Page 6 of 9
Re: [Page update] At Grade Connectors
Posted: Fri Aug 29, 2014 7:47 pm
by PhantomSoul
qwaletee wrote:if an AGC connects segments with three different types, do we have specific guidance? The general rule:
>> The connector segment's type should be set based on the lowest of the types of roads it connects.
Seems to imply take the lowest of three, but it may not have anticipated all cases of three types. Do we take into account primary use, prevalent connections, count or discount transitions of road type that HAPPEN to take place at the junction with the AGC?
Look at this example:
https://www.waze.com/editor/?env=usa&lo ... s=32170542
It is basically a two-way AGC, with three road types connected.
I actually see the selected segment as dual-purpose:
1) Northbound connector from S Cedarbrook Rd (mH) to Rte 73 (MH)
2) A segment of Meyers Ave (PS)
In its strict interpertation of an AGC's lowest type, the selected segment would be PS. The NJ FC map shows Meyers Ave in this immediate area as a collector road, but also never shows any connectors, ramps or otherwise. I would imagine the selected segment needs to be at least mH to preserve road type continuity when traveling from S Cedarbrook Rd to Route 73.
I guess what's unclear to me is how significant is that continuity, vs. just leaving the selected segment PS?
Re: [Page update] At Grade Connectors
Posted: Tue Sep 02, 2014 5:33 pm
by PhantomSoul
So, in summary about the AGCs, we want to use the lowest connected type that maintains continuity, which is not necessarily the lower of the top two types, per se, but that's how it will usually turn out. Right?
Re: [Page update] At Grade Connectors
Posted: Wed Mar 19, 2014 4:17 pm
by qwaletee
CBenson wrote:Doesn't the "What links here" link in the "Tools" box at the bottom left of each page provide the information.
Yes, it does. Wiki even reminds you to check for such redirects when you do a move.
I just fixed a double-redirect yesterday.
There are also tools we can run to find stragglers, assuming we're still mostly compatible with wikimedia.
Re: [Page update] At Grade Connectors
Posted: Wed May 28, 2014 7:07 pm
by qwaletee
I would favor the type of the road transitioned from, which will also solve the problem PZ mentions, and be more obvious and consistent for the driver viewing the map. I have heard objections to both versions, in that it will visually make it look like there are a lot of odd appendages to the major roadway that is color highlighted, but
1) I believe on the app screen that only shows a small chunk of the road, it would not be offputting
2) I would still take clearer directions over aesthetics of the segments any day
This could be solved with microdoglegging instead, but that's more effort for thr editors, and a bigger hack that might one day be problematic due to some update that makes them more visible.
Re: [Page update] At Grade Connectors
Posted: Mon Jun 02, 2014 3:31 pm
by qwaletee
sketch,
Subjective. I actually like the way it looks, both night and day. Your "tentacles" say to me "oh, look, that's how my road connects to the next, glad I can see the intersections to the smaller streets so clearly."
For pure aesthetics, I don't think it is ugly, either. Art, no, but ugly neither.
Re: [Page update] At Grade Connectors
Posted: Sun Aug 03, 2014 9:12 pm
by qwaletee
sketch wrote:I will be adding the exceptions (MUTI, RCUT, DLT, Jughandle) to this article to achieve parity with the Road types/USA article.
I'm not sure if the entirety of the content discussing those intersection styles should be placed here, as it can get kind of lengthy for some of them.
Also, should this page be made into a subpage of the junction style guide? It is, after all, a junction style issue.
Subpage sounds good, though it doesn't matter that much. Keep the summary in the main article as short as possible to avoid overloading the page, which I think is what you were saying you were inclined to do anyway.
Re: [Page update] At Grade Connectors
Posted: Sun Aug 24, 2014 9:15 pm
by qwaletee
sketch wrote:
The commitment point is a good place to put the "stay left" if the AGC is designed to give a "stay left" followed by a "turn left" for particularly long or physically separated turn lanes or particularly complex intersections.
However, for AGCs designed to give a "turn" instruction at the start point of the AGC (in the case of, say, physical separations), the commitment point is not the optimal location. I'm not sure if you meant to include or exclude those, but I want to make it clear. Putting it too early has the disadvantages of (1) the turn instruction leaving the screen too soon, (2) the next prompt spoken too soon, and (3) inconsistency with timing of non-AGC turns. Consider a 40 m solid white line for a normal right-turn lane, then a 40 m solid white line leading to a 30 m physical separation. We wouldn't add an AGC to make the former sound sooner, so we should keep that in mind when designing the latter. The "turn" instruction is expected to be based on the actual point where the turn is made, whether you are guided by a physical island or not.
Sketch,
In the case you are giving (40m solid white followed by 30m physical separation), we could force a keep right for the initial turn lane followed by a turn right (using hackery unfortunately), if we needed to give the dreaded early lane guidance without compromising the turn direction itself. Or am I missing something here?
There's also the case of the extra wide turn and diverging main road that cause Waze to lose track of the driver position, but I don't think those would typically need ramps to begin with, just careful geometry of a street AGC (more hackery!).
Re: [Page update] At Grade Connectors
Posted: Fri Aug 29, 2014 7:24 pm
by qwaletee
if an AGC connects segments with three different types, do we have specific guidance? The general rule:
>> The connector segment's type should be set based on the lowest of the types of roads it connects.
Seems to imply take the lowest of three, but it may not have anticipated all cases of three types. Do we take into account primary use, prevalent connections, count or discount transitions of road type that HAPPEN to take place at the junction with the AGC?
Look at this example:
https://www.waze.com/editor/?env=usa&lo ... s=32170542
It is basically a two-way AGC, with three road types connected.
Re: [Page update] At Grade Connectors
Posted: Fri Aug 29, 2014 8:45 pm
by qwaletee
So the second highest? That's an interesting calculus. How do we arrive at that?
Re: [Page update] At Grade Connectors
Posted: Fri Aug 29, 2014 9:06 pm
by qwaletee
I am talking about Roundabouts for the moment, not AGCs, but only because you brought it up 
[/i
I understand the continuity issue, but you said something a little different than that. That rationale you just stated doesn't lead to "lowest of the top two," in your previous post. In some cases it might -- if you assume that MH and Fwy are mostly equivalent, then a roundabout that connects Fwy, MH and mH (never!) could indeed use MH, since it is not perceived to be a problem, whereas mH is perceived to be a problem. So that would leave MH and Fwy, and MH is the lower of those two.
But say it connected MH, PS and Street (which is actually a reasonable combination given FC). There, the lower of the top two, PS, is a significant continuity issue compared to MH, so why would you choose the lower of the upper two? I would think you would need to use MH to ensure continuity, i.e., the absolute highest.