The AONB boundary clearly goes around the village of Alderton there. The only sensible thing to do would be to follow the boundaries (or on-the-ground boundary features) of the village properties, probably as depicted on OS Mastermap.

As far as I know, the digital Land Registry boundaries are also drawn by reference to OS Mastermap. They sometimes contain errors, but if you match them to various visible boundaries on Bing imagery, you can get a good idea of where they are right, and where there may be discrepancies.

Either way, I think it's unlikely that the official boundary of the AONB would have been deliberately chosen to run a metre or so inside the boundary of the properties in the village, rather than following the boundary itself.

I've detached the AONB boundary and put it back to the Natural England GeoJSON dataset: changeset/120298934 .

I'm not sure it's settled that such boundaries shouldn't be attached to other objects when they are indeed coterminous. In general I would expect common boundaries to share the same nodes, though I can see practical reasons for not wanting to do so here.

More importantly, the reason I moved the boundary was to align it with the Land Registry Index Polygons around the edge of Alderton. Almost certainly this is what the AONB boundary should be aligned to. However, it looks like the data from Natural England has an offset of a metre or so from this -- possibly due to coordinate transformation errors. How confident are you that the Natural England data is correct, and are their GIS files definitive anyway?

I've heard back from the FSA, who've been in touch with the City Council. Apparently, the Council made an error in their upload, which caused some of the ID's to be suhffled. They've now uploaded a corrected version to the FSA. Hopefully this means that the FSA data will revert to the IDs I added by tomorrow, and then the FHODOT tool will update too.

Hmmm. I was using Greg's FHRS tool, with the automated JOSM links. So I think when I added those FHRS ID's they would have been correct.

If you go to https://gregrs.dev.openstreetmap.org/fhodot/?lat=52.13146&lon=0.18814&zoom=9&layer=fhrs and click on the Cambridge City polygon on the map, it will load a graph of matches against time. In March 2020, you'll see a drop in the green matches (which was caused by the Council switching to some new software and assigning new numbers -- see the 18th Feb item at https://ratings.food.gov.uk/Status ). It then bounces back, which would be me updating most of the entries. Then shortly afterwards the pink block of mismatches appears. This is presumably caused by the assigned IDs changing internally for some reason.

See https://www.whatdotheyknow.com/request/food_hygiene_rating_scheme_data for details of how the IDs are assigned by the FSA. I guess either Cambridge City Council's software as mixed up the IDs they assign, or the FSA has done the same with the mapping from the Council's IDs to the ones they issue. Either way, it's a pain for OSM if the ID's aren't stable, immutable, and non-reusable. I'll see if either Greg RS, or the FSA have any insight into what's happened here.

Yes you're right -- thanks for spotting that. I've now removed that way from the relation.

The St Edmund Way was previously mapped as following the Stour Valley Path around the edge of Sudbury. But when following it the other day, the signs clearly take it into the town centre, though they're a bit patchy there.

The gap where I couldn't be sure is around Bullocks Lane. There's a sign sending you SW down Meadow lane, but there isn't any way out there. I couldn't find any signs along the old railway line either.

I've now removed the rjw_update tags.

Sorry about that. There were added as an intermediate step in part of JOSM workflow to flag ways that I was going to update. They should have been removed before I hit upload, but it looks like I forgot. Feel free to remove any you come across. I'll sort them if not out when I get a moment.

On the other hand, most of the businesses along there list there address as "St Benedicts Street" or "St Benedict's Street". Given the discrepancies, I think we need to determine the official name from Norwich City Council, and use that.

Regarding the list of streets the references to "Saint" are all in the description fields of other road segments. It's quite common elsewhere for the road names in descriptions to be out of date, so I would conclude that the descriptions aren't always updated when road names change. It's quite likely someone has decided to standardise the names in the name column, the question is whether or not that was done officially by Norwich CC.

For what it's worth, OS Open Roads (which should be authoritative) also lists the name as "St Benedicts Street". Have you contacted the City Council yet and got a response?

I don't know I'm afraid. I would suspect that there was only ever one phonebox there. The changeset was done using Map.me on my phone, so it's entirely possible that I missed the other nearby node that had already been mapped.

I don't know I'm afraid. I was just passing through and didn't have time to stop. There are definitely signs for NCN 13 along the cycleway just north of Chadburn Road. I looked quickly at the junction of Lackforth and Bull lane and didn't notice and signs either way there. I've added a note at note/3029426 for someone to check.

In this changeset you seem to have edited node/1281726035/ to change the brand tags from Budgens to Co-Op, but have left the name as Budgens. Can you confirm if the store has changed to a Co-Op, or are there perhaps two different stores in that building?

(It's not a suitable source for OSM, but I see that both chains list a branch at 24 University Way on their websites: https://www.budgens.co.uk/our-stores/wharley-end and https://stores.welcome-stores.co.uk/bedfordshire/bedford/24-university-way.html .)

You seem to have changed brand:wikidata=Q152096 to not:brand:wikidata=Q152096 on way/61845189/history in this changeset. Can I check if this was deliberate? Or perhaps it was an unintended consequence of clicking the "tag as not the same" option in iD?

As the street naming authority, the name according to the council is by definition correct. The street name from the council will also be used in addresses by Royal Mail, and we want those to match the street name too.

In this case, "Saint Benedicts Street" is arguably incorrect, despite the etymology.

That said, I'd regard it as a bug in Nominatim that a search for that doesn't find the street. Equally, it seems that a search for "Saint Swithin's Road, Norwich" (with a grammatically correct apostrophe in it, which is missing from the name in OSM) returns no results. I'd also regard as a bug in Nominatim.

Yes, I'm aware of that. But if the official name is "St Benedicts Street", then then "St" in the OSM tag isn't an abbreviation of the name.

(So this is different from e.g. taking "West Main Street" and abbreviating it in OSM to "W Main St")

Hi Emily, Many thanks for your contributions to OpenStreetMap.

I see you added a bank at node/9320540695 in this changeset. Do you know its name (i.e. which bank it is)? That would be useful information to add.

Similarly, does the cafe at way/1010179149 have a name?

In the absence of a response here, I've done my best to move node/5619417432 to a more appropriate location, based on the address.

This was a while ago, but an odd tag "addr:vector=opposite" appears to be been added to way/43303737/history in this changeset. There's only one other instance of this key worldwide, so I was wondering what it's purpose is, and if the information would be better represented some other way...

This was a while ago, but an odd tag "addr:vector=backward" appears to be been added to way/59551258/history in this changeset. There's only one other instance of this tag worldwide, so I was wondering what it's purpose is, and if the information would be better represented some other way...

Hi,

It looks like you've changed the address of node/5619417432 in this changeset, but haven't updated the position of the node. Could you move it to the correct place on the map? (Presumably it's somewhere around osm.org/#map=17/53.71981/-0.45933 .)

Could you have another look at way/401032045 ? The postcode isn't right for this location. (See https://osm.mathmos.net/addresses/pc-stats/B/B36/0/#0DD )

From a bit of searching online I think Summerfield Education Centre may not be occupying new buildings on this site: way/84658252 , which would fit with the postcode. But can this be verified?