(asking for community input per this post)
From what I understand, _link roads should not have names. Yet, I found some that do. What is the correct tagging for these roads? Thanks!
Attached is a screenshot of the area in question, with the problematic roads selected.
A true name on a highway=*_link road is rare but does happen. I would be suspicious of it sharing the exact same name as the road it’s next to and message the mapper who originally added those names to see if they have more information.
It appears as though these roads were originally tagged as highway=secondary, but were later changed to highway=motorway/secondary_link while still keeping the name=* tags. I have asked the respective users about this.
The link ways currently named “South Columbus Drive” should be much shorter, beginning near the traffic island, not where the lane change restriction begins. Without more context, I can’t tell why these classifications were chosen, but as long as it’s a link way, it should go unnamed.
I improved line geometry in Changeset: 184631223 | OpenStreetMap
turn lane tagging can be probably added to Way History: ‪South Columbus Drive‬ (‪1532602180‬) | OpenStreetMap
I would map this then I would put https://www.openstreetmap.org/way/23825832/history%20https://www.openstreetmap.org/way/1000504230/history as highway=unclassified as extension of Way History: ‪East McFetridge Drive‬ (‪377665850‬) | OpenStreetMap - this does not look like a slip road
but _link continue to confuse me so not sure about last paragraph
Internationally, about 9% have names (type name into the search there), less so in the USA
It’s certainly not that common, but I’m always worried when I see things like “should go unnamed”, because there are always exceptions. If you’ve been there and verified that there is (a) no signage on the ground and (b) it would make no sense for there to be any, because of the nature of the _link way, then yes, remove what someone might have previously added. If not, don’t?
I agree that we should be open to the possibility of exceptions.
I ran into one case near me where I placed a highway=primary and highway=motorway_link as separate carriageways in a two-carriageway road.
I strongly considered putting the road name on both carriageways – the primary and the motorway_link, on the basis that it looks like a dual carriageway road when you’re on the ground.
I got the name of this road, “JT Connell Connector” from local news reporting at the time, since it was a brand new road that was installed where no road had previously been.
In that news report, it actually says:
Drivers will still be able to turn right onto the JT [Connell] Connector Road to access the Pell Bridge.
(emphasis mine)
Now the Pell Bridge is a highway=motorway, the eastbound carriageway of JT Connell Connector is a highway=motorway_link because it only goes to the bridge. And local news reporting refers to it as a named road.
So, while I’ve resisted putting a name on that motorway_link, I actually think it would be 100% correct to give it name=JT Connell Connector.
I intended to give guidance about the specific way in question, not a blanket statement to be quoted out of context, not even for the U.S. as a whole. If that’s how people read forum posts, then I would be worried too.
That said, in the U.S., the two main uses for link ways are unnamed almost as a rule in the real world: a slip lane at an intersection or a freeway/expressway entrance/exit ramp. If this way were a slip lane, it would’ve been misleading to tag it with the name of the road that led to the slip lane.
Back in the day, I did experiment with naming any freeway ramp in Ohio based on the reference location signs, but later on I moved these names to official_name=*, because they serve the same purpose as carriageway_ref=*.
“Connectors” like @ZeLonewolf’s example are relatively rare. Of course I’ll add a name tag if the signs make clear that it has a name or if the name is otherwise verifiable. However, I would not make up a descriptive name where the signs only indicate a destination.
The case in question was something else entirely: the continuation of South Columbus Drive into East McFetridge Drive at a bend where an on-ramp continues straight ahead. I agree with the current approach of tagging it as a bona fide part of South Columbus Drive, and therefore highway=secondary. Now please don’t selectively quote “I agree” to mean that I agree with everything always as a rule. 
Haha, thanks for the input! (I hope this is what DWG meant by “Ask the community first before making any changes, and make sure they’re unanimous, too!”)
