Only way in not detected

[arnt]

Hi,

https://www.sprawlmap.org/#15.9/48.132642/11.617775 shows a dead-end street (which perhaps isn't, but forget that for now). It's connected to the wider street network using a single east-west blue street.

Shouldn't the westernmost part of that blue street be either green or red?

Now that I look closer, there's an intriguing bit slightly northwest... https://www.sprawlmap.org/#15.9/48.137231/11.61352 shows a red street being connected only to a short blue north-south street. However, in the real world there is a footpath westwards from the point where red meets blue, and cars can go eastwards from that point too. Is the blue street blue because the code knew about those two?

Arnt

[arnt]

https://www.sprawlmap.org/#15.9/48.123098/11.628919 shows another misclassification. Here's another view of the same area, based on ISM: https://www.naviki.org/en/naviki/plan-route/#p=48.12283047855735,11.62888443957479&z=16

The two blue streets near the middle (Gögginger Straße and Vinzenz-von-Paul-Straße) are connected by a path, but ignore that for the moment. Assuming that Sprawlmap doesn't consider them to be connected to each other, shouldn't they be red?

Arnt

[Adam Millard-Ball]

Yes, the pedestrian and bicycle paths in this example are contributing to the connectivity. That likely accounts for the blue street being blue, since the algorithm considers the pedestrian path.

Of course, the problem is that pedestrian paths may not be as complete as the road network in some parts of the world.