Visualization of CoViD-19 data live
News are posted on Twitter from @c_haider .
I started this project out of dissatisfaction with the visualizations of CoViD-19 case data. Most interesting to me, was how the daily changes develop, i.e. how fast is the change changing?
The first step was to get accustomed to Amber, a nice Smalltalk running in the browser. After solving all kinds of issues with the tool stack, I started with the app and the access to the JHU data.
For gathering data, I use VisualWorks, since it is my home environment with all my tools. Besides, I think that it is one of the most productive development environments available on this planet.
The 3 data files (confirmed, deaths and recovered) are downloaded and combined into series of datapoints, each with the 3 attributes confirmed, deaths and recovered. The result is a list of all countries or states with its data series. This list is then written as JSON to the data directory on the server where the amber app sits.
The latest change to the data structure is that I put the list into an envelop together with metadata (currently just the modification timestamp). Data (#meta #data).
This JSON is then read by the Amber app.
The German official data on county level is also downloaded and transformed into the JSON schema from above. Interestingly, the data have more interesting attributes: age group and sex! Also, the structure of the data is different: it is a stream of the official reports from the counties of that day. Therefore, the data had to be accumulated to create the time series.
Now I want the code to be so flexible so that other data sources can be added easily. AFTER I manage to automate the downloads and distribution.
The assembled JSON data can be used by anyone, also from within the browser.
GET https://covidcrt.uber.space/data/CoViD-19_JHU.json
GET https://covidcrt.uber.space/data/CoViD-19_RKI.json...to be written...
First, you need to install Amber, as shown in Amber Instructions.
Then, you get and setup this project. In a dirctory of your choice do:
git clone https://github.com/ChristianHaider/CoViD19UI
cd CoViD19UI
npm run initAt last, start the server with amber serve and go to http://localhost:4000/ in your browser