Hiya, thanks for adding this. Is this a newly created nature reserve? I’ve not heard of it before (only stuff about planting and conservation in Ryelands Park). Where’s access to it from, is it off Austwick Road or Marshaw Road? Thanks :)

You’ve avoided responding to any suggestions that these edits should be done under the automated edits/imports policy, and now you’re asking the local editors to not review your edits and effectively allow you to mark your own homework.

In the meantime, while this discussion has been doing on, you’ve logged another ~17 similar edits in this area. Are you trying to make your set of changes too big to fail?

Your most recent edit (177988990) still has all the same problems:
- Mass-adding address and UPRN details while also changing unrelated things (pipeline URL) which makes it harder to review or revert a changeset without collateral.
- Adding UPRNs to things which I’m not confident the UPRN definitely represents (way/1465345074).
- Mixing geometry and tagging changes.

I’ve reported this to the DWG, because despite the weight of local mappers telling you to stop, I think we need a third opinion. It’s also gone beyond my abilities to revert things.

Hiya, welcome to OpenStreetMap and thanks for your recent improvements around Maryport.

Just a hint — if you press ’Q’ while a building is selected in the editor, its corners will be automatically squared up. This can make drawing regular buildings a lot easier!

Happy editing :)

Cheers!

Was way/1472817871 meant to be industrial=railway landuse=railway? Those tags don’t quite seem to match the rest of the tagging in this changeset.

A lot of the house number and house name data in Ambleside is contributed by tourists, who are here for a limited time, probably distracted and don’t know the area that well, and hence I wouldn’t really trust it to interpolate from.

I don’t think it’s safe to guess at house numbers like that, even if you are using multiple sources to cross-check. While you might be right most of the time, house numbering in the UK can be surprising at times and I think it will end up introducing errors.

This kind of thing is much better left to an on-the-ground survey, and apps like StreetComplete make that really easy. Someone will get to surveying it eventually.

I don’t think you’re fully understanding the impact of reviewing these changes.

There are currently about 25500 individual changes in my area of interest from you, which have not been reviewed yet.

Reviewing your previous edits has so far resulted in tens of follow-up changes, from you, me and other local mappers.

The only way to review those remaining 25500 changes is either to review them one by one (in which case splitting edits up geographically helps with tracking progress, but doesn’t actually reduce the number of things to review; but more on that below); or to revert them en-mass (to remove the need for review; more too on that below).

Having now spent hours going through some of your changes, I am convinced that the automated edits policy (osm.wiki/Automated_Edits_code_of_conduct#Scope) applies to them:
- You are importing from the UPRN database.
- You have at least a partially scripted workflow; or
- you are manually changing tags without adequate review.

The fact that rskedgell has spent considerable time planning similar imports (osm.wiki/Import_of_UK_postcodes_and_UPRNs_(England_%26_Wales)) suggests the same.

You may say that your workflow can be (and has been) improved as a result of feedback, but that doesn’t deal with the huge pile of currently unreviewed changes, and it still means that any future edits you do will need to be reviewed.

As I’m sure you know, formalising your editing process by following the automated edits policy and publicly planning and documenting the import process ahead of time is designed to remove the need for that one-by-one review of changes, by checking that the process itself is not going to introduce errors into OSM.

That might mean fewer UPRNs get imported as quickly, because the underlying geometry isn’t yet in a state to match well with them, but I think that’s fine. It’s not a race!

So: subject to me being able to work out how to do it without introducing further errors into OSM, I plan to revert the larger of your unreviewed changes in the area. Please don’t make any further edits. Once you’ve documented your import process and got it reviewed (or joined in with rskedgell on their already-in-progress import), it should be straightforward to re-import the UPRNs here.

Sorry about this, I realise you’ve put some time into these edits. But I think this train was set in motion as soon as you started submitting multi-thousand-object changesets without following the imports process.

And the house numbers on Greenbank Road (e.g. way/931037222), where did they come from? A few of them have come from splitting terraces (e.g. way/461754049/history) but unless I’m missing something many of them haven’t?

sigh, I pressed enter too early.

I’ve undone it as changeset/177825330

I’ve undone it as

Where did the house numbers for some of the houses on Kirkfield Rise come from? e.g. way/423109955

Why change way/423109954 and way/423109955 from a more specific building type to building=residential?

Wonderful, it’s great to see some more attention being paid to the map around Horton :D

Nice work! Thanks for these changes.

If you’re going to be making more changes to farmland tagging, consider taking a look at osm.wiki/User:Gurglypipe/landuse, which is a writeup of the consensus tagging from a few of us for farmland and meadows in the north west.

In the case of this changeset, I think it would mean tagging the fields as meadow=pasture as well as landuse=meadow, as from memory all the land around Horton is improved or upland sheep grazing.

Hope that makes sense, happy to discuss further if you have thoughts on it. Happy editing :)

I’ve now checked and fixed everything in Heysham and north up to The Battery in Morecambe.

Well spotted! Sorry I missed that one. There’s still a lot to review and clean up :(

Whoops, sorry I missed that one when cleaning up some of the UPRN mess.

I’ve reverted this as changeset/177651865, as we don’t know exactly what the UPRN represents.

From a couple of minutes of reviewing this (I am nowhere near finished, nor do I really want to waste any more of my time on this):

1. way/1228810819 and way/1228810820 have a suspicious addr:street

2. Same for way/1191358663 and various other buildings around there

3. UPRN for way/171150926 should probably be on the surrounding site

4. 4 UPRNs for way/83934347 looks like it needs more investigation

5. Similarly for way/158856135

6. Buildings like way/1158031835 should be addr:city=Backbarrow not addr:city=Ulverston, to match the rest of the buildings in the village. (Or if you’re going to insist on using postal towns in addr:city, then addr:suburb needs to be set as well)

‘Lathe’ (or ‘laithe’) means a field barn, so you can probably safely retag buildings like this as building=barn.

https://en.wiktionary.org/wiki/laithe