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 SuperDave1426
sketch wrote:What I mean is that the box should be unchecked by default in both editors. Otherwise, users who are not aware of the change when Validator auto-updates may inadvertently start drawing 2-way segments where 1-ways are needed.
I see what you're saying in the rest of the message; however, given the ratio of two-way to one-way streets (and the fact that WME runs opposite of that, requiring you to have to remember to go change the direction after you draw a road), I'm not sure that's a problem.

Someone who's used to it will click on it, see the direction is already two-way, and move on after wondering about it for a minute or two.

I admit that I certainly don't know about where you live, but here on the west coast the vast majority of roads are two-way, not one-way. It seems to me to make a lot more sense to have the editor default to "two-way." (I'm constantly having to deal with roads that should have been two-way, but the person who created them didn't bother to change the direction from the default "one-way" setting.) Failing that, having it be a feature of an extension is nice.

People have their preferences, though. I wouldn't mind it being a clickable option, so long as it's one of the options that will "stick" so that you don't have to keep going back to change it to what you want, if you happen to want the opposite of the default setting.
SuperDave1426
Country Manager
Country Manager
Posts: 1047
Has thanked: 97 times
Been thanked: 325 times
Send a message

Post by SuperDave1426
berestovskyy wrote:
sketch wrote:Should be 10 km at the minimum. Rural and even suburban freeway segments can be much, much longer than 1 km.
There is a minor technical issue with long segments: if a node is out of the screen, WME does not load it. So Validator does not even trigger the group of checks which analyze both end nodes (for instance, "No inward connectivity" and 'Same endpoints')

If no complains on that - I'll increase the length limit.
Can I put in a request for a clickable option, then, that says something along the lines of "show unneeded junction nodes regardless of segment length?" :D

A lot of the stuff I edit is less than 1K in length and I want to get unneeded junction nodes off the road whenever possible. Having this tool call attention to them has been a big help. I'd really hate to lose that.
SuperDave1426
Country Manager
Country Manager
Posts: 1047
Has thanked: 97 times
Been thanked: 325 times
Send a message

Post by SuperDave1426
sketch wrote:It would be nice to have the extension give a pop-up warning when a new version resets all your settings, btw.
That would be nice, yes. :D
SuperDave1426
Country Manager
Country Manager
Posts: 1047
Has thanked: 97 times
Been thanked: 325 times
Send a message

Post by SuperDave1426
berestovskyy wrote:
ituajr wrote:Can roundabouts be excluded from the "same endpoint" test, please?
It was discussed here: page 7 and we decided to leave them. Let me know if I shall exclude roundabouts for you country.
I was involved in that conversation, and with all due respect to the others who commented on it, I remain unconvinced that currently there's any real probability of someone being routed down the wrong way in a raoundabout that only has two roads going to it (which is the only thing that triggers this particular alert in the Validator). CBenson gaven an example during that discussion of a toll road where toll was one way but not the other (on two one-way roads) and so people with "avoid toll" set would be routed down the wrong way street because it had the lower penalty. Since I've never seen a roundabout where it was toll in only one segment, I fail to see how the segment going the wrong way, which has a higher penalty than going the segment the correct way, is going to be the route that the navigator selects. Maybe there was a problem with it at one time, but I haven't seen anybody cite anything current where someone has been routed the wrong direction on a roundabout.

The "add roundabout" function in WME doesn't create an extra, third empty junction node and the Toolbox "redo roundabout" function doesn't do it, either.

Adding a third node by way of the usual trick (connect a blank road and delete it) results in a roundabout which no longer has sequential segment ID numbers, and that has been demonstrated to cause routing problems occasionally. Fixing that with the Toolbox "Redo Roundabout" tool properly removes the extraneous extra node because, well, there ain't no road there. :-)

I really think that this decision to have a two-road roundabout flagged with that "same segment" warning needs to be rethought. As it stands, I feel this same-segment warning on two-road roundabouts is going to encourage editors using the validator to create more problems on roundabouts than the alert is trying to solve.
SuperDave1426
Country Manager
Country Manager
Posts: 1047
Has thanked: 97 times
Been thanked: 325 times
Send a message

Post by SuperDave1426
berestovskyy wrote:
kentsmith9 wrote:I could have missed the discussion on this one, but how does this determine it is unnecessary?
If this one-way Ramp has a turn enabled to another one-way Ramp. Here is an example: permalink
I fail to see why that is an "unneeded name," given the length of the ramp. I've seen lots of named segments of that nature out here, and have been doing it that way all along, since it helps to provide guidance to drivers making their way to the freeway.
Sorry, maybe the condition is not sufficient for US, so please let me know if I shall disable the check for US or add more ifs.
I'd vote for that. :-) I'm still not sure it's applicable for anywhere else, though. Is there something in the Wiki that I've missed regarding freeway onramps?
SuperDave1426
Country Manager
Country Manager
Posts: 1047
Has thanked: 97 times
Been thanked: 325 times
Send a message

Post by SuperDave1426
kentsmith9 wrote:Otherwise we have to add a node in the middle of one of the two segments on the roundabout. That seems like unnecessary work.
Not to mention that the minute you do that, you create a roundabout with non-sequential segment IDs, and I've read that can cause problems with navigation through a roundabout. The Toolbox will specifically flag such a roundabout as one which could cause potential problems (and if you use the "redo roundabout" tool, it will (quite properly, IMO) remove the extra node you just added because there's not actually a road attached to it).
SuperDave1426
Country Manager
Country Manager
Posts: 1047
Has thanked: 97 times
Been thanked: 325 times
Send a message

Post by SuperDave1426
berestovskyy wrote:03.02.2014 v0.5.10:
- NEW for ALL 'Too short segment' (less than 2m long)
Just a "point of curiosity" question, but how does a segment less than 2 meters in length (which is a bit over 6 feet) cause a problem? I agree, it's kinda pointless to put in a segment that's that short, but I'm wondering why there's a check for it (not opposed; just curious).
You might need to update Validator, since it was disabled few days ago:
22.01.2014 v0.5.2:
- UPD 'Same endpoints drivable segments': roundabouts are temporarily excluded
That's a relief. :-) When he posted that, I was afraid that you'd turned it back on for some reason. :D
SuperDave1426
Country Manager
Country Manager
Posts: 1047
Has thanked: 97 times
Been thanked: 325 times
Send a message

Post by SuperDave1426
sketch wrote:
SuperDave1426 wrote:
berestovskyy wrote:If this one-way Ramp has a turn enabled to another one-way Ramp. Here is an example: permalink
I fail to see why that is an "unneeded name," given the length of the ramp. I've seen lots of named segments of that nature out here, and have been doing it that way all along, since it helps to provide guidance to drivers making their way to the freeway.
Because it will automatically take the name of the next named ramp. They have the same name, so there's no need* to name the first one as it'll already give the second one's name as the instruction. It makes zero difference* client-side, so there's no reason* to have the extraneous name taking up space in the Waze database. It's just streamlining.
Ok, I see what you're saying there.

Although there are still times where naming individual segments apply (as you mentioned further in your message), so it seems that a check of that type is probably still not a good idea. Both in the case that we're talking about and which you acknowledge, but also times when the name changes to provide directional information.

I.E., an offramp that splits; the split segments have information about what they're taking you to beyond simply the road name that they're connecting you to.
SuperDave1426
Country Manager
Country Manager
Posts: 1047
Has thanked: 97 times
Been thanked: 325 times
Send a message

Post by SuperDave1426
CBenson wrote:
SuperDave1426 wrote:Not to mention that the minute you do that, you create a roundabout with non-sequential segment IDs, and I've read that can cause problems with navigation through a roundabout. The Toolbox will specifically flag such a roundabout as one which could cause potential problems (and if you use the "redo roundabout" tool, it will (quite properly, IMO) remove the extra node you just added because there's not actually a road attached to it).
So recreate the roundabout with the extra node.
How exactly do you do that, given that the "add roundabout" tool only adds nodes for each road coming into it?

I suppose it could be managed by temporarily creating a third road leading into it and then deleting that road once the roundabout is created, but to me that falls back into what kentsmith9 was saying about being more unnecessary work.
I still don't really understand the point of mapping roundabouts with only two segments.
Since when is a roundabout about how many roads enter it? A roundabout is a big honkin' (or in some cases, a small honkin' :-)) circular road, running one way, around something in the middle of it. Other roads attach to it and those coming from the entering roads are required to yield to the traffic already in the roundabout.

In U.S. dictionaries the terms "roundabout", "traffic circle" and "rotary" are synonyms. There's no requirement that there be more than two roads attached.
But I have seen them give similar problems as any two segments that are connected to the same two nodes. I've had more issues with two segment roundabouts than with non-sequential segment ID roundabouts.
I'll take you at your word for it, though I haven't seen any such. If such is the case, then it seems to me that those of you with access to Waze staff need to encourage them to fix that, since that's clearly a bug that needs fixing.

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.
SuperDave1426
Country Manager
Country Manager
Posts: 1047
Has thanked: 97 times
Been thanked: 325 times
Send a message

Post by SuperDave1426
dbraughlr wrote:
sketch wrote:So, this check should probably be adjusted to alert for any segment under [X] m long
... for any drivable segment except a terminal (dead-end) segment under [X] m long ...
+1 :-)
dbraughlr wrote:
sketch wrote:Remember that a lot of the checks in Validator are not necessarily wrong, they're just things that might need a look.
...
I'll reiterate the point I made, that many of these are just checks. Not everything needs action.
In practice, you are mistaken. There are many editors that expect the Validator to give their area a clean bill of health. Come to chat and talk to them. Editors have been deleting the loops flagged with "Same endpoints drivable segments" error. I recommended installing the updated Validator instead.

Any arbitrary "rules" like this should be turned off by default, clearly marked as informational when turned on, and marked with green or some color never used for errors.
http://i230.photobucket.com/albums/ee1/ ... hssign.gif
SuperDave1426
Country Manager
Country Manager
Posts: 1047
Has thanked: 97 times
Been thanked: 325 times
Send a message
Last edited by SuperDave1426 on Mon Feb 03, 2014 6:19 pm, edited 1 time in total.