Don't forget to add the roles (in most cases this will be "outer") to the boundary ways

Hi Si,
I am a little wary of giving these community boards a place in the administrative hierarchy, as they don't seem to have any statutory basis and have no form of elected or otherwise representative government. In any case adopting a specific admin_level value should probably be checked in talk_gb first?
Regards, Colin

Hi Paul, the source for the boundary you added is strictly speaking not for the civil parish, but for an electoral district or ward of Ryedale DC. They may be coincident, but that is not guaranteed. A better source for civil parish boundaries is Ordnance Survey, which was actually used in this case (see the tagging of the boundary ways).

Remember this object represents the area, not the council. The council name provides a bit of context, but the councils themselves are not represented in OSM as they are not "geographic data". This council calls itself "Comhairle nan Eilean Siar" in English as well, so this is its name:en. But it's not worth fighting over so I will let it go.

I disagree with this change. "Western Isles Council" is a translation, and not the name the council uses. Even in an English context, the council name is "Comhairle nan Eileanan Sar" - check out their website if you don't believe me.

No, relation #13466955 is not an admin boundary. However the ways of which it is composed, do represent (parts of) admin boundaries.

The area will stay unnamed. It is not defined directly in law. Only parished areas are defined explicitly, and unparished areas are areas which do not fall within a civil parish. This boundary represents a "hole" in the administrative hierarchy, and the only reason a polygon exists in OS Boundary-Line is to ensure the topology of the whole country has complete coverage.
In the past there may have been some kind of council or other "thing" with a name that "owned" this polygon in some sense, but that is not currently the case.
I'm sure nobody needs reminding that any similarity of naming between government and Royal Mail is "purely coincidental."

Hi! Can you explain how removing nodes from an admin boundary is an improvement? The source is reasonably authoritative, so these nodes are not there for nothing. I am referring specifically to way/1053157085

Do you mean points with identical coordinates? Or just close together? The effect of the DP is clearly visible... What tolerance did you use? In any case it would be bad practice to sacrifice accuracy in the database, unless a certain limit has been agreed beforehand.

Hi Baz,
your change to the parish boundary seems to be at odds with official records... How did you determine the new run of the boundary?

Hi,
Why did you simplify the coastline? The input data from Ordnance Survey is already generalised. Why remove detail? Where do you draw the line in terms of accuracy/resolution?

I am curious... Why did you remove the wikidata tag from the Glascwm relation?

Hi Billy... I am not convinced Sealand should have any kind of admin boundary as it is not recognised by anyone except its own inhabitant(s). But it should most definitely not be at admin_level=1!!! Normal sovereign countries are at admin_level=2, and level 1 is kind of "reserved" for supranational administrations if they ever become hierarchically superior to sovereign states.

Hi Mateusz... This change is not right. The previous wikidata link, Q23939248, reference the correct area (including the City of London), but this one is wrong as it excludes the City of London. There are about a million interpretations of "London" and it's a confusing mess, sorry about that...

Hi! The issue is solely around the admin centre of the civil parish administrative boundary. The "rule" is that the admin centre is set to the place node (representing the place where the main council offices are located). So it's not about the geometric centre, or a good place for a label, or the exact building. In this case is should be the place node for the place of Kings Hill. The admin centre in the boundary relation was correct (and now is correct again), but you moved it to some other node.... Why did you choose that node specifically?
More info on osm.wiki/Relation:boundary#Relation_tags

NB: your change to the admin_centre of Kings Hill parish has been reverted - please check the wiki for how this role is used

Please don't fiddle with UK admin boundaries. If you change something, please state your source in the changeset comments.

My bad, I think I copied one too many relations over to the new boundary way. Thanks for fixing!

Please note the correct admin level for Greater Manchester is 5 as it is now a Combined Authority and no longer an administrative county.

Why are you moving the admin boundary between Llannor and Pwllheli? Do you know something the Ordnance Survey don't?