I agree that there is a need to define the correct way to tag the center-line of a two-sided embankment or earthworks.
This was mentioned previously in the discussion starting here: https://lists.openstreetmap.org/pipermail/tagging/2019-May/045798.html and continued here: https://lists.openstreetmap.org/pipermail/tagging/2019-May/045798.html It could be called an "earthwork": "a raised area of earth made, especially in the past, for defense against enemy attack", or "embankment" or "rampart" instead of "berm". This is in use 200 times as barrier=earthworks (https://taginfo.openstreetmap.org/tags/barrier=earthworks) or historic=earthworks - 196 times (https://taginfo.openstreetmap.org/tags/historic=earthworks) or perhaps barrier=earth_bank - 184 times (https://taginfo.openstreetmap.org/tags/barrier=earth_bank) There are a few uses of barrier=rampart and military=rampart. In contrast, barrier=berm has been used only 68 times, and man_made=berm 14 times. But there are even more uses of embankment=yes as a standalone tag along the center of an earthworks/berm/rampart/embankment, and there is also quite a number of features tagged man_made=embankment + embankment=both or embankment=two_sided (https://taginfo.openstreetmap.org/tags/embankment=both and https://taginfo.openstreetmap.org/tags/embankment=two_sided > 600) I don't think it would be a good idea to redefine man_made=embankment to be two-sided. And using embankment=yes alone is a bit of a problem since it is a unique key. But I'm not convinced that "barrier=berm" or "man_made=berm" is better than "barrier=embankment" Also, I would oppose mapping berms as areas, especially if they are under the "barrier=" key - these features are (almost) always linear, and because "man_made=embankment" can be used to make the exact location of the top of the embankment. On 11/28/19, Graeme Fitzpatrick <[email protected]> wrote: > On Thu, 28 Nov 2019 at 20:47, Volker Schmidt <[email protected]> wrote: > >> >> >> - First of all when I see a both-sided sloped linear elevation I do >> not know I see a berm.because it is a rarely used term >> >> Can you see a pile of dirt? If so, it's a berm. If it's a bricks, blocks, > rocks, concrete, wooden, steel or any other type of construction, it's not! > I agree it's not very common in normal usage, but would be quite common in > any engineering context, & there are any number of OSM terms which are very > specialised! > >> >> - Second it is ill-defined from the shape point of view: it can define >> a step in an earth wall and it can mean an earth wall with sloped >> sides >> >> Yep, either one could be a berm. > >> >> - Fourth it can describe man-made or natural objects. >> >> I saw reference to berm also being used to describe a line of debris > thrown up on a beach by a storm, however, that wouldn't be a permanent > feature, as it would be changed by the next storm, or human use of the > beach, so we wouldn't map it. > > Let's go back and define what we need >> > > Agree with everything you say here > > >> >> - (this tag is purpose-free) >> >> Sorry, I don't understand what you mean by "purpose-free"? > > Thinking about it we may only need to add a new man_made=slope_base tag. >> Packaging this together with the opposite man_made=embankment in a >> "slope" >> relation, this would give us the possibility to model even complex >> objects. >> Without a relation a closed man-made=slope_base way could be used to >> draw >> the footprint of levees/dykes. The use could be similar to >> water=riverbank. >> > > I did suggest area=slope or similar, to map the area of levee walls in > discussion about large flood control levees a couple of weeks ago, but it > didn't seem to go down very well? > > Thanks > > Graeme > _______________________________________________ Tagging mailing list [email protected] https://lists.openstreetmap.org/listinfo/tagging
