aigeo
here's the pitch
Banner ads are crap. Nobody reads them on purpose, and they aren't relevant to users, and yadda yadda they suck. Map based ads are only heavily used in Waze (owned by google). They are super relevant to users, not intrusive, native to the content they are looking at, and all around better! The goal of this project is to figure out how to make a doubleclick type ad exchange where advertisers can buy campaigns based around locations, and applications or web services with map views can display the ads inline for revenue. Users see the content they are looking for, and get ads highly targeted at something they've explicitly expressed interest in as they express that interest. If you look at a map view, it tends to be because that's where you are, where you're going, or somewhere you're interested in.
what works
I've got a domain with github pages hosting the client content (an angular web app). I spun up a linode linux box to run a json server, that can hold some basic information. Independently I got Bootstrap, Angular, and a sample map from mapbox to display. which leads to
whats horribly broken
I wanted to be able to display a marker on the mapbox app from a geo json object from the json server, and dynamically inject that using Angular. That did not work as I hoped. When you put the map inside an angular directive breaks everything. That's what I spent the whole time trying to fix, and if I had finished that the next step would be an input for someone going to the site to be able to input a lat-long and information that would be inserted preferably with 2 way data binding, or on reload if dataservices were needed.
what I learned
New ways that angular and booststrap interact, and most importantly geo-json. I didn't know that that was a format that exited. That makes the entire idea much easier. the map-box api is powerful (and finicky) and i need to spend a good soul to soul with it to understand what it wants.
Log in or sign up for Devpost to join the conversation.