I'm talking about the east side ramps. Those are the ones you edited to add the ref=11 and name tags. The west side is tagged correctly, that is, as a trunk_lnk without any names or ref added.

I believe the tags you added are incorrect. They aren't harmful, in fact, they could be helpful to a simplistic routing algorithm. However, the reason I was editing that intersection was to add destination and destination refs to the first sections of the highways branching off the 1136. I think that will supply information to routers in a more appropriate way.

I'm not going to remove any of your tagging but I think you should.

Respectfully,

Dave

Hi,
You tagged the ramp on the east side of the Super Highway at the junction of route 1136 as highway=trunk and name:en= Super Highway
I believe this way and all it's sections should be tagged highway=trunk_lnk and should not have either a ref=11 or a name. It merely provides access the the Super Highway and therefore should be tagged just like the similar ramp on the east side of the Superhighway.

What do you say about this?

Dave

Hello again and thanks for your reply

I'm quite aware of your point about not mapping for the renderer. The "Map tool" I'm using is a Garmin-compatible map I created myself and I've never before seen it exhibit this behavior nor have I ever seen this tagging scenario. I can adjust my compiler coding to display the actual Exit 49 destinations but I thought I'd politely check with you first. I did not intend to delete the ref tag.

As far as I can tell, and I may be wrong about this, the ref that you applied to all the ways making up Interchange 49 is still problematical. The Wiki text you cited might or might not apply here because the ways in question, those following the junction node, are not part of any relation nor are they signed in any way. Furthermore, the exit is not "signed as 49" as you state but is signed as "Exit 49" as are most other exits I've checked.

So, to reiterate, I'm not questioning the use of the junction:ref=49 tagging which, although new to me, seems correct but your use of the ref=49 on the other ways in that interchange. None of the six or seven other I-90 interchanges that I inspected in for purposes of this discussion have any ways tagged with a ref=nn as you've done for the Exit 49 interchange.

I have written and answered many Changeset comments before and believe it is an accepted way to carry on these discussions. Sorry if that offended you.

Dave

Hi gadjet,

Yesterday I noticed that my GPS was telling my to turn at "49" while driving north on I 90 near Buffalo. I've never before seen notation for an exit on my GPS. It should be telling me to turn to reach NY 78, Depew Lockport, or something similar.

When i tried to understand why it was reporting "49" instead of the more usual destination(s), I discovered what I believe to be a redundant ref that you added to the way following the motorway_junction node. Here are the tags for that way:

destination:ref=NY 78
destination=Depew;Lockport
highway=motorway_link
junction:ref=49
lanes=1
maxspeed:advisory=40 mph
oneway=yes
ref=49
... Etc.

The junction:ref = 49 is fine but the other ref (the last line above) is confusing my GPS and I believe it is also incorrect.

What do you think?

@quincylvania
I updated my river tagging preset to use your scheme with rapids:name. I use the same scenario for named bridges with a bridge:name tag. Many OSM mappers change the name of the highway to the name of the bridge in such cases but that is obviously incorrect. The same logic applies here.

Why relations don't supply tags to their members really puzzles me. It seems only logical that unless specially tagged differently, all members of the main_branch of a river should have the same name tag and the tag waterway=river as it would if the it were only a simple way.

Good talking with you....

Thanks for the clarification quincylvania. I was feeling embarrassed that I had misunderstood the Wiki. My memory wasn't clear enough to recall how I made the decision to tag a segment of the river with waterway=rapids but now I understand. Interestingly, the McKenzie River relation includes that segment as a member of its "main_branch" (I might've done that myself but again, my memory not clear) but apparently renderers don't look at the relation, only the way itself.
Anyway, I retagged the segment in question to use waterway=river and rapids=yes. This doesn't break the river way yet still carries the appropriate information so overall it seems a better approach.

Is everybody okay with that?

Another thought is that in this case, and I don't recall the specifics of my decision, when a rapids has a measurable length, say 100m or whatever, then drawing a weir across the river isn't a very accurate way to represent it. My intention was possibly to somehow model these rapids that way, I dunno.

I think you should change the way it's mapped. When I did this, I was under the impression that a tagging a segment of the waterway was a legitimate way to map rapids. It looks like the tagging you suggest is correct, or more correct. I may have taken my guidance from an older Wiki article, I really don't know.

Hi,
I believe the tagging is correct as is but feel free to replace the line of rapids I've drawn as a way with a node if you wish.

Splits in ways are very common in OSM. We do it for highways all the time, for example, whenever the speed limit changes. Dealing with such splits is a rendering issue.

But, as I said, feel free to change it to a node or weir if you wish.