I am fine with using the junctions of the ramps if the cross road goes over and the distance is reasonable like shown above. When the road goes under and the ramps are longer like this I like to see it broke at the bridges not the ramp junctions.

https://goo.gl/photos/h9rju97rwd3MsAYo9

Lance (Sdtahoe)

Fredo-p wrote:I beg to differ.

When we asked the staff at meetups to help overcome the elevation issues, 0 was well, what is 0 when it came to editing the map. We assume it can be one thing. So to put a more definitive definition to it that many and most could relate, Ground is the best solution. It means that it essentially is on solid earth, there is no segment passing under it or it is not passing under another segment. It was to mitigate the elevation conflicts that were encountered between segments, to map more in reality and not some figurative situation.

Even that didn't help solve complex interchanges where on, an uninterrupted segment by means of next junction node, happened to go 'over, under, over and over and under' several times before the next junction.

Hence why the Seagull rule helps further mitigate this situation and scenario by allowing the segment to be more reflective of reality with the bridges being represented and the segment could then have that one elevation change over the other segment. Which it will also help to prevent crossing segment MP conflict reports that it believes there should be a junction node, if they happen to be at the same elevation because the long segment happened to be unnoticed. Instead of 8 elevation bumps because of the 10 different segments below it, it can now be maybe more realistically 1, maybe 2 or 3. But it doesn't get fictitiously elevated because the segment is now crossing so many others as it had before at the various elevations.

In the end, elevation setting was never to be indicative of slope or grade change. You would not use elevation change to reflect the 6% grade change along I-17 between Phoenix and Flagstaff. Those segments are on the ground, unimpeded by other segments crossing under or over it, hence why it is on the ground setting. If it had a bridge along the way, preferably in my interpretation only when another driveable segment crosses underneath, than yes, place that brief segment at 1. This way, 10 miles of segment aren't again, fictitiously raised because of one bridge. Which if another driveable segment crossed it 8 miles down the segment, it would not have to be at 2 because the main segment was now 1. Following the Seagull rule it would be 1 because it only had to cross the one segment at ground elevation setting now.

Now, if Seagulls were flying into the ground, borrowing under and flying out the other side... we might need a new definition of what ground was, and study this new breed of Seagulls.

dmcrandall wrote:Interesting. The OCD part of me won't let elevation go on for miles and miles. I've always clipped overpasses to reasonable lengths, mainly because I don't like the colors TB uses for elevations.

I've always edited with a road type heirachy in mind. The highest level road type was almost always "ground level." Anything that crossed over, or under, it was shown as +1 or -1, and the segment was cut to a reasonable length so that the elevation change only covered the actual overpass area (with a little leeway on either side).

This talk of miles of continual +1 elevation on roads is incomprehensible to me.

There are plenty of rural situations where this arises, where the old paradigm of using elevation to represent the segment above or below (which we later determined that -1 are tunnels), can be for long distances when there are not exits for 15 miles but it may still cross over segments along the way because it was cheaper to build the smaller road under instead of over.

I clip the bridges at the expansion joint. As long as I am not leaving segments <15m I believe all is well and makes for pretty accurate and realistic representation, though I know that it isn't the communities goal explicitly, it has helped with those ultra complex interchanges with high over passes when it comes down to it.

Fredo-p wrote:

the1who wrote:In the end, elevation setting was never to be indicative of slope or grade change. You would not use elevation change to reflect the 6% grade change along I-17 between Phoenix and Flagstaff.

Fredo-p wrote:But that part of the road is also elevated. I'm not asking that any road that goes up in elevation be mapped. I'm saying that the ones that begin to rise and fall as a part of an overpass should be mapped in its entirety. That is mapping to real world.

Understand you more clearly now. I was simply stating for the rule of the clarification. I still don't think that the elevation change because of the geographical rework to make an overpass should be marked as 1, since it still has earth below the road segment and a Seagull can not fly under, which is why I prefer the expansion joint placement.

I don't normally see bridges mapped the other way, and in most cases when it was simply an indication of a bridge, it was due to the lack of resolution that the map was printed on, meaning the level of detail was already zoomed out as it was.

Fredo-p wrote:

qwaletee wrote: I'll repeat, we have information from the devs that roadways with a positive elevation WILL be drawn as bridges. Ground and negative will not, nor will they be drawn as "tunnels" unless teh new tunnel attribute is ticked.

Can you link to where this information is?

Fredo...

I know, like many would, you want very specific, definitive information from staff in "quotes" but sometimes that is physically and practically impossible for most cases. On many cases, the details and discussions involved were on very proprietary means. We have had discussions with devs on a 1-on-1 level a time or two. That is their goal and intentions. I don't see why you have to have such a dramatic "quote" stating otherwise, when champs like PZ, I, and others have expressed the intentions. Continually asking us to populate that information is really not your place. With that being said, when we see room for improvement, we aim to make that move.

As far as chopping up freeways. I don't find it necessary as I have told the Plains SMs and in relevant chats that it would be necessary for roads to be chopped because it crosses a stream or just because a Seagull can fly under it, such as over non-driveable. On the other hand, I expressed, if a driveable segment does indeed go under it, it should be raised, the one on the ground will remain as ground, regardless of the type of road classification. That in effect, exemplifies the Seagull rule. I also stated that I do not find it necessary to also chop up those segments for non-driveable segments, but to have the non-driveable go up or down around the other driveable segments affected. Such as trails and railroads.

dmcrandall wrote:

Fredo-p wrote:When I see the word ground, I see ground level.

If the road is attached to the ground...even if it's an artificial incline, that is still ground level.

Stop trying to split hairs.

This is my understanding. The purpose of elevation is to separate roads that intersect but do not create a route-able intersection due to differing elevations. This new policy (or guideline) takes it a step further to say that if a seagull can fly under it, it's a +1 and so on... Elevated ramps that are built up on raised Earth are still ground just like the roads that take you through the Rockies. Note that I-70 west of Denver is all ground even though you top out at over 11,000 ft where you enter the Ike Tunnel which becomes -1 elevation because it's below ground.

Normally no one would care when the elevation started or stopped as long as the intersecting roads were included where applicable. This is why you see miles and miles of Interstate at a +1 to accomodate a ~300 ft section of overpass. It's also why we see roads at a -1 that pass under an freeway where the freeway is at ground (when it should be +1). I believe this looks tacky in the livemap (but that's probably just the perfectionist in me). I welcome the policy and prefer the more exact elevation drawing, except of course when it's already close to an intersection (Fredo's first example).

Understanding why we used to do things the way we did them, it seems like Waze now has some use planned for the elevation setting. The screenshot is from one of our typical freeway intersections where the "ground level" segments are set to -1 to keep from having to adjust the freeway elevation. Now they oddly appear underground with how the livemap now renders.

So if we don't base our decisions on the guidance from Waze, or how it's affecting the map rendering, what should we base them on?

Fredo-p wrote:The dash lines mean underground? I bet you if you polled as many waze users as you could, they wouldn't know what it meant nor care.

I think it's convention that dashed lines indicate the road is behind or under something, in this case the ground. Regardless, the representation implies there is a change of grade for those road segments when there is not.

Merry Christmas Fredo-p, go get that stupid bird!

https://i.imgflip.com/wbw03.jpg