Sample Tree

Inspiration

The art of sampling is a tradition in many genres of music such as Hip-Hop and R&B. To sample a song means to use a snippet or musical element from it to create a new song. This practice has been so prominent over the past half century that it's possible to trace networks of sampling, where a song made today could be sampling a song from the 90's, which itself samples a song from the 70's.

What it does

Sample Tree takes a chosen list of influential songs and generates a visual network to demonstrate which artists have sampled it, and where those songs have been sampled as well.
The site acts as a digital

How we built it

We started by scraping song and sampling data from the website Whosampled from a short list of songs which have had a significant musical impact in history. The scraper would then recursively go through those songs which sampled the original and see where they were sampled as well. Then we created an algorithm that groups the songs visually into a web to illustrate the multi-level sampling relationships. Finally we created a website to display the network and show song information for each branch.

Challenges we ran into

There is a limit to the amount of data we could scrape from Whosampled, so we had to limit ourselves to only showing the networks branching from a few songs we thought were important. Without that limitation we could have let the user search for any song and see the relationships it has.

Accomplishments that we're proud of

The site is a very interesting way to visualize how sampling plays into musical history. It's possible to search online for direct relationships from one song to another but currently there does not exist a platform which clearly demonstrates multi-level sampling as a phenomenon. We're also proud of use of several different programming languages and principles to bring the website together.

What's next for Sample Tree

We could partner our website with Whosampled and be among the several websites and applications which use their library of sampling data to show interesting stories in music history. This will allow us to have a much more comprehensive network showing any song that someone might be curious about.