[Tagging] tree rows vs individual trees

[Martin Koppenhoefer]

sent from a phone

> On 10. Feb 2019, at 23:10, Tom Pfeifer <t.pfeifer at computer.org> wrote:
> 
> Thus it is more comparable to the addr:interpolation which we use before all addr:housenumber are mapped individually. Once we have achieved that, the interpolation line becomes obsolete.


I believe this is disputable. Personally, I have never used the tree row tag, I always mapped single trees, in a row or not, but I think there could be a sense in explicitly stating a tree row that goes beyond address interpolation, which really is redundant when the individual numbers are there. For example there can be anomalies like missing trees or corners where different tree rows meet, and an algorithm might not be able to get it right (also because information like tree types could be missing, a human looking at the scene would understand the rows because she sees more than just a binary tree or not, she will see age/height/color/shape (and other species related properties) and be able to distinguish them, even without expert knowledge like knowing the species, just by looking at it. A short way to tell which trees belong together (or which row is interrupted) would be the tree row tag.

I see no harm in having both, provided the individual trees are part of the tree row way.

Cheers, Martin