Hadrian's Wall is no longer showing up in rendered tiles (it should be & at lower zoom levels!)

[jeffreyameyer]

Style change requested

Describe what you want to see added or modified in the style. Is it something easy to explain like a color fill or rendering something not currently showing? Or is it a more complex need specific to historical mapping?

We should have some sort of representation of historical walls, particularly those of a military and defensive nature (although probably agricultural and wooden fences, as well...)

I believe this was rendered in the past, but it is now invisible and is a key missing piece of visual context for the towers and encampments that were located along the wall.

There are a variety of possible treatments for this type of historically significant barrier, and we can add whatever tags might be appropriate to differentiate / signify this importance.

Affected Tag or Tags

If you know what they are, let us know the specific tags we should target for style changes barrier=wall military=cordon

maybe something others, to show:

Importance? important=yes When it should start rendering? scale_rank=6 ???

[tsinn]

Looking into this issue a bit, it seems as though the Hadrian Wall is not being rendered because it is not in the vector tiles. Digging a bit further using overpass turbo and looking for walls in the data, it looks like this wall isn't in the vector tiles because it is a relation. Walls that are ways are in the vector tiles and rendering in OHM. So, once they are brought into the vector tiles, I think they should render on the map.

[tsinn]

In this map, blue lines are ways, and they are in the vector tiles. Pink lines are relations, and they are not in the vector tiles.

[danrademacher]

So that's probably in here that we would need to pick up relations in addition to nodes and ways: https://github.com/OpenHistoricalMap/ohm-deploy/blob/d2733ab25aace49f536dbb6f58b4a2f11dce8f8b/images/tiler-server/config/config.toml#L260-L280

Similar to https://github.com/OpenHistoricalMap/ohm-deploy/blob/d2733ab25aace49f536dbb6f58b4a2f11dce8f8b/images/tiler-server/config/config.toml#L303

So we'll need @Rub21 to consult on this and see if that's an easy change -- since relations can include nodes and ways, I am not sure whether we have to make a new layer like other_relations to go along with other_points, other_lines, and other_areas?

[flha127]

Related threads : #276 #284 & #439

OHM only renders multipolygon relations, multilinestring aren't (however, this part of the Hadrian's Wall is tagged as type=collection, it should be multilinestring for the defensive wall).

@danrademacher as for multilinestrings they should be added to existing layers if it is possible so as to avoid dupllicating the rendering rules.

As for object types, multilinestrings should only include linestrings. Public transport routes as here should use the older V1 schema as we don't need routable relations (Public transport V2) and as to avoid having route relations with both linestrings (roads or train/tram tracks) and points (stops).

[jeffreyameyer]

@vknoppkewetzel - fyi - this is a great example of a physically small (i.e. narrow) feature with big historical significance.

[danrademacher]

And also an example where cartographic treatment is rather blocked by a mismatch between the data (Hadrian's wall as a multi-linestring relation), the vector tile database (which doesn't pull in multi-linestring relations), and long-stymied issues (like this one) around how we handle relations (vs nodes and ways)

[vknoppkewetzel]

@jeffreyameyer thanks for this example