Updated `natural=rock` and `natural=stone` presets by bgo-eiu · Pull Request #471 · openstreetmap/id-tagging-schema

bgo-eiu · 2022-05-27T22:57:45Z

These updates mostly reflect the discussion in #469 , with the objective of arriving at presets for natural=rock and natural=stone that are more clear in meaning and still relate to how they are used.

Included in this version:

"rock," "stone," and "boulder" as terms for both since enough people are using these interchangeably that it would help to see both options. Added some miscellaneous related terms for both that I found from searching ("cairns" appears to be a term that can be used to describe rocks used as trail markers for example).
"Glacial Erratic / Boulder" as the name of the natural=stone preset because as @westnordost clarified, this is being used more specifically than the name "stone" or "boulder" suggests alone. This would help make it more clear how people are using this tag.
Just "Rock" for natural=rock. This is the more common tag and people have been using it for a variety of types of rocks so the simple name seems OK as long as the terms are there.
Added "Exposed bedrock" as a checkbox field for natural=rock. Per @matkoniecz asking about this, I think that this has a much more clear meaning for adding geological=outcrop to natural=rock. Attached is too ambiguous and was confusing in the first place.
Added dimension field options (height by default; diameter, width, circumference in additional) since these are often used in combination and it would help to know what type of object people are actually mapping.
Added a "Lava field" checkbox field in the additional options for natural=bare_rock which adds geological=volcanic_lava_field. This tag combination is used a lot in Iceland for example. I noticed bare_rock didn't have very much in the presets so that could be a helpful detail to have in there.

1ec5 · 2022-05-28T00:34:12Z

+{
+    "key": "geological",
+    "type": "defaultCheck",
+    "label": "Exposed bedrock",


Field names should be in Title Case.

1ec5 · 2022-05-28T00:42:40Z

+        "erratic"
    ],
-    "name": "Unattached Stone / Boulder"
+    "name": "Glacial Erratic / Boulder"


It’s been a long time since my geology appreciation class in school. “Glacial erratic” is a term I’ve never encountered before, despite hailing from in a region where terms like “terminal moraine” are well known. If there’s no colloquial term for this specific concept, then “Glacial Erratic” is a good name for the preset. But seeing “boulder” next to it might make me inclined to choose this obscurely named preset over “Rock” when tagging a decorative boulder. Maybe “Boulder” should be an alias but not part of the name? (“Decorative boulders”, marketed as such, are so common at garden stores that it could even be a term for that preset.)

I was kind of on the fence since boulder is a more commonly known term. The German concept of "findling" that the tag is used for does seem to be a near exact match to "glacial erratic" though so boulder could just be an alias. Good point about decorative boulders / landscape boulders, I was thinking a lot of what people would describe as boulders count as glacial erratics anyway but those would not.

Thank you for the feedback, I will update 👍

1ec5 · 2022-05-28T00:47:17Z

+    "moreFields": [
+        "width",
+        "diameter",
+        "circumference",


This field is currently called “Circumference (at Breast Height)” because the natural=tree tagging scheme derives circumference from a diameter at breast height (DBH) measurement. (Go figure.) We’ll need to split out a different field for circumferences of other things, like rocks, for which “breast height” would be baffling.

/ref #455

Didn't realize it was that specific - I think there are already petrified trees on the map as natural=rock so it's not totally out of the question but diameter will do.

… usage

bgo-eiu · 2022-05-28T01:09:22Z

Updated to reflect feedback above

bgo-eiu · 2022-05-28T19:01:47Z

Updated to avoid duplication in terms/aliases

westnordost · 2022-05-29T16:35:22Z

-    "name": "Attached Rock / Boulder"
+    "name": "Rock"


Just "Rock" for natural=rock. This is the more common tag and people have been using it for a variety of types of rocks so the simple name seems OK as long as the terms are there.

The terms are not visible to people using the iD presets. Only aliases will be, but then only if one specifically searches those

Well this would lead to natural=rock being tagged (a lot) on free-standing non-attached rocks which seems to be wrong according to the previous discussion/research.

If I understand your point correctly, you are saying that this is what is happening anyway already a lot, so renaming the preset name to just "Rock" is merely following the de-facto tagging situation and not redefining the meaning of a tag. The latter, we absolutely do not want for obvious reasons that I hope do not need explaining.

What I am missing is proof for the claim that this is indeed the de-facto tagging situation and if you have proof, it should first be documented in the wiki. How much percent of natural=rock is not really a rock attached to bedrock and how much percent is on the other hand tagged correctly according to the wiki definition? Without this, we cannot simply shrug this distinction away in an iD preset PR. After all, the very first two sentences of the wiki article mention this distinction:

natural=rock describes a notable rock feature or small group of rocks, attached to the underlying bedrock, mostly used on a single node element.

On the contrary, single boulders not attached to the bedrock are better tagged as natural=stone.

... and iD (presets), too, has been denoting the meaning of natural=rock correctly according to the wiki definition since the preset had been added.

Good point - I have a strong suspicion the de facto use is for any rock due to how the word is understood in English, but I will do some more formal queries on this for documentation first.

westnordost

See https://github.com/openstreetmap/id-tagging-schema/pull/471/files#r884296415

As it stands, this would change the definition of a well-documented tag.

westnordost · 2022-05-29T16:41:53Z

Also note that the wiki definition of natural=stone is

A large (whatever that might mean) natural boulder or rock fragment that is not attached to the underlying terrain.

So, this includes more than a glacial erratic - any boulder or rock that is not attached to the underlying terrain is a natural=stone. This is why I proposed that you use the aliases field.

bgo-eiu · 2022-05-29T17:11:53Z

I am going to look into:

How is natural=stone used outside of Europe, where mappers likely did not understand the idea of a findling? It does seem like it follows "glacial erratic" quite closely within Europe, but if there is a noticeable pattern in other uses that could complicate things. In any case, it does seem worth updating the wiki in some way to explain the findling concept and link to your research on this to clarify things.
How is natural=rock used generally/de facto?

And document the findings for this.

Co-authored-by: Martin Raifer <martin@raifer.tech>

westnordost · 2022-05-29T17:18:27Z

Great! Though, in regards to this, just a disclaimer that I am not the maintainer of the iD presets, @tyrasd is.

In particular, I am not sure what happens if your finding is that indeed, many tagged natural=rocks are actually stones or the other way round: Whether the guideline for contributing to this repo would allow to disregard the (original, documented) definition of a tagging and replace it with a more general one. Maybe there is a precedent for that, I do not know.

tyrasd · 2022-05-30T17:28:29Z

Great that you're going to look into this in more detail @bgo-eiu !

In particular, I am not sure what happens if your finding is that indeed, many tagged natural=rocks are actually stones or the other way round: Whether the guideline for contributing to this repo would allow to disregard the (original, documented) definition of a tagging and replace it with a more general one.

In general I do abstain from modelling presets in a way that contradicts or otherwise mismatches the documentation on the wiki. If you find that the documentation on the wiki does in fact not match real-world usage of the tags, then we first need to find out why that is: If it is because of editor presets which are not conveying the intended meaning well enough, these presets should probably be improved/clarified. If it is because the tagging concept is not well enough defined or too complex for mappers, then the tagging needs to be redefined, optimally through a tagging proposal. It could also be the case that the wiki just contained some misleading information, then it would suffice to update the wiki.

tyrasd · 2022-05-30T17:29:16Z

-        "name"
+        "name",
+        "height",
+        "geological_outcrop"


Something that irritates me a bit is that the proposed added field for geological=outcrop to the natural=stone preset currently only has very few usages (only 30 map features have this tag combination). The wiki also doesn't mention this particular usage.

@bgo-eiu ping?

I have not looked much into PR in general, but I plan to close it as abandoned given no reaction to comment over 3 years. Just to confirm: is it abandoned?

tyrasd · 2022-05-30T17:30:05Z

 {
    "icon": "temaki-boulder3",
+    "moreFields": [
+        "geological_volcanic_lava_field"


For the geological=volcanic_lava_field tag the combined usage with natural=bare_rock seems to be quite popular. Perhaps it could make sense to have this combination as a dedicated preset, so that it can be more easily found when mapping such features?

matkoniecz · 2026-03-20T21:23:51Z

waiting for whether it is abandoned, see https://github.com/openstreetmap/id-tagging-schema/pull/471/changes/BASE..02aed633bb3252085518ae07feb114c2c505c0ad#r2862909126

bgo-eiu added 6 commits May 27, 2022 17:57

added geological outcrop field

bd1d019

added more terms + fields for rock

1c9309b

added dimension fields and more terms for stone

ca17efb

Create geological_volcanic_lava_field.json

d76ecd0

added lava field field for bare rock

20fc9a1

added more terms for rock + width field as option

6be7bd2

bgo-eiu mentioned this pull request May 27, 2022

Semantic correction to natural=stone and natural=rock #469

Closed

1ec5 reviewed May 28, 2022

View reviewed changes

bgo-eiu added 4 commits May 27, 2022 20:58

title case field name

eabc7de

title case field name

2a6ca1b

added decorative boulder terms based on feedback; removed circumference

e2ea3ce

added boulder alias; removed circumference to avoid conflation w tree…

b5251bf

… usage

tyrasd reviewed May 28, 2022

View reviewed changes

Comment thread data/presets/natural/stone.json Outdated

bgo-eiu added 2 commits May 28, 2022 15:00

avoiding duplication of labels

209f71b

avoiding duplication of labels

eb7a322

tyrasd reviewed May 29, 2022

View reviewed changes

Comment thread data/fields/geological_outcrop.json Outdated

tyrasd reviewed May 29, 2022

View reviewed changes

Comment thread data/fields/geological_volcanic_lava_field.json Outdated

tyrasd added the enhancement New feature or request label May 29, 2022

westnordost reviewed May 29, 2022

View reviewed changes

westnordost suggested changes May 29, 2022

View reviewed changes

bgo-eiu and others added 2 commits May 29, 2022 13:12

add missing brace

afcba36

Co-authored-by: Martin Raifer <martin@raifer.tech>

add missing brace

02aed63

Co-authored-by: Martin Raifer <martin@raifer.tech>

tyrasd reviewed May 30, 2022

View reviewed changes

tyrasd force-pushed the main branch 2 times, most recently from 9d3204d to 49f529e Compare June 22, 2022 16:19

tyrasd force-pushed the main branch from bc29c6c to c5e947f Compare July 12, 2022 16:04

matkoniecz added the new-field create a new field (see add-field for cases where field from presets is added to new entries) label Mar 13, 2026

matkoniecz changed the title ~~Updated natural=rock and natural=stone presets~~ Updated natural=rock and natural=stone presets Mar 14, 2026

matkoniecz requested changes Mar 20, 2026

View reviewed changes

Comment thread data/fields/geological_outcrop.json

matkoniecz added waitfor-requested-changes This PR is waiting for an update to incorporate requested changes from a PR review. waitfor-pr-author-response and removed waitfor-requested-changes This PR is waiting for an update to incorporate requested changes from a PR review. labels Mar 20, 2026

Conversation

bgo-eiu commented May 27, 2022

Uh oh!

1ec5 May 28, 2022

Choose a reason for hiding this comment

Uh oh!

1ec5 May 28, 2022 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Choose a reason for hiding this comment

Uh oh!

bgo-eiu May 28, 2022

Choose a reason for hiding this comment

Uh oh!

1ec5 May 28, 2022

Choose a reason for hiding this comment

Uh oh!

bgo-eiu May 28, 2022

Choose a reason for hiding this comment

Uh oh!

bgo-eiu commented May 28, 2022

Uh oh!

Uh oh!

bgo-eiu commented May 28, 2022

Uh oh!

Uh oh!

Uh oh!

westnordost May 29, 2022 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Choose a reason for hiding this comment

Uh oh!

bgo-eiu May 29, 2022

Choose a reason for hiding this comment

Uh oh!

westnordost left a comment

Choose a reason for hiding this comment

Uh oh!

westnordost commented May 29, 2022

Uh oh!

bgo-eiu commented May 29, 2022

Uh oh!

westnordost commented May 29, 2022

Uh oh!

tyrasd commented May 30, 2022

Uh oh!

tyrasd May 30, 2022

Choose a reason for hiding this comment

Uh oh!

matkoniecz Feb 27, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Choose a reason for hiding this comment

Uh oh!

tyrasd May 30, 2022

Choose a reason for hiding this comment

Uh oh!

Uh oh!

matkoniecz commented Mar 20, 2026

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

5 participants

1ec5 May 28, 2022 •

edited

Loading

westnordost May 29, 2022 •

edited

Loading

matkoniecz Feb 27, 2026 •

edited

Loading