Talk:Tag:barrier=hedge

From OpenStreetMap Wiki
Jump to navigation Jump to search

What's illustrated on the main page is, indeed a hedge. It's important in the UK to document and map "hedgerow" as distinct from the manicured hedge found in and around gardens. A hedgerow is very important historically, culturally, naturally, legally and for mapping, especially in rural areas. Wherever possible hedgerows should be mapped, and the tag 'hedgerow' not used interchangeably with the hedge described on the main page here. See here for distinction [1] [[User:UrbanRambler|UrbanRambler}] 10:09, 27 January 2010 (UTC)

Mapping hedges as areas?

There is not current discussion on the wiki page about mapping hedges as areas, though this is sometimes done and it's noted as a possiblity in the ValueDescription box (onArea=yes). Should it be stated clearly that "area=yes" is required in this case, to avoid ambiguity? Most closed way features with barrier=hedge have a tag like landuse=meadow and were not intended to be interpreted as areas. --Jeisenbe (talk) 03:17, 3 August 2019 (UTC)

At github, Pnorman suggests not mapping hedges as areas, because most of the current mapping is mistakes due to including barrier=hedge on an area feature. See https://github.com/gravitystorm/openstreetmap-carto/pull/3844#issuecomment-538698692 and previous discussion in https://github.com/gravitystorm/openstreetmap-carto/issues/971 and https://github.com/gravitystorm/openstreetmap-carto/pull/3834.
I was wondering why my carefully mapped thick hedges with area=yes tags (and hedge=thick_hedge from an earlier OSMWiki definition which seems to have been changed/removed now) were no longer showing as nice green fills. Seems to be a result of discussion here:- https://github.com/gravitystorm/openstreetmap-carto/pull/3844
Surely the best course would have been to initiate a mapping project for worldwide mappers to correct the tagging of problem areas rather than alter the rendering to accommodate tagging errors? Tagging errors should be corrected anyway and there's also the "don't map for the renderer" rule to consider - these seems to be a case of "don't render for the map." Thick, heavy hedges - some of them so old that they fall into the category of ancient monuments in the UK for the purposes of protection from removal - that have been mapped as thick hedges using area=yes with barrier=hedge and possibly hedge=thick_hedge are not accurately shown as open areas with a 1-pixel green boundary (as they are now) as they are an enclosed area that is dense vegetation throughout. The removal of render on barrier=hedge, hedge=thick_hedge, area=yes stand-alone tag sets makes the map suggest that there is an open area bounded by an enclosing hedge, which is not a reflection of the situation on the ground.
Where, say, a field bounded by a hedge has been drawn as a polygon and tagged with landuse=farmland and barrier=hedge (often these may be mapped as two separate polygons but then the validator throws a "two identical ways" error (or however it is worded) and mappers combine them into one) then the renderer coding needs to be set to prioritise landuse - however, in this case there is no area=yes tag so it shouldn't be a problem. Before this change I hadn't noticed fields I'd mapped this way rendering as a "solid" hedge instead of a field - possibly because there was no area=yes tag.
Could the change to the renderer (and the attendant OSMWiki section on the barrier=hedge page) not be updated to use the hedge=thick_hedge tag as the cue to shade the area? Then, in the landuse=farmland/barrier=hedge/area=yes scenario described as the rationale for change in the linked discussion above the hedge would render as a hedge around the perimeter and the field would render as farmland. Where large hedges are mapped to the spec' I first read herein appear (barrier=hedge/hedge=thick_hedge/area=yes) they would render as they once did, regardless of the presence of the area= tag. It is also unlikely that the hedge=thick_hedge tag would occur where landuse polygons also have barrier= tags attached - if it has been used correctly as per the original definition.
I've noticed on older-mapped sections of the map when editing that the validator has thrown up "deprecated tag" errors for closed wags tagged with area=yes that no longer need it so maybe a campaign to clean up the map and remove non-applicable area=yes tags would be an idea? Indeed, when I first discovered OSM back in 2015 the in-browser iD editor was adding area=yes tags to closed ways that now do not require it. I've put some time into going over my early edits and removing these where they are no longer required.
Clearly, having OSM editors adhere to a common set of "rules" is also desirable to avoid the situations described in the linked discussion (I notice that the current JOSM and current iD differ in the presets they offer for the same feature in their tagging dialogues in some instances, for example). Also, bringing new mappers' attention to the existence of a standard and its documentation - the OSMWiki - more prominently would be a good idea. For example, when you launch JOSM you get links to JOSM help and introduction pages but there is nothing to flag up the existence and need to map in accordance with the content of the Wiki on that initial screen. Editors should have this information - with a link to the Wiki - front and centre and big n' bold, in my opinion.
In any case, in light of this change to the rendering style the hedge Wiki needs updating as it still specifies using area=yes for thick hedges. Personally, I think the renderer should be put back, at least partially, to how it was so that closed ways tagged solely as hedges still render as they once did - possibly conditional on the use of hedge=thick_hedge described above. Perhaps a grown-up will mull this one over and look into it, please?
Either way, I'd like my filled thick hedges back, if that's alright with everyone (I live in and map an area that is rural, with many, many ancient and very thick hedgerows)! It also allows one to correctly map those nice landscape garden features knowns as "box hedges." Ta everso. --John Grubb (talk) 21:54, 4 May 2020 (UTC)
I also live in a area where thick hedges are common. And with "thick" I mean that they have two paths with a 3-way crossing inside of them. If these aren't "hedges", I don't know what they are. They are not "woods", not "garden", not "grass". Are they "scrubs" perhaps, despite being managed?
But if these thick hedges are "scrubs", then shouldn't "grass" and "grassland" be the same thing? The current situation is like mapping a forest only using "tree row". Either way, I'd like my filled thick hedges back, if that's alright with everyone. --Christoffre (talk) 22:23, 26 May 2020 (UTC)