I didn't check the history who created the duplicated service roads. Instead of creating additional duplicated ways the correct method would have been to verify that the already mapped highway=construction ways and change these into highway=residential in case that the construction work is over. But as I do not have any information about which of those ways are now finished I decided to remove all the superflous ways instead and wait for a local mapper to make these changes after verification.

Sorry that I messed this up. I do not have local knowlege. From remote the most likely oneway highway=primary situation which just ended without any connection was that this was an accident. Thanks for correcting the situation now!

Thanks for cleaning up the first part. Here there are still a lot of problems:
https://tools.geofabrik.de/osmi/?view=routing&lon=71.60584&lat=33.77498&zoom=11&baselayer=Geofabrik%20Standard&overlays=unconnected_open_ends_1

But the situation here looks a bit different. The AI might have detected real missing ways. But the connectivity is broken as can be seen on the red marked items from OSM inspector.

It would be great if you could manually go over the import, connect the new ways if you spot them on one of the image backgrounds and delete the bad data if you cannot detect them.

There are still lots of useless ways in the middle of nowhere tagged as residential and unconnected. See https://tools.geofabrik.de/osmi/?view=routing&lon=76.18536&lat=35.74031&zoom=8&baselayer=Geofabrik%20Standard&overlays=unconnected_open_ends_1
Please stop importing this useless data generated from broken AI!

Thanks for the clarification. Just for your information: I did check the history before and to me user @Charmscobb didn't look to be a vandal to me.

I'm following the community forum and as you (or somebody else, I don`t remember) announced that the revert has finished I thought that now is the time again to re-start with improving the data. If this is not the case please announce the status in a clear way on the community forum. Thanks!

The kerb in the west of way/1223987957 was not connected and marked as error by the OSM Inspector. See osm.wiki/OSM_Inspector/Views/Routing

Sorry, I thought it was finshed.

Hi!

Your edits for way/231954252, way/1211671215 and way/1211671214#map=15/47.2350/16.6565 look very inaccurate as they are lacking connectivity to the existing road network. Please re-check and fix, otherwise I plan to revert/delete these changes.

I now added a service road in order to fix the currently broken ferry routing. Please verify if this is edit changeset/141918016 is correct and remove the "fixme" tag if you know better. It if this is correct for cars it is unclear to me what is the correct entry point for pedestrians and bicycles.

Hi,
yes it was this user who destroyed it. And no, I did not report this vandalism.

Thanks!
I now fixed the connectivity issues and added the (logical) oneway=yes tag, please re-check.

Hi, please note that usually we don not map roads as areas. I therefore deleted all your imported road areas.

Hi and welcome to OSM,

can you please explain and improve your changes around way/1210497303? This tunnel and the connectivity to the existing ways do not make much sense. It seems that the newly created way should be oneways, but I doubt that it is possible to have a tunnel below of the roundabout, this does not make any sense to me.

Hi and welcome to OSM,

Please note that we do not map routes as ways as this leads to duplication of ways. The correct method to map a route is using a route relation. See osm.wiki/Relation:route

Please note that this segment is already mapped with a two route relation for hiking, see relation/15667312#map=14/40.5938/8.9693 and relation/15641575#map=14/40.5938/8.9699

You seem to missing a third route for cycling.

But please also note that we only map routes which are marked on the ground. I'm not sure if this is the case here. Do you know this?

Hi and welcome to OSM,

please note that your tagging of oneway=yes for a highway=pedestrian does not seem right. First highway=pedestrian is usually used for pedestrian zones. See highway=pedestrian#Vehicle_access. The way in this area seem more like ordinary highway=footway. Second a oneway for a footway is very unusual. How is this enforced ?
Third you created a couple of connectivity issues because of the parallel running ways. So please consider turning the ways into turning to highway=footway and deleting all the superflous parallel ways.

Hello xbarnada_nexus!

This changeset disconnected most of the ferry routes in the port of Barcelona from the road network. As a result the ferry routing from/to Barcelona is currently broken.
Please check node/11071043689#map=19/41.36956/2.17710, this node is not connected.
If I roll back your changeset I can see that the previous road connectivity looked more accurate. It used these still existing ferry_terminal nodes: node/468523299 and node/30717964

As it looks as you have on-site knowledge I would ask you to answer the following questions:

1) Why did you introduce the new node node/11071043689 ?
2) Where is it possible to enter/leave the ferries on a vehicle? The two still existing ferry_terminal nodes also do not seem to be 100% correct as they are only connected to service roads, which are only indirectly connected to the main roads.

Hi,

thanks for pointing this out. I was confused because I didn't see the undergroud parking and on esri the underground part was not visible.

I fixed it by splitting, putting an layer=-1 on the segment and amenity=parking_entrance tag on the entrance node.

I tried to revert this changeset via the JOSM reverter plugin, but did not succeed because of a hell of conflicts

I'm seeing multiple changesets from different users according the same pattern, just duplicating data, not responding to comments. I just informed the DWG and asked for a block.