A sequence of 26 changesets over the last 6 months* has replaced several tags used to define the area of landforms created by landslides with hazard=landslide.
The tags that were replaced with hazard=landslide are:
natural=area of the landslide
natural=شلال معركة السلالم (“Staircase battle chute” according to Google Translate)
natural_2=landslide
natural=lansdslide
natural=land_slide
natural=Landslide
natural=slide
natural=landslip
natural=slip
natural=broken_ground
geological=landslip
natural=landslides
geological=landslide
geological=mass_movement
The replacement with hazard=landslide seems to have been a misunderstanding of how the tag should be used. The OSM Wiki describes the hazard=landslide tag as indicating designated areas where landslides are known to occur, e.g., by tagging a sign placed to warn people about the area. So this tag would seem to be appropriate only where the danger of landslides has been officially designated.
In contrast, the natural=landslide page describes the use of the tag to mark the area of a debris field of a landslide. This page also proposes the use of the geological=landslide tag for “historical” landslides.
Four of the 26 changesets were recently reverted by DWG as undiscussed mechanical edits. However, the remaining changesets are still in place. I have three questions about the remaining changesets:
Should the remaining changesets be reverted in part to restore the tags that were in OSM before these changes?
If the original tags are restored, should some of the obvious synonyms be changed to natural=landslide?
Should the hazard=landslide tag be removed from the landforms created by recent or historical landslides or should this tag be retained? (That is, are all former landslides currently hazards?)
* The OSMCha link above includes two extraneous changesets in the results.
Thanks for the comment. Regarding the hazard=landslide tag, since the changes were made broadly and in some cases as mechanical edits, I think it’s likely that the nature of the hazard was not verified before this tag was added.
That said, it might not be a problem to leave this tag until someone can verify it.
It looks like there’s some agreement to restore the original tags for landslide features, but some differing opinions about whether to correct some of the tags in the process.
To clarify my question 2 above, should the following tags be changed to natural=landslide or be restored unchanged?
After Andy’s comment, I think the right approach here is to partially revert the changesets in question to restore the OSM tags exactly as they were before the changes.
That is, yes to 1, and no to 2 and 3.
I may revert the tags in a day or so. If anyone disagrees with the approach, please let me know before then!
Following the discussions in the Rock glaciers retagged as moraines thread, I’m holding off on reverting these landslide changesets in case the mapper would like to join the discussions here.
Many a road here in the region tagged hazard=landslide per the triangular signs with exclamation mark and subsign “Frana”. Occasionally “Crollo” where the road, full width or part went off the hillside. Poor as this region is some like a road section from Penne to Farindola it’s permanent movement so they just patch it with a cement like gravel. One gets used to it.
Yes. That’s exactly the right usage for the hazard=landslide tag.
And that’s a good illustration of why the natural=landslide tag means something different when it’s used to map the outline of the landform created by a landslide.
The mapper that made these changesets deleted their account after being temporarily blocked by DWG. I have reverted the tag changes in the changesets as discussed above in this thread.
If there’s an interest in cleaning up any of the tags that have been restored, let’s discuss that in a separate thread.