Kind request to reconsider rendering natural=fell

[apophenic-treehugger]

Dear renderers,

Your effort to create a beautiful rendering of our data is highly appreciated!

I know the topic has been closed and re-closed (#545). However, as some years have passed since, and in light of a recent discussion, I'd like to kindly encourage you to re-evaluate your decision not to render natural=fell.

Expected behavior

Render natural=fell in a colour somewhat lighter than natural=grassland.

Actual behavior

natural=fell is not rendered.

---

This is not technically an "issue". I wasn't able to find out how to create a feature request.

[pnorman]

I'm leaving this open because #545 was a meta-issue that considered several tags at once.

@apophenic-treehugger can you outline what has changed from when it was previously considered and how that is connected to the link?

[apophenic-treehugger]

Thank you for giving this another chance!

I believe the rationale in #545 was that you don't want to render natural=fell because some mappers use the tag in a wrong way. But to be honest, I did not fully understand the discussion.

What has changed from when it was previously considered:
I have concerns that the carto team was perhaps not aware how prevalent this landuse type is and how big a loss it is not to render it, just because some mappers use the tag in a wrong way.

How is that connected to the link:
The link shows that there is a strong consensus in the mapping community that natural=fell is the correct tag for a large portion of alpine areas.

Why I think it should be rendered:
Fell is a common landcover type. If it is not rendered by the most popular renderer, this incentivizes mappers to use other (=wrong) tags for fell to force its rendering or not map it at all.

The mapping community seems to agree that no other tag than natural=fell is correct for this landcover type. Most areas of the map where this tag would apply are currently void of any landcover data -- I strongly believe that this is because it is not rendered (mappers get little reward for adding it to the map).

[imagico]

@apophenic-treehugger - i have unlocked #545 now - usually we do so after some time but we missed that in this case and it was left locked. All previous discussion of natural=fell rendering can be found in #545. Anyone interested in the matter is strongly advised to read up on that there before commenting.

I hope the discussion started by martianfreeloader leads to some improvements in differentiated mapping of herbaceous vegetation (which is underdeveloped in OSM compared to mapping of woody vegetation) - though so far the comments there seem to be mostly dominated by OSM-Carto haters without much interest in improving tagging concepts. Keep in mind that natural=fell has, by its proponents, always been promoted as a landform/ecosystem tag and not as a vegetation/landcover tag. Hence the practical mapping problem martianfreeloader brought up (which is about mapping herbaceous vegetation) has not much to do with the tag natural=fell. The only connection is that the case example presented with photos is something proponents of natural=fell would consider suitable to be mapped with natural=fell (as a landform). As a landcover tag it would clearly be considered an umbrella tag for a multitude of different landcovers for which more specific tags exist and are rendered by us (as discussed in depth in #545 and also explained on the OSM wiki).

@pnorman - our established principle is that we try to keep discussions on the same subject together. We have in the past closed all new issues opened on natural=fell (like #2923, #3165) as duplicates of #545. Since there is extensive (and relevant) past discussion of the matter on #545 i think disconnecting any new discussion from that would be quite counterproductive.

[hungerburg]

Creating a mock-up of what OSM-Carto might look like if natural=fell was rendered immediately helped me discover two wrong mappings.