View Single Post
Old 19 Dec 2018, 14:17 (Ref:3871414)   #3372
grantp
Subscriber
Veteran
 
Join Date: Oct 2008
Posts: 6,396
grantp should be qualifying in the top 3 on the gridgrantp should be qualifying in the top 3 on the gridgrantp should be qualifying in the top 3 on the grid
Quote:
Originally Posted by Mike Bell View Post
The most worrying thing is that the app is for EV charging.....
Maybe they change GEO location when a charger is out of action?

That said, back in the 00s I got hold of several years of UK Serious Accident data for interesting database application testing.

The accident records contained geo positioning data and so it was relatively easy even back then to fire up a mapping system to see the location of the accidents in the report.

For the best of the records one could compare the position of the dot to the rest of the data in the record and see that it was accurate to within inches.

However some Police Forces seemed to have more difficulty with either ascertaining the accurate location or, possibly, transcribing it from handwritten notes into the database. The clusters of serious accidents off the coast of Scotland in the North Sea were difficult to explain with any certainty. Also those some way to the West of the Scilly Isles.

Given that by the time the data sets were available to the those who had been allowed access, they had, one would hope, been through some sanity checking processes and some quality assessment. The large number of errors - not all as obvious as records pointing to the middle of large areas of water - made one wonder about the quality of the rest of the information.

There was, for example, clear statistical evidence that ages 20, 30, 40, 50, 60, 70, 80 and 90 were by far the most likely to see people involved in some way in a Road accident. Intermediate ages in those decades were, generally, much safer although 25, 35, 45, etc. seemed to appear somewhat more frequently than one might have expected.

I never did think to check whether the accidents at sea and the specific age bubbles coincided in any way. I might revisit the data just to see what I missed ...
grantp is offline