What if you made it say yes to everyone. Then it takes only the ones who matched. Then you could start culling people by with the word 'YOLO' or hashtags or instagram accounts in their bio or some shit. Then start running facial recognition based on manually selected ideals. Then it could send an opener line from a list. And then finally, it could send you an alert if the match replies to your automatic opener, as well as recording all their details.
Make some fake facebook accounts and start running it in different cities etc to collect a bigger data set, you could find your own personalised statistically most successful pick up line, by age group, 'likes' etc.
It uses Eigenfaces to figure out if it should like someone for you, and then has a basic conversation to judge if they are sufficiently interested for Tinderbox to alert you.
From the README: "built at the 2015 Stupid Shit No One Needs & Terrible Ideas Hackathon". I'd say it definitely doesn't need a public API, but for their apps on various platforms, it's probably useful.
I think the OP may be referring to a public API. I can't think of any positive practical uses for the public API as-is. In the hands of a more malicious person, it could be a tool to automate harassing users.
I havent seen a truely positive use of the API either. In fact they've had problems with it.
There was an exploit at one point that allowed you to ask the API "how far from my current position is this person" and it returned a very precise distance based on both user's gps. If you then moved your location (just the argument, no need to actually move), you could triangluate the other users exact location. Tinder fixed that by fuzzing the distance and giving less precision.
There are also apps that auto like everyone. This sort of defeats the purpose of the app. Tinder didnt seem to really mind though. They did eventually limit the number of "likes" each user can have in a day, but added a pay option ot bypass it too.
> There are also apps that auto like everyone. This sort of defeats the purpose of the app.
It's what a lot of people do anyway, though. Why not automate an action if it's the action people want? (Farming/grinding bots in MMOs, for example.) If the service doesn't like it, they can always put a stop to it by changing the incentive structure of their "game."
That is in fact the use case. One popular case was a bot that played guys off other, pretending they were taking to a different (woman) with s different pic.