If I’m not mistaken, the usage in the first sentence has been in place for at least two years, if not more. It’s interesting that we spend voluminous amounts of time discussing the Junction Style Guide, which incorporates all manner of “road-meetings”, including both simple intersections and roundabouts. I won’t be any more of a jerk by quoting more Glossary statements, the wiki is fluid, after all, but it seems to me that, once upon a time, geopoints were referred to as “geometry handles” in the wiki, were they not? I like “geopoints”; I think we should move that direction.
“Node” is not uncommon in mapping. My Synchro software refers to junctions simply as “nodes” and assigns a unique number to each one. There is no other “node” reference in that program. The handles that force a curve into a segment are simply handles… and we only get two of them so getting the geometry correct without having some REALLY funky junction geometry can be a challenge.
The dogleg that I showed you at Winona was a very simple curve segment… I should have shown you some of the offramps.
The wiki usage predates any understanding of the internal data model.
IMO all the Junction stuff in the wiki should really just be replaced with “Intersection” to remove the conflict with Junctions in the data model which are something completely different. Also the standard shorthand name for geometry nodes is geonode, but since OpenLayers calls them points (because they are ), geopoint works.
No Junctions are intersections in the data model, they’re all what we call roundabouts, traffic circles or rotaries. It’d be more like saying your knees are fingers
Some clarification on this… Does this also apply to non-drivable segments?
Pedestrian boardwalks that cross at grade, railroad, etc… Should be be elevating these non-roads?
This is Waze, and we need to be able to communicate clearly amongst ourselves without ambiguity about the different functions of the Editor.
I’m opening a new thread in the Wiki forum to discuss this, as we’ve now sped at about 900 mph directly away from the original topic and issues raised in this post. (Thanks, me! :oops: )
Exactly what I’m saying. I see this atleast once or twice a week.
As far as doing non driveable segs at -5. You bring up a good point. We can’t wiki that they need to be -5 to avoid navigation issues, and in the next breath propose that elevation has no affect on navigation…
I understand the lock. But there is also in WME chat, the unlock forum and also us here monitoring this thread. There’s multiple resources to ask someone to unlock or fix a locked road.
The plot thickens. My home in Cordelia to Mammoth no longer works, and it’s too soon for the Green Valley / Lopes Rd overpass elevation to have taken effect. I did not try it immediately before I changed it.
NOW, you’re beginning to see what I saw the other day. There is inconsistency in routing all along this stretch of I-80E (I started around Rockville). If you try it a few more times, it will alternately work or not work, from exactly the same spot. At least that’s what I was experiencing.
As far as the 4 is concerned, the theory is that it shouldn’t matter what Elevation you set it to, as long as it’s different from the roads it crosses.
Edit: I should clarify that there is inconsistency at least as far as routing to Mammoth is concerned. I didn’t do much experimentation with other destinations.
I’m becoming more and more convinced that this is a pruning/complexity issue:
The routing typically takes upward of 30 seconds, when it works (for comparison, routing the other direction typically runs about 6 seconds, due to the relative lack of choices leaving Mammoth)
Time of day used for routing in Live Map seems to have an effect on success of routing. I found that early afternoon tends to have the most difficulty here.
Only one route is ever presented when searching from this stretch of I-80E to Mammoth, as opposed to multiple routes when starting from a bit north, typically splitting somewhere around Sacramento, or coming from somewhere in the town of Fairfield, which sometimes offers a route to the south.
Only one of many examples:
I just routed successfully from west of Green Valley, through the Suisun Parkway exit area, at the current time, 5 PM. Change nothing other than the routing time, to 2PM, routing fails.
Things also change behavior as the destination moves north of Mammoth on the 395. Multiple (northern) routes begin to show up, even though there is really only one functional route, 395, between there and Mammoth.
The last thing I’m going to say about Elevations is that if there were truly an issue here, I would have expected to see a failure when routing anywhere east on the 80, and that didn’t happen; Davis, Sacramento, and Tahoe destinations all routed fine from that Green Valley spot, even before the elevation change.
nnote, I would consider it a kindness if you would please send me a solid example of elevation change failure, as described above in the A/B/1st/2nd example, next time you run across it; I’d love to see it in action. There’s just WAY too much wiggle here for me to confirm that Elevations were/are an issue in this vicinity.
That would the developers (and every math- or computer science-oriented editor). They’re all nodes. A node just means a connecting or ending point in a graphing system or a tree.
Kind of like saying “place” which could mean Area Place, Point Place or both, depending on context and qualifier.
Except for intersections that are built of multiple junctions. In which case, not every intersection is a junction, and not every junction is an intersection. But you need junctions to build intersections, while you don’t always use junctions exclusively to build intersection.
To confuse further, the common definition of intersection is more restrictive than the engineering definition. To an engineer, a ramp split is an intersection, because the roadways “intersect,” while in common usage, that woud not be called an intersection.
), geopoint works.