Distribution doesn’t extend beyond the country borders:
Something about the algorithm: The used data only based on the waiting time votes of hitchwiki.org users. The map is rasterized by 0.1 degree. Every database record inserted by normal distribution (sigma = 50 km) and weighted with the number of the votes. 10 hours or longer waitings are filtered out, because angry people could make false data. Motorways, main roads, ferry lines and populated places are also visualized.
What is the difference between the two version? I inserted each waiting time experience to an exact location by normal distribution with 50 km deviation. (If I just insert values without any distribution, the map wouldnt be useful because it would only contain information per point.) The difference between the two map is only that normal distribution go through the boundaries or not. For example when I inserted an experienced data of Austria near the border that cause effect only in Austria and nothing in Hungary, Czech or any other country. This is the ” Distribution doesn’t extend beyond the country border” version. In the other one borders are ignored, (but showed in the map of course).