It's an odd name, but it's real: https://www.ubyurbeo.com/finding-your-home/u-hansfield

At least they didn't call it "Ur Hansfield"

What was the source for this mapping? I've already updated the tagging to reflect the fact that no construction has begun, but I'm concerned that the source may be a copyrighted map.

Hey, I'm curious about the western end of the Sallins Link Road, which is showing as a cycle way only. We have highway=cycleway. Because of lockdown, I haven't seen this for ages, so maybe this reflects the main roadway being unfinished, but just in case...

Yes, I drove the route, the junctions and some key crossing roads, though in most cases I was tweaking earlier work performed by others during construction.

Interesting. As a local, I'm surprised at the names, but some name is probably better than no name, and they are attested.

Corduff Roads 1 and 2? New one on me - where do these names come from?

The gate you added was on the roundabout, indicating a complete traffic blockage. I've removed it.

I'm concerned about these edits. highway=trunk is the correct tag for all primary A roads (green signs). It has nothing to do with the (invisible) concept of a Trunk road. Similarly a non-primary A road is tagged as highway=primary (yes, this is wacky). From what I can see, your edits have left us with a large number of green-signed roads now tagged as though they have white signs. Not good. For at least some of your edits, I know for a fact that these are primary A roads. I surveyed them myself. Unless I've misunderstood your intent, these changes will need to be reverted.

I need to flag an error in the source tag for this edit: the Mapillary footage I was looking at was for another nearby roundabout. I'm still convinced there could be no stop signs at an approach to a roundabout, so leaving the edit in place, but if somebody knows better, I'd love to hear the story :)

If so, yes, but it’s not for certain that a link road to R744 is part of R744 (or wasn’t to me). I changed it from motorway_link, which is certainly wrong, but wanted to play safe. Feel free to retag based on better info.

These are fantasy edits. I have restored the previous state. Please don't spoil the map for others!

No, you moved some shops to completely wrong locations. You understand that you are editing the map that others are using, right? I have reverted these edits - please be more careful.

A long time ago, when we imported this data set of place names, we wondered whether we might wish to track future updates to the source data set. For that reason, we included a lot of meta data of this kind from that source. By now, it is clear that we will never do that, and I have been removing the tags when I encounter them.

Hi - this edit contains highly detailed outlines of newly built houses. The only hinted data source is the Bing aerial imagery tagged as part of the change set, but that is several years too old to feature the buildings. What is the actual source?

I recall it being wall mounted

The change set appears to relate to some old ODbL cleanup. I may have done a tagging fix or something on this node, but I have no knowledge of the POI itself. I do see buildings nearby that are consistent with the label, but I'm not prepared to move the node without a real source. I have added the likely buildings, so perhaps somebody local will finish the job.

It would be anachronistic if Dunleary wasn't the name in daily use by most people. But it is. The vague similarity between the spellings of the Irish and English names coupled with an initiative to get people to move to the Irish name has led to an odd situation where people say "Dunleary" but write "Dun Laoire" or some other Creole of the Irish name. There is no more valid candidate for name:en and it is furthermore valid to allow nomination to match searches for it. The name tag remains at the Irish name, which reflects on the ground signage etc.

The bit problem with this idea is that most of these particular junctions already have a highway tag set, highway=traffic_signals. And what we realise, I think, is that the status of these nodes as junction points on this particular route is not their primary purpose, making it problematic to set a highway or ref tag specific to this one secondary purpose. This is why I'm convinced that a route relation is the correct way to solve this (though even this has challenges in how to tag, say, the junction refs).

Hi. It's not correct - the motorway_junction tag is only appropriate for an exit from a controlled-access road (doesn't have to be a motorway). Everything about the tag is intended to be interpreted by, say, navigation systems, as a routing decision point where the only action is to exit, usually to the left (or right, depending on country). Likewise, rendered maps interpret the tag as something that should be displayed prominently even when zoomed out aggressively. Keep in mind that, for most of these junctions, it is in any case impossible to apply the tag highway=motorway_junction, since highway=traffic_signals is already in place. A strong clue that this tag was never intended for a traffic situation of this kind.

The Dublin city through route scheme is a bit of a misfit anyway for digital maps and especially for machine navigation, since it's essentially intended to solve a navigation problem manually (signs on the ground and printed maps depicting the routes). It's not clear that a user of machine navigation should even care about those routes or junction points. But if it is appropriate for us to map them, we need to do so with some assumption of use case. How will software or map renderers want to use this? I adopted the highway=city_junction tag as a way of retaining the information captured by those mappers who did the existing work until we can work out what we actually want it for. But even that tag isn't satisfactory, because of the traffic signal conflict I mention above. In reality, I think the key to managing this properly is to define a route relation for the circular routes, with the junction nodes being included in the relation with a suitable role ("junction"?). A relation is surely the only way to go for a bunch of other reasons too. Even the ref tag causes us problems - when you have highway=traffic_signals, ref=J42, doesn't it seem like this should be the ref for the signals?

So this is where I'm coming from. I'm happy to discuss any valid use cases for this data, because without doing so, I fear we will not work out a useful tagging schema.

Please start explaining your edits using a commit message. The exit sign for this junction names R772, and until a new routing via Wicklow can be clarified, it is best to indicate this connection as R772. http://www.mapillary.com/map/im/qcHHM-WDdyfwSZBnXAaoLQ/photo