Get a sneak peek at whats next for Permanent Hazards on our April 7th Office Hours!
Discussion for the unofficial, community-developed addons, extensions and scripts built for the Waze Map Editor.

The official index of these tools is the Community Plugins, Extensions and Tools wiki page.
Post by CBenson
ctpoole wrote:I agree with TXEMT. If I put in a node I eliminate the same segment warning but in its place I get an unneeded node warning. I tried to question this once before but didn't communicate the problem well.
Could you provide an example permalink? I do not get this behavior from the script.

It's been recommended since I started editing that no two segments should have the same two end points. The guidance has been to split one of them so there are always at least three segments involved in any "loop." If you don't do this several minor problems can occur. First if you start or end a route on a segment that shares its endpoints with another segment, the route line may display on the other segment. Second, routes over such a segment may show using the other segment in URs. Third, if you deviate onto the other segment, it seems that waze does not recalculate the route (admittedly not a significant problem as you end up at the same point anyway).

I believe that somewhere it was explained that these issues are caused by the routing engine identifying routes by junction nodes rather than segments. Thus if you have two segments that connect the same two junction nodes, when the various display engines get the instructions from the routing engine the route is ambiguous. However, I can't remember or find who posted this explanation.
CBenson
EmeritusChamps
EmeritusChamps
Posts: 10330
Has thanked: 608 times
Been thanked: 1642 times
Send a message
Regional Coordinator: Mid-Atlantic, US
Verizon, Nexus 6, Android 6.0.1, Waze 4.7.0.902

Post by CBenson
So I put a node on Vanda Ter. Validator is not showing an error for having identical adjacent segments (or any other error) to me. Is showing an error for you?
CBenson
EmeritusChamps
EmeritusChamps
Posts: 10330
Has thanked: 608 times
Been thanked: 1642 times
Send a message
Regional Coordinator: Mid-Atlantic, US
Verizon, Nexus 6, Android 6.0.1, Waze 4.7.0.902

Post by CBenson
Well sure. When I started many were deleting nodes from the end of dead end segments using essentially the same logic. Then waze came in and said the end nodes were necessary. If the node serves a purpose, its not an extra node. Same here, when there have been complaints about routing to loop roads, waze says the solution is to add the "extra" node. If we only mapped primary streets and up, then there would be fewer places we'd have to check for errors. The goal here isn't to have fewer places to check for errors, its to have a map that works the best we can make it work.

Seems to me you can either change the your clean up process to comply with the wiki or get waze to change the way it handles multiple segments that connect the same two junction nodes.
CBenson
EmeritusChamps
EmeritusChamps
Posts: 10330
Has thanked: 608 times
Been thanked: 1642 times
Send a message
Regional Coordinator: Mid-Atlantic, US
Verizon, Nexus 6, Android 6.0.1, Waze 4.7.0.902

Post by CBenson
We were changing self connectivity at dead ends before the direct ability to edit them was in the editor and most seem to be continuing to do so now. If the underlying routing issues have not changed, I don't see why the validator check should change. Its no different than warning us about reverse connectivity on one way roads, which can't be directly edited in editor either.
CBenson
EmeritusChamps
EmeritusChamps
Posts: 10330
Has thanked: 608 times
Been thanked: 1642 times
Send a message
Regional Coordinator: Mid-Atlantic, US
Verizon, Nexus 6, Android 6.0.1, Waze 4.7.0.902

Post by CBenson
PesachZ wrote:Only because pass asked is to stop disallowing the u-turns at dead end nodes, so that they can see if their fix is working, and tweak it to work better anytime, its found to have failed.
Has staff actually clearly stated this, because if so I missed it? I continue to see the same issues I always have with u-turns allowed at dead end nodes. I also see no change with those that are currently disallowed. If turns aren't allowed now and they have removed the ability to directly edit them, then how do they expect the turns to be allowed for testing? There was some talk about them being all allowed with the removal of the ability to edit, but that doesn't appear to be the case to me.
CBenson
EmeritusChamps
EmeritusChamps
Posts: 10330
Has thanked: 608 times
Been thanked: 1642 times
Send a message
Regional Coordinator: Mid-Atlantic, US
Verizon, Nexus 6, Android 6.0.1, Waze 4.7.0.902

Post by CBenson
Well imho there nothing clearly stated in the beta threads about what is really happening with U-turns.

I would also note that I'm incorrect that there there is no ability to dissallow U-turns at a dead ends. Even without JNF, Q seems to dissallow the reverse connection at a dead end. Although, oddly, if they are looking to have the dead end reverse connections enabled, W does not seem to allow the reverse connection.
CBenson
EmeritusChamps
EmeritusChamps
Posts: 10330
Has thanked: 608 times
Been thanked: 1642 times
Send a message
Regional Coordinator: Mid-Atlantic, US
Verizon, Nexus 6, Android 6.0.1, Waze 4.7.0.902

Post by CBenson
Taco909 wrote:Unless we are in the UK, we are not supposed to do this anymore... we've been asked in multiple threads by various global champs and Waze staffers.
Validator needs to ignore the u-turn, and the "Q" function in JNF needs to be changed so it likewise ignores u-turns if they are at a dead-end.
I continue to disagree. The waze stance seems to be that they will* enable all the u-turns at which point they would like them to remain enabled to confirm that the short detour prevention mechanism does solve the problems with enabled u-turns at dead end. First, the whole point then will be to see if problems are due to the u-turn being either enabled or not. As they have taken the ability to see whether the U-turns are enabled away from us in the WME interface, I find it now somewhat critical that this indication remain in Validator so I can troubleshoot and report problems with enabled or disabled dead-end roads. The point is if waze is going have routing affected by a property that can be changed by editors, then we should be able to see that property.

* If "soon" means "we might get around to it someday" then "will" means what?
CBenson
EmeritusChamps
EmeritusChamps
Posts: 10330
Has thanked: 608 times
Been thanked: 1642 times
Send a message
Regional Coordinator: Mid-Atlantic, US
Verizon, Nexus 6, Android 6.0.1, Waze 4.7.0.902

Post by CBenson
More or less. Basically my position is that it absurd for routing to depend on some setting that we can inadvertently change (say by disconnecting a road that has become a dead end and leaving a rev con in place) where we have no way to see that property. What waze should do is not make routing decisions based on the condition of the reverse connectivity at dead-ends. In essence if all dead-end u-turns should be enabled, then they should remove the ability to have disabled u-turns in the model. If that's too much of a change at this point, they should completely ignore the property. Then we could ignore it too. But they don't. The idea that good routing or routing to a stop point on dead end roads is going to depend on having the dead-ends reverse connections enabled when they have removed the ability to easily enable them does seem absurd to me.

So yes, it would be better for Validator to highlight the disabled dead-end u-turns.
CBenson
EmeritusChamps
EmeritusChamps
Posts: 10330
Has thanked: 608 times
Been thanked: 1642 times
Send a message
Regional Coordinator: Mid-Atlantic, US
Verizon, Nexus 6, Android 6.0.1, Waze 4.7.0.902

Post by CBenson
ditchi56 wrote: - there's no reason to remove the link to Google maps since Google own Waze, they're not going to object to their own property being enhanced?
To the contrary there is no reason to believe that all of the Google map data suppliers entered into agreements that include using the data for waze.
ditchi56 wrote: - there's no reason to remove the link to OpenStreetMap since that prides itself on being copyright-free?
OpenStreetMaps is certainly not copyright free. See here and here.
CBenson
EmeritusChamps
EmeritusChamps
Posts: 10330
Has thanked: 608 times
Been thanked: 1642 times
Send a message
Regional Coordinator: Mid-Atlantic, US
Verizon, Nexus 6, Android 6.0.1, Waze 4.7.0.902

Post by CBenson
sketch wrote:
Taco909 wrote:Likewise, there may be some cases where it is desired to execute the turn instruction early, but to overlay the departing segment onto the feeder segment for a short distance.
As for the early instructions, IMO it's better as an editor to lay the departing segment just next to, but not directly coincident with, the continuing segment.
I agree in part.

My understanding is that if you overlap segments, no turn instruction is given. So overlaying a departing segments on the road it departs from to achieve an early instruction I think is futile. So I agree that such segments should not be directly coincident with the continuing segment.

I would go further however as In my personal opinion angles of less than about 20 degrees should be avoided whenever possible. My reasoning is that the road closure feature in the app is next to useless for intersections with small angles. But there is the question of whether we should be mapping to the make the road closure feature work or whether the road closure feature should be made to work with the map that conforms to reality.
CBenson
EmeritusChamps
EmeritusChamps
Posts: 10330
Has thanked: 608 times
Been thanked: 1642 times
Send a message
Regional Coordinator: Mid-Atlantic, US
Verizon, Nexus 6, Android 6.0.1, Waze 4.7.0.902