CBenson wrote:A couple of thoughts.
1) I didn't say they weren't roundabounts, just that they weren't worth mapping.

Perhaps not to you.

2) Although yielding to the traffic is a consideration, its not a universal quality of traffic circles. There are certainly urban signaled circles where "at the roundabout take the second exit" is a useful instruction, but the traffic in the circle doesn't have any right-of-way over the traffic entering the circle as the entries are all signaled.

Actually, one of the defining characteristics of the modern roundabout is that traffic entering yields to traffic already in the circle unless otherwise marked (at least, within the USA; obviously, I can't speak regarding other countries). (By "otherwise marked," I mean that there will be signs in the circle directing traffic to yield to the entering traffic.) A "modern roundabout" is a type of looping junction in which road traffic travels in one direction around a central island and priority is given to the circulating flow. Signs usually direct traffic entering the circle to slow and to yield the right of way. See this document from the Federal Highway Administration for further information.

I even found the following image that kind-of illustrates the whole roundabout thing:

http://i.imgur.com/AlknQTN.jpg

While it's true that US dictionaries use "roundabout" and "traffic circle" as synonyms, in technical terms, they're different.

----------------------
Common distinctions between modern roundabouts and older rotary type intersections:

Typically, modern roundabouts are:

• smaller than rotaries
• designed for slower entry, circulating, and exit speeds
• always following a “yield-at-entry” traffic control principle
• designed with a raised splitter island to slow and deflect traffic prior to entry
• designed to facilitate safer pedestrian crossings
• designed to follow a same lane entry/lane exit principle at multilane roundabouts (NO LANE CHANGES in the circulatory roadway)

Signalized Traffic Circles are NOT Roundabouts. As an example, Dupont Circle in Washington DC is not a roundabout, it's a Traffic Circle with signals (I.E. a Signalized Traffic Circle).
(Source)
-------------------------
As with anything in life, there are bound to be exceptions to the above - however, they're just that: Exceptions, not the rule.

I realize that this is possibly going beyond the scope of what needs to be considered from a Waze "mapping it" standpoint, but I feel that the point needs to be made - after all, we're talking about mapping roundabouts and whether or not the Validator should be annoying us with non-warning about a two-entering road roundabout. (Also, I felt that the distinction between the two should be pointed out, even if the map software doesn't give us a way to map them differently.)

I'm sure there are those who might say that there's no difference from a mapping standpoint since we only have a "Add Roundabout" function within the WME. I expect you're probably one of them. However, if a distinction needed to be made, I'm thinking that for those editors who have the Toolbox installed, the "Change roundabout to standard road" (which would still be a one-way circular road) tool could help with that and thus a signalized traffic circle could be represented that way. I'm not advocating this; just suggesting it as a possibility if it were felt important enough to treat them differently in the map.

CBenson wrote:

SuperDave1426 wrote:One question, though, and I didn't see an answer to this the last time we were discussing it: You had indicated that in the past you've seen that to be the case. Is it, in fact, still the case now? Have you seen any recent occurrences of whatever the problem behavior is that you've seen? It could very well be that it's already been fixed.

I have not seen recent occurrences, so roundabouts may now be treated differently with regard to this issue.

Then can I make the request that when you're arguing against two-road roundabouts that you drop that as a reason? If it's no longer happening, then it seems to me that it's no longer a valid reason to use in a "do it or not" type of consideration. But maybe that's just me.

CBenson wrote:

SuperDave1426 wrote:Then can I make the request that when you're arguing against two-road roundabouts that you drop that as a reason? If it's no longer happening, then it seems to me that it's no longer a valid reason to use in a "do it or not" type of consideration. But maybe that's just me.

Sure you can make the request. But if I understand correctly, the validator is simply applying the same rules regarding two segments connecting the same two junctions that is applied to any segments to roundabout segments.

Actually, no it's not doing that anymore. That check was disabled (at least in the US) a couple of versions ago.

I do believe that there are currently problems when the same two segments connect the same two junctions. I guess I'd like to see more evidence that roundabout segments are somehow treated differently with regard to these problems.

Time will tell.

A feature request of my own.

Is there any chance that a check can be added to highlight a road with a blank elevation? I've been running into two situations:

1) As I'm editing, I'm discovering blank elevations that I'm having to set as ground and then save, when then shows "ground" the next time I select those roads. This then results in my checking a lot of extra roads nearby that I'm not otherwise planning to edit to see if they've got the blank elevation as well. If the Validator highlighted empty elevations, then I'd be able to simply look around and know which ones need to be fixed.

2) I'll edit a road that already has a elevation of ground, but after I've made my edit(s) (none of which involve elevation), when I select the road again it will now be blank instead of "ground." This does not happen all the time - only for some roads. As a result, I'm wasting a lot of time editing a road and then having to select it again after saving to make sure the elevation is still right. If it is, then I move on and if not, then I change it back to ground and resave. If the Validator highlighted empty elevations, then I could eliminate the need to click on something I just finished editing simply to ensure that it's still set to "ground" and will only have to do it if it lights up after I save. (I sometimes have dead-end U-Turns that were green where the save at hard restricted didn't "stick" for some reason and reverted back to green after the save - since the highlighter lights up those roads, know I need to go back and do it again without having to waste time reselecting every dead-end I edit just to be sure the restriction is in place.)

So, any chance this check can be added? Maybe make it a selectable option so that those who don't care won't be pestered by them?

BellHouse wrote:

SuperDave1426 wrote:Is there any chance that a check can be added to highlight a road with a blank elevation?

Blank elevations do not exist in the database, they are just sometimes displayed in WME by a long standing bug that's masked by the Junction Node Fixer Script (JNF). It happens after saves and the features are reloaded before the changes have fully propagated. Causing the features to reload (reload page, permalink, pan/zoom far enough, WME Toolbox key) clears it.

Ok, thanks for the info. Question: How does that account for times when I've gone into an area, and even before editing anything I've got empty elevation segments showing? I've tried zooming, moving around, etc., and it still shows empty instead of "ground."

By "masked by the JNF," do you mean that when I've selected a segment that was showing ground but then ended up as empty, that the JNF was masking the out-of-range elevation and thus it was showing as "ground" even though it wasn't actually that way in the database? And that for an apparently short time after I saved an edit it was showing as empty again until the JNF catches up to it? I'm just trying to make sure I understand what you're describing above.

If that's the case, then maybe the validator could be looking for things that are out of range and either give you a one-button fix or highlight them so that they can be looked into? The JNF may be making it so that out-of-range looks like "ground" in the editor (if I'm understanding you correctly), but that doesn't change the fact that the value is wrong in the database under those conditions....

berestovskyy wrote:I couldn't find any blank elevation around, but I found few segments with out of range elevation (i.e. level 20) and WME displays it as a blank. Here is an example: permalink

Based on my understanding of what BellHouse said in his post, that's probably what I've been running into.

That being the case, is there a way that you can have the Validator check for things like that or, better yet, give us a way to quickly fix them when we encounter them?

berestovskyy wrote:19.02.2014 v0.7.0:
- DISABLED for US 'Two-way Ramp segment'

Can you make this an option that can be turned on/off, so that those of us who live in areas where two-way ramp segments simply don't happen won't miss one that gets created by mistake?

berestovskyy wrote:

SuperDave1426 wrote:Can you make this an option that can be turned on/off

You can "turn off" any check you want using 'Reported as' field in the "Search" tab.

For instance, to disable #40 'Soft turns' highlighting/reporting you may put the following:

Code: Select all

!40,*

Hover your mouse over the field for more examples.

Unfortunately, there is no way to force "turn on" a check once it has been disabled for your country

Then how about re-enabling it and letting the people who don't want the check to use the above trick to turn it off? That way, everyone gets what they need from it.

IMO Two-way Ramp is not a error and at the moment the check is enabled just for one country.

I'm sure that will be a great comfort to the first person who gets routed the wrong way onto a freeway because an editor didn't see that it was allowing traffic in both directions.... (Yes, I'm being somewhat facetious here.)

Has there been a change made to the way that the Validator handles roads marked with (CONST ZN)?

I ask because this morning when I checked on an area that I manage, an road that's marked that way lit up like a Christmas Tree.... I thought that construction zone roads would be not checked/highlighted. Did that change?

AlanOfTheBerg wrote:

SuperDave1426 wrote:Has there been a change made to the way that the Validator handles roads marked with (CONST ZN)?

The blue Note reminder was added because there are often old construction zones for which the segment named never were changed back. I found two yesterday in different areas which I wouldn't have found before without this feature.

This was lit up in red complaining about no inward/outward connectivity, which really kinda made it stand out. Blue would have been less annoying.

Example.

AlanOfTheBerg wrote:

SuperDave1426 wrote: This was lit up in red complaining about no inward/outward connectivity, which really kinda made it stand out. Blue would have been less annoying.

Example.

Ah! So, it wasn't red because of CONST ZN, but because of the connectivity issue.

Perhaps there should be an exception for the inward connectivity for segments which are marked CONST ZN then. That way, if happened upon or in a scan, they are "caught" and user alerted for the CONST ZN with a BlueNote and not the more serious RedNote.

Yea, something like that. The odd thing is that it used to light up, then the change was made that any road marked with CONST ZN wouldn't be checked (or whatever it actually was) and then that road stopped showing up highlighted. Then out of the blue it light up that way.

Hopefully something can be adjusted.