Yeah, I'm not particularly keen on this either but there's not currently any consensus on the best way to tag these.

I think disused=yes might be the best approach actually. Maybe I'll just document that on the wiki and implement it on OpenInfraMap.

Hi,

You added some names for unpopulated places here as place=suburb. These place tags are intended for populated places only, so using them could be considered "tagging for the renderer":

osm.wiki/Tagging_for_the_renderer

In the case of Dartford Marshes, this was already mapped as a wetland area, which does actually get shown on the map:

way/572429422

I have removed and changed a few of these place nodes accordingly.

Aha, thanks! Looks like this is the 132kV underground cable for that project - I've re-tagged it.

Hi,

I honestly thought this usage was documented, but apparently it isn't!

As the voltage tag is semicolon-separated, I think this usage of the frequency field is useful to indicate which voltage is AC and which is DC.

The validator error has been annoying me, and I was planning to get that fixed at some point, but I will see about getting some consensus for this usage and documenting it first.

Russ

Reverted by changeset #110620381

Reverted by changeset #110620381

Reverted by changeset #110620381

Reverted by changeset #110620381

Reverted by changeset #110620381

Reverted by changeset #110620381

Reverted by changeset #110620381

Reverted by changeset #110620381

I have posted about this on the talk-gb mailing list with some more detail, and I'm pausing mapping for the moment:

https://lists.openstreetmap.org/pipermail/talk-gb/2021-August/027595.html

I'm fairly sure a multipolygon is the correct approach here - there's not really any other way of tagging this (and there should only be one "correct" way).

If it were a designated nature reserve, we could put the name on a separate `boundary=protected_area` area, but Kielder isn't a nature reserve.

Many large forests are tagged as multipolygons - there are quite a few in continental Europe like this. There are also some famously massive multipolygons for bizarrely-shaped national forests in the US, but these are generally protected areas.

Part of the issue with Kielder is that it's mapped as a load of small landuse areas. I think many of these could be combined, with any gaps denoted by `man_made=cutline`, which would make the relation less unwieldy.

I'm not the biggest fan of relations and I do try and avoid them where possible, but I think in this case it's the correct approach.

Ah, my apologies. I can't remember what the context was with my previous edit as it was a while ago - I see you've fixed it now.

Hi,

I believe this change broke the coastline - coastlines in OpenStreetMap are handled in a special way and need additional care, because a broken coastline anywhere in the world will prevent all coastline updates.

JOSM should warn you if there are coastline errors, and you should pay particular care to these warnings.

I have fixed the issue in this changeset - hopefully I haven't broken anything else:

changeset/107149078

Cheers,

Russ

Hi James,

Just to let you (and Bernard) know that I have fixed these issues with the coastline.

If you do make any changes in future, it's worth keeping an eye on the coastline inspector to see if you've inadvertently broken anything:

http://tools.geofabrik.de/osmi/?view=coastline

Hi James,

Just to add a bit more context to this:

Coastline is handled in a special way by OpenStreetMap, which means you have to be careful when editing it. Your edits have generated several errors in the coastline, which will prevent the coastline from being updated worldwide until they're fixed.

Unless you are confident in editing coastlines, it's best not to try and add or remove segments. Moving existing coastlines is much safer.

Secondly, coastline should be mapped to the high-water mark (which is normally at the top of beaches), and I think your edits are closer to the low-water mark. The "OS OpenData StreetView" imagery layer shows the high-water mark.

Oops, fixed. Thanks!

Hi Alex,

Oh, I missed the photo! This location is definitely "Burghfield Generation Plant" - it matches the image (google maps screenshot) on SSE's website:

https://www.ssethermal.com/flexible-generation/operational/burghfield/

From the sign in your photo it looks like the location to the northeast is Burghfield Landfill Site. That one is likely burning methane from the landfill.

I moved your photo back to the other object but it might be worth getting it renamed on Wikimedia Commons.