A visualization for FE3 Assessment 1 by @danoszz. This documentation will show my process and thoughts.
As the workflow guided through the process; a visualization always will start with looking at the raw data. I chose the Population without indoor toilet dataset in particular because of a few reasons:
- In the past I had some problems with JSON formats for visualizations, good time to learn.
- Most data points from each dataset: year, percentage and country.
- π©
I came up with an initial idea of creating a scatterplot in the shape of an toilet called Splatterpot.
Since the idea is a bit out of scope for the project and above my own expectations it is fair to say the outcome would never match 100% with the initial idea. I've learned that if I focus to much on the style, concept en outcome that the code will be the weak chain in the end-product. To match the learning objectives the visualization is focused on a working D3.js code instead of a laughable, not working, toilet with π©. So therefore styling is out of scope (but not out of mind).
But.. I failed to deliver it before the deadline because I was stuck at one point for 5 hours:
- The 'years' (2005 - 201^) are stored in JSON as same level as 'country' and 'code' and as a string. To get them on the Y-axis was impossible.
One and a half hour before the deadline I realized I had to make a big change to still hand in a proper working code. So I did. It needed to be fast and solid. So no crazy ideas. They can be handed in during the re-sit at the end of the semester. The original code can be viewed here.
βοΈ The only challenge for myself to write the code as in few lines as possible.
Although I am not 100% happy with the result the D3.js visualization contains a decent amount of features:
- d3.pack
- d3.tsv
- d3.format
- d3.hierarchy
The prettified code is only 48 lines long!
- Pick another dataset
- Find a example code for visualization
- Clean up file structure and make it work locally
- Again some cleaning, but now in the HTML, CSS and most importantly JS
- Change D3 V3 to D3 V4
- Get a working visualization without any errors
- Upload master branch to GitHub Pages
- Edit meta data
- Workflow described in documentation and code
- Replace this document in your fork with own readme!
- ES5 -> ES6
- Add functionalities (ex. hover display amount, force & drag function)
- Style the hell out of this boring bubble chart