Even this website posted is running off of the old data.
The CSV dump from the NYT repo is out of date, although the heavy website and very fluffy news article from the data source has much fresher numbers.
I was hoping be able to quickly look up localized data with a quick bash command every time my threshold of morbid curiosity is breached, but the data is a couple days old.
P.S.,
Thank you OP for bringing awareness to this data source.
Refreshing the Wikipedia page for the pandemic with numbers by country was starting to lose its relief/reward factor to my feedback loop.
The chart makes it appear that LA is a raging wildfire, while SF is under control.
Actually, LA has 2 cases per 10000 while SF has 4 cases per 10000. SF is twice as bad but you wouldn't know it from the chart.
You could use a log scale for the Y-axis, and while this doesn't solve the population problem, at least you can compare the rate of growth or trajectory to judge how well the county is controlling the spread.
7 comments
[ 5.2 ms ] story [ 12.6 ms ] threadEven this website posted is running off of the old data.
The CSV dump from the NYT repo is out of date, although the heavy website and very fluffy news article from the data source has much fresher numbers.
I was hoping be able to quickly look up localized data with a quick bash command every time my threshold of morbid curiosity is breached, but the data is a couple days old.
P.S.,
Thank you OP for bringing awareness to this data source.
Refreshing the Wikipedia page for the pandemic with numbers by country was starting to lose its relief/reward factor to my feedback loop.
What it looks like is that updates are posted once a day to the news article, and later on the NYT pushes the previous day's data to the repo.
So, afaict, there's a >1-2d delay for NYT's latest data being pushed to the repo.
For example, Los Angeles vs. San Francisco -> https://imgur.com/a/k4SykrJ
The chart makes it appear that LA is a raging wildfire, while SF is under control.
Actually, LA has 2 cases per 10000 while SF has 4 cases per 10000. SF is twice as bad but you wouldn't know it from the chart.
You could use a log scale for the Y-axis, and while this doesn't solve the population problem, at least you can compare the rate of growth or trajectory to judge how well the county is controlling the spread.