Peter Bedrosian and Nick Ciereck - Final Project

Proposal:

Peter Bedrosian and Nick Ciereck
For this project, I am thinking about combining either google maps API or the leaflet API with YouTube to make geo-referenced video searches. YouTube now gives users the ability to tag their videos with a specific location or geographic coordinates. I think it would be interesting to create a visual map where a user can click on a specific location to see videos that were tagged there. For example, a user using my site could enter "sunset" and set it to show YouTube videos with keyword "sunset" that were uploaded this week. The basemap would then generate points for every video that YouTube can find that matches those search parameters. The user can then click on a point and a box will pop up, just like in the google maps lab, except this box will have the title of the video, location, date, link to the video, and the actual embedded YouTube video. If this is too broad, I could focus on a specific group of YouTube videos, such as "sports", and then allow the user to select between soccer, basketball, football, and so on.

YouTube has its own API that I plan on implementing with GIS. Here is another link for users interested in combining YouTube with Google. Here is a link for an example that someone else created that is similar to my idea. In this example, the code for the video is "baked" in, and I am hoping to have these points and paths generated based on what the user specifies. Here is another link for a similar example, except the GUI isn't that great and it doesn't allow the user to search for specific videos. According to cnet, the previous example website works by implementing "Google Maps API, Google Feeds API, YouTube Data API, and Google's Location API to find geotagged videos near your location." I am not sure how ambitious of a project this will be, but I think I could deliver something with at least the same quality of this example.


Demo:

Demo Link
Here we have a map of Ohio State University and a marker placed over the football stadium. When the user clicks on the marker, they see a video representing something at that location (in this case, a video of the osu marchingband). Unfortunately, the video doesn't quite work yet but we will fix that. Eventually we will populate all of Ohio State University with markers to videos showing different features at those locations.

Final Link:

Here is the link to our completed final project: final.html

Report:

We decided to change our project to one that resembled the final lab in this class. We still implemented YouTube in our project but without using any YouTube API. The process for gaining access to the API and implementing it in leaflet proved to be much more difficult than we imagined so we scrapped that idea. Instead, we created a database of geolocated YouTube links that corresponded to real world landmarks scattered around The Ohio State University's campus. The idea is that you could either enter an address or search by your current location and find all interesting OSU landmarks surrounding you within a specified radius. We lowered the values assuming that a student would not be interested in landmarks more than walking distances. Then leaflet would generate markers that the user can click on to see a YouTube video of something interesting about that landmark. One of the biggest limitations of this project is that we have to hardcode the YouTube links into a text file. This means that if the link goes down, it will remain down unless fixed by one of us. It would have been much more interesting to actually use the YouTube API to search videos by geolocation and have these coordinates generated into leaflet every time you hit search. This way, the search would be much more dynamic, adding and deleting videos as they come and go. However, I have not seen any examples online where this has been done so it is entirely plausible that the YouTube API can't actually do this.