OpenStreetMap logo OpenStreetMap

Changeset When Comment
118038711 almost 4 years ago

Supt,

I respect that it is frustrating to have many changeset comments coming to you, and I really appreciate that you have remained civil and responsive in our interactions. I get the sense you feel as though this is a personal disagreement, and I really want to assure it is not.

There are always going to be differences in style, and I have tried to refrain from badgering you about stylistic choices that aren't wrong, even if they aren't my preference.

Unfortunately, modelling intersections to include ways like 1036691022 (a left turn way where there is no slip lane or physical separation) is just, well, wrong. I'm not trying to force a new standard or mapping rule onto you, I'm trying to bring your attention to how your methods differ from what is generally accepted.

You mentioned you get annoyed or disheartened when your changes are reverted. That's not going to stop if you and I simply agree to disagree personally: other mappers will eventually change these intersections back to what is considered "right".

I don't think more changeset comments are going to be productive, but I have raised the topic in the talk-au mailing list, again: https://lists.openstreetmap.org/pipermail/talk-au/2022-March/015925.html.

Please, engage with the discussion there with other mappers. We are all bound by the community consensus and some objective, impartial input would be best to resolve this.

Take care,
Dian

118038711 almost 4 years ago

Hey Supt. The intersection at Point/Narre Warren-Cranbourne Road has the same problems as we've previously discussed for dual carriageway intersections. I'm resolving a note in the area and have removed the unnessary/overlapping ways.
---

Published using OSMCha: https://osmcha.org/changesets/118038711

116487794 almost 4 years ago

This popped up on the list of recent comments, so I had a look. If it is a shared_zone, if might be best tagged as a living_street? :)

117383192 almost 4 years ago

Hi HighRouleur.

I'm going to undo the chang you made here.

Kingston Central Boulevard has been extended across where this taxiway once was. I only very recently updated this road to match its new geometry.

I would strongly suggest that when a road looks glaringly out of sync with the aerial imagery that you check a few different sources. In this case, I have uploaded numerous Mapillary images driving down that road.
---
#REVIEWED_BAD #OSMCHA
Published using OSMCha: https://osmcha.org/changesets/117383192

117344410 almost 4 years ago

Sorry for the belated reply; I wanted to have a proper look at this on my home computer. :)

There are two options I would suggest:

The first (and best IMO) would be to model this as a regular dual-carriageway intersection. (https://imgur.com/a/DnCXKWe) It's a little bit awkward because the two parts of Homan Street don't align, but modelling the intersection in this way would allow all the required vehicle movements with accurate descriptions: Vehicles continuing on Homan would get a "continue straight" message or similar). Bonus is that this woud make it a lot easier to introduce no_u_turn restrictions where needed.

Alternatively, you could model it as two seperate T intersections. https://imgur.com/a/R0kdHK0 This accounts for the misalignment in Homan Street. Vehicles travelling along Homan would receive a "Turn Right then turn left" message". I wouldn't choose this option, as the turns are so close together to make it a bit unnecessary, but it would easily get the job done if you prefer that as well. :)

Dian

Incidentally, very interested to see the new bridge they are building there. Not sure if you are a local or a tourist there, but I can remember getting stuck in traffic there during summer school holidays!

117344410 almost 4 years ago

Hi Supt,

Yes, I can see that there are new traffic islands at the intersection on the Maxar Imagery. However, the left hand turn from Warren onto the northside of Redman is not physically separated: way/1030856092.

There is also no physical separation that requires way/1030856098 or way/1030856097; both of those right hand turn lanes create an extra set of intersections that don't exist.

117344410 almost 4 years ago

Hello Supt,

This got flagged on OSMCha as "Crossing Ways".
The way way/1030856092 doesn't seem necessary as there is no physical separation there. Similarly, you're modelling of Warren/Homan street has unneeded turn lanes that break routing.
---

Published using OSMCha: https://osmcha.org/changesets/117344410

116207837 almost 4 years ago

Hello;

I was in the area and swung by (there really isn't a substitute for going in person sometimes :) )

As mapillary seems to be having issues with iOS uploads, see this pictures fresh off the presses:
https://commons.wikimedia.org/wiki/File:Seaford_Foreshore_Trail_at_Station_St.jpg
https://commons.wikimedia.org/wiki/File:Seaford_Foreshore_Trail_at_Riveria.jpg

The signs looked new-ish, but not having been before I can't tell how long they have been there. I also didn't walk the entire track, so it's possible that there were sections signed differently.

Then again, the ground really was quite sandy....

Dian

117144673 almost 4 years ago

No problem

I hope you are able to sort out what the error was :).

117144673 almost 4 years ago

Hey Zulu;

Came across this node (9488792036) while updating the school next door, and I can see this changeset has a lot of "knooppunten generator" in unusual locations. What are they?

117099145 almost 4 years ago

Hi Cheena,

Thanks for your edits. What is the source of increasing the speed back up to 110 km/h? Another user only recently reduced it as it is a new limt: way/199164004/history
---

Published using OSMCha: https://osmcha.org/changesets/117099145

117046200 almost 4 years ago

Hey Chonker,

Similar to the comment on the other edit, in this changeset you've moved the point where 103401900 splits from the main road north quitte a bit beyond where the physical separation ends, and "curled" the turn slightly as it arrives at the left turn.

Moving the point where the carriageway splits will cause routing software to give mistimed turn instructions to drivers; effectively asking them to turn before the "actual" turn. In this case, there isn't another maneouvre possible to confuse them, but it can cause issues with driveways etc.

Similarly, the "curl" onto the main carriageway isn;t really ideal. It makes the anngle of the slip road look tighter than it is, and the turn onto the main road tighter. This can affect calculations/routing for long or articulated vehicles, or with restricted turning radii. I'm happy to be incorrect, but I do not believe there is any utility to modelling turns like this beyond aesthetics.
---

Published using OSMCha: https://osmcha.org/changesets/117046200

117065625 almost 4 years ago

Hey Chonker!

Impressive detail work, as always.

I'm curious though, what was the reason for splitting the carriageways at Partridge Cres and Buxton Cres, as there doesn't seem to be any physical separation?

Dian
---

Published using OSMCha: https://osmcha.org/changesets/117065625

116986338 almost 4 years ago

Hey Immortal Boredom!

Thanks for the edit around the McClelland Drive area!

I noticed the way that you've modelled the roundabout at McClelland and Skye means that traffic from the forestry area can't go onto the roundabout; they would be forced to turn left. Would you be able to link the road with the roundabout junction?

It has also left a traffic island node sitting on it's own that can probably be deleted. :)

Dian
---

Published using OSMCha: https://osmcha.org/changesets/116986338

116986338 almost 4 years ago

Hey Immortal Boredom!

Thanks for the edit around the McClelland Drive area!

I noticed the way that you've modelled the roundabout at McClelland and Skye means that traffic from the forestry area can't go onto the roundabout; they would be forced to turn left. Would you be able to link the road with the roundabout junction?

It has also left a traffic island node sitting on it's own that can probably be deleted. :)

Dian
---

Published using OSMCha: https://osmcha.org/changesets/116986338

116889786 almost 4 years ago

Thanks for confirming Cassie.

I have moved the node for the physiotherapist to their new address.

Dian

116888548 almost 4 years ago

Hey Leon,

This appears to be a descriptive name you've added to this drivethrough. It's not a street with that name gazetted. osm.wiki/Naming_conventions#Don.27t_describe_things_using_name_tag

I've removed it accordingly.
---

Published using OSMCha: https://osmcha.org/changesets/116888548

116889786 almost 4 years ago

Hello,

It appears that you were trying to relocate the physiotherapist that was at this location; you've accidentally changed the address of the building where they presumably used to be.

I will correct it for you now, let me know if you have any other questions or concerns.

116517384 almost 4 years ago

Hey mate,

I've done some research and now know much more than I did previously about helicopter operations!

The markings on the east of Bentinck Street are a "aiming point": https://www.icao.int/APAC/Meetings/2016%20Annex14VII/ICAO%20Bangkok%204-2016%20Visual%20Aids.pdf

Helicopters don't set down there, but descend until they cross the road. I've altered the tags for the aiming point accordingly. :)

I've also added an additional tag to the taxiway: taxiway=air, to indicate that this is a "special" taxiway.

Let me know what you think. :)

116517557 almost 4 years ago

No problem at all.

It's okay if you don't feel like this particular place is suitable for the emergency landing site tag; there is always some grey in the wiki. :) I appreciate you saying that some of the others can be retagged.

OSM is addictive. :) If you are feeling social, I'd encourage you to join the talk-au mailing list. It's a cool group of people. https://lists.openstreetmap.org/listinfo/talk-au