Unintended dragged node repaired. This can happen when trying to move the map in iD while editting, if the cursor was on a map point. No harm done.

Unintended dragged node repaired. This can happen when trying to move the map in iD while editting, if the cursor was on a map point. No harm done.

At least the ones that match are candidates for clearing the name without loss of information.
But there are errors too, I had noted w.1425467008 (and 3 adjacent) where one field has BN1 2xx and the other BN2 1xx.
And setting your query to "!=" I see w.1432593337 with a random typo/mismatch.
But there are also shape errors, I noted one of the final ones (added *after* your discussion had concluded!) at the N of Peacehaven,
w. 1436971553 managed to get both 10 Telscombe Road and 10 Greenacre - but at least this sort of error suggests they might be deriving the areas themselves not copying directly from a copyright source.
I can see where one could get the areas where a small road has 1 postcode, or 2 for odds vs evens, but not sure how they got (or how we can verify) where a longer road has a change midway eg w. 1416829146 BN2 4TE starts from no.96, and 4TF up to no.94.
But there is probably no big hurry, if we're happy to change these tags by very careful overpass+josm edits rather than reverting.
A bunch of unwanted St->Saint edits are getting gradually edited back.
There's some bad 3D work but the worst can get fixed up. Some arbitrary/random? colour, height values though, and a general lack of care.
And there are also some type=site relations by the user which are just grouping stuff in categories and can go at some point.
Cebderby (Clive)

Good. There are around 1300 landuse=residential + name from this user along the coast from Ferring to Friston when one might expect maybe below 100 named areas in an region this size. On overpass I found 958 with name="BN"... The user did add a few more in 2 changesets dated after this discussion, but seems to have stopped now.
There are also a load with the town/area as name, which is slightly less wrong, but still wrong. eg for "Peacehaven" you would expect either 0 or 1 residential area named this, but there are 173; 59 as Woodingdean; 51 Newhaven; 31 Saltdean etc.

Excellent work. The only alt_name I would consider is the version without the apostrophe eg St George's Road (way/26726656) had alt_name=St Georges Road at v#14 before it got wrongly expanded at v#15. That form seems increasingly frequent on road signs, so maybe worth recording if you see any addresses using it, otherwise just stick with the basic correct form I think.

I am entirely familiar with it. I suggest you read the page you link to, including the parts you don't agree with. eg paragraph 2 not just paragraph 1.

For info, there are entrances at both ends, both more or less equal status and mapped in some detail. At both ends, the entrances share parts of the buildings with offices etc above, the station being underground. At the north end, even the ticket hall is underground, under the Conductor pub. At the south end, a central part of the ground floor of that building is dedicated to the station.
The station node used to be at the south end of Fleet Place which was definitely confusing, but the current mid-platform location is probably the best we can do.

Where the sidewalk is continuously adjacent to the carriageway (no big grass areas etc just an immediate kerb), and has no special access (eg shared cycleway), then the extra footway is a duplicate way. The public highway is mapped and carries the foot traffic. We don't add a second footway beside a farm track if it has a public footpath, we add foot= and designation= to the way. This is just the same; a duplicate way offering no choice of route but adding massively to the complexity of the foot routing calculations and the maintenance of the map. The path here is a massive GPS routing trap, which says that if you route to or from anywhere in the block around Walmer Gardens (gardens) then you MUST go via the E connection of Walmer Gardens (road) and Erlesmere Gardens (road). This is fiction, bad mapping, bad routing.

It's a marginal case. Compared to how it was, it was made dual and feels like it. The river bridge has old and new halves with a physical division, and the W arm of the T junction has a more-than-minimal island and about a 13m spacing between E and W bound. So this section is divided. Beyond that, it's a question of where these 2 ways meet - functionally it's good as it is and OS OpenMap Local agrees. For OSM, the join points back to a single way could be shorter if people want.

Good spot, thanks. Tidied up in ch. 165894556

The addition (not mentioned in your changeset comment) of an airfield, 3 runways and a perimeter track - all marked as if active - is reverted. The remnants of the perimeter way were already accurately mapped, the rest is historic/gone and correctly mapped as farmland etc.

Not sure about that:
https://archaeology.co.uk/articles/news/the-rose-rises-again.htm

Thanks for the clarification. Unfortunately - as my comments and changes showed - I don't take this interpretation; seems we'll likely not agree between us and this will probably need others to comment here or somewhere like talk.gb to get a consensus.

I don't understand. This is where you seem to be claiming all slip roads should be link, but admit that some cannot be. Hence my suggestion above that after the previous discussion you've seen that not all slip roads can/should be link. Is this not the case?

To be completely clear, can I just check with you that you've seen that the A516 runs from the Derby inner ring road to the A50 N of Hilton, and the only part that is not the A516 is where the A38 ref takes precedence. The roundabout outside the Royal Derby Hospital is not the termination and the A516 does not run along the (now tertiary) Uttoxeter Road any more. The change from OSM primary to OSM trunk at the roundabout is probably because of the newer road design/build standard not because an A38 connection exists; the A516 (OSM) trunk continues to N of Hilton.

In the absence of roads linking between A roads or terminating to/from lower types, I do not see that the trunk_link page description has any relevance. The wiki description of trunk roads says highway=trunk. Does this trunk status (which I regard as higher status) not 'win' over OSM's provision of a trunk_link concept to mark roads which connect to/from such ways and generally have no ref or name themselves.

I'm trying the keep the map correct, not to disturb your editting.

So why is it you think 2 parts of the main route of the A516 (but not others) need to be labelled as _link ways?

You came to the previous discussion with the assertion that 'all slip roads shall be link' and after some discussion you concluded that assertion to be false. In the absence of ways linking between A roads, no _link is present.

Wrong, the 2 slips leading to/from the east are the A6, not _link. (The other two to/from the west are connections between the A6 and the A50 and were correctly marked differently as _link). Not every slip road is a _link as previously identified to you.

Wrong at changeset/140063384 (see discussion there) and wrong again. A reminder that there are no link roads here, is a direct merge of two A roads.

This is merely more of the existing tagging along the HS2 route which seeks to differentiate between the land which will become railway (landuse=construction,construction=railway) and that which is currently construction temporarily and is presumed destined to be returned to be useful farmland (assuming the ground level is returned to 'natural'). I don't know where it started. It's imperfect and could be construction=farmland but that is not the purpose of the construction. No construction= tag/value at all could probably be better but then validators would bleat.

"roof:shape=dome" non è un valore previsto o significativo per "barrier=hedge".

Quando riparerai l'etichettatura errata?
changeset/148518846

È necessario risolvere questo problema prima di continuare ad apportare nuove modifiche.