match.com, which was founded in 1995 and was followed by eHarmony and OkCupid in the early 2000s. These advertised sites their surveys, compatibility scores, and science-based approaches to bringing couples together as the best way to find lasting love. These interoperability-based approaches to online dating dominated until 2009 when the gay dating app Grindr entered the scene and changed online dating forever.
Grindr, like a mobile app, organized romantic options not by compatibility, but by distance — the main person was closest to you. This is still the default on Grindr today. When Tinder took the idea of Grindr to the real world in 2012, it duplicated that notion that it was based on distance, if not quite in its code.
“When you think about platforms like OKCupid and eHarmony, it would be hard to use them and not know the algorithm exists because it is at the forefront of what they do,” said Liesel Sharabi, a scientist at Arizona State University. studying dating apps. “But when I talk to people who use Tinder, they don’t always know there’s an algorithm. A lot of people think it’s just showing the people around them and it’s a lot harder.”
In 2016 Tinder confirmed he used the Elo score traditionally used to rank chess players to rank users by desirability and match them accordingly. The media storm was fast and strong; by 2019 Tinder was declaring that it no longer uses an Elo score, although it probably still uses some, if not many, algorithms. Since then, most dating app companies have been using a “black box” approach and not publicly talking about what factors influence their algorithms.