[talk-au] Assistance with ongoing disagreement regarding intersections

[osm.talk-au at thorsten.engler.id.au]

I’m in agreement with you that this is not the correct, widely accepted, way of mapping these intersections.

 

Ways should only split at the start of physical separations. Intersections of dual carriage ways should result in a # like pattern. Turn lanes without physical separation should be mapped using appropriate :lanes tags.

 

To give counterexamples of what I think are correctly mapped intersections:

 

https://www.openstreetmap.org/#map=19/-27.24256/153.02079

https://www.openstreetmap.org/#map=19/-27.24173/153.02469

https://www.openstreetmap.org/#map=19/-27.23018/153.02236

 

 

From: Dian Ågesson <me at diacritic.xyz> 
Sent: Friday, 4 March 2022 15:31
To: OSM Australian Talk List <talk-au at openstreetmap.org>
Subject: [talk-au] Assistance with ongoing disagreement regarding intersections

 

Hello,

I'd like some assistance resolving a disagreement I'm involved with regarding the correct mapping of dual carriageways at intersections. I have previously mentioned this topic on the mailing list here: https://lists.openstreetmap.org/pipermail/talk-au/2021-September/014968.html <https://lists.openstreetmap.org/pipermail/talk-au/2021-September/014968.html,> .

To summarise briefly, a very active contributor prefers to model dual carriageway intersections in a manner that I don't believe is correct.

Turn lanes are split from main carriageways at the start of the new turn lane, then cross over each other in an "X" shape, rather than a Box shape that I've seen documented. (Examples, because I am bad at explaining: Burwood Hwy/Mountain Hwy <https://www.openstreetmap.org/node/8917929878> , Smith St/Dandenong Rd <https://www.openstreetmap.org/node/2951838115/history> , Burwood Hwy/Dorset Rd <https://www.openstreetmap.org/node/8925914559/history> , Princes Hwy/William Rd <https://www.openstreetmap.org/way/779286918/history> ) Additional highways are introduced for left hand turns where there is no physical separation (eg, Mt Dandenong Tourist Rd/Mountain Highway <https://www.openstreetmap.org/way/113685299/history> , Greville St N/Sturt St <https://www.openstreetmap.org/way/976666655/history> , Glenleith St/Church St <https://www.openstreetmap.org/way/75040109/history> ). This editor has been an extremely active contributor for many, many years: I found these examples by just zooming in on a given town or suburb, found intersection that was modelled this way, and checked the history to confirm the source.

I initially engaged with the user in September (111051481 <https://openstreetmap.org/changeset/111051481> ), and after some initial delay, we have engaged in a productive conversation <https://resultmaps.neis-one.org/osm-discussion-comments?uid=206929&commented>  since. To the user's credit, they have been patient and understanding in our interactions, and have made adjustments to their mapping style based on my feedback. Unfortunately, we have reached a fundamental point of disagreement <https://www.openstreetmap.org/changeset/118038711> , and I don't believe further changeset discussions are going to be productive.

I'm now a little too close to this discussion to be objective, and I would really appreciate some assistance with this disagreement. Due to the extraordinary output of this user, simply avoiding editing in similar areas isn't going to be practical. But am I incorrect in my assessment of intersection modelling? Is this a question of style, or of accuracy?

Kind Regards,
Dian.

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.openstreetmap.org/pipermail/talk-au/attachments/20220304/8ea5e253/attachment.htm>