Hi @ECDC_EU, I'm looking at COVID-19 case numbers you're publishing and I noticed that Spain has negative number on 19/04/2020. That looks like a mistake. What's the way to sort it out?
-
I was looking at case numbers that I would expect to be less prone to retroactive corrections. But even if there was a data mistake, I would expect ECDC to correct the original mistake within the record it occurred instead of compensating in future dates.
The reason they wouldn't do that is that it creates two versions of historical truth, and once you allow mutability of past data, there's a risk that using derivative data (e.g. cumulative deaths) from different versions of the past could lead to further errors.
-
-
If the raw data is published as an immutable event log, it's more resilient. An alternative would be to use bilinear time coordinates (T, t) for every datapoint, ("this is what we thought, at time t, about time T"), but that's an order of magnitude more complex.
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.