11–20 of 65 posts
Why 6 years?
It started with an idea to allow users to connect within polygon geofences anywhere on earth. Getting this system to work on Android with all the backend code takes a lot of time. The system itself loads polygon geofences 180° E and W longitude and 90° N and S latitude. And it uses perimeter-based loading system that crosses the antimeridian, equator and north and south poles.
Reminds me of Jodel ( https://jodel.com/ ), an app originally focused on students. It lost quite some activity in the last decade though, gaining fewer users than it loses.
There's so many ways to fake your location data. There's one way that you can't really fake: Send them a secret code on a piece of paper in the US mail to their physical address. NextDoor used to do this at one point.
I worked on a location-based app a few years ago and this was the exact validation method we used (after having learned about it from using NextDoor). It’s incredibly slow and tedious though. We abandoned the app for other reasons but I always wondered how one could continued to manage this approach once network effects kick in and the app really starts to grow.
In an experimental identity system I prototyped as a civic tech project[1], I paired this with scraping a government "voter registration check" form and comparing against, and it was a two-fold guarantee: someone had to either submit false info to the voter registry or intercept mail. In theory it was very cheap to get very high assurances, for only the cost of a postcard API per user
[1]: https://github.com/patcon/id.c4nada.ca?tab=readme-ov-file#ab...
There's so many ways to fake your location data. There's one way that you can't really fake: Send them a secret code on a piece of paper in the US mail to their physical address. NextDoor used to do this at one point.
There's so many ways to fake your location data. There's one way that you can't really fake: Send them a secret code on a piece of paper in the US mail to their physical address. NextDoor used to do this at one point.
There's so many ways to fake your location data. There's one way that you can't really fake: Send them a secret code on a piece of paper in the US mail to their physical address. NextDoor used to do this at one point.
Then once people exploit the app, that doesn't mean they wont add value (e.g. contribute positive content). Maybe they are just a high school kid that wants to talk to his friends in his last town?
Once you have users, then there will be other easy signals to detect: Is the person teleporting? Do they hit rate limits freq? Is their GPS location the exact 'center' of the city? Is there GPS a nice pretty number? Does their GPS location never move?
WTF is wrong with these social apps!?!? Who wants to chat on a tiny screen when they have a computer available. Especially for local apps that function only when you're home.
Instagram had photo filters; Strava had activity stats. What could this have?
No web version? No sale. WTF is wrong with these social apps!?!? Who wants to chat on a tiny screen when they have a computer available. Especially for local apps that function only when you're home.
Let me start by saying that I really like this idea. Obviously social apps like this are faced with a chicken-and-egg dilemma of how to acquire users. I'm no marketer, so I don't have any suggestions on how to solve this one. For myself, I avoid non-free/open-source programs in general, but especially chat apps. I think that especially the programs we rely on to communicate should at least be transparent on the clien…