I like the criteria, although to me the second is a clarification of the first and makes it unnecessary. Also, it's hard to know whether a given rail line has scheduled traffic on it. Because of that I wonder if we can just use the second criterion.
I support simply leaving unjunctioined railroads at ground with elevation conflicts. Gaps look weird or will give warnings in regression checker and Magic. Lowering isn't a problem now but will also look weird once the app shows elevation.

And this is what you are proposing, right?

Proposed article, draft #3 wrote:• In specific circumstances, create junctions between drivable roads and railroads at grade crossings.(rr) Otherwise, do not junction grade crossings. The resulting elevation conflicts are acceptable.

(rr) Adding a junction node for a grade crossing only benefits routing if (1) between the existing junction nodes on either side of the grade crossing, there is at least one destination on one side of the crossing and room for cars to back up waiting on the other; and (2) the tracks support regular, scheduled train traffic. If either of these elements is missing, adding a junction node for the grade crossing will not benefit routing and may in some cases degrade routing. If it is difficult to determine whether regular, scheduled trains use the tracks, you may assume they do if the tracks appear active and well maintained.

I'm good with adding that to the railroad page now. Looks like we reached a consensus about that part back in July.

I vote A

Option A.

If Railroad road types were brand new, or no wiki guidance had previously been promulgated, I would wholeheartedly endorse this proposal.

However, a lot of the U.S. has put considerable editor elbow grease into elevating and junctioning railroads according to the present wiki guidance. At a minimum, the elevation guidance needs to acknowledge this and state that leaving RRs at ground is recommended if that work has been done properly. Junctions are another matter - it may well be worth the effort to review existing junctions and delete RR junctions that are not providing benefit and could be detracting from collection of reliable segment/turn timings.

Thanks, this works and represents an improvement from current wiki which encourages adding RR junctions without regard to the trade-offs.

Indeed, and I didn't mean the draft guidance to suggest setting all RRs back to -5 (although given what has happened on a large scale since the junctioning guidance was introduced, I think we were, perhaps with localized exceptions, better off under that system).

The bullet-point item I'd like to change was excerpted from a long bulleted list in the Railroads section of the Road Types article. The immediately preceding item in the list reads...

• Set the elevation just as you would a drivable segment. When tracks junction a road on the ground, the Elevation should be set to Ground.

...and I don't propose to change this guidance. To make that clearer, the following updated proposal eliminates the suggestion to change elevation:

Proposed article, draft #2 wrote:• In specific circumstances, create junctions between drivable roads and railroads at grade crossings.(rr) . Otherwise, do not junction grade crossings. The resulting elevation conflicts are acceptable. If desired, elevation conflicts may be avoided by breaking the railroad segment to open a very small gap for the road segment.

(rr) Adding junction nodes at grade crossings only benefits train-related routing if all three of the following conditions are present: (1) the drivable road does not already have a junction node nearby; (2) destinations are present along the road between the nearest junction nodes on either side of the grade crossing; and (3) the tracks support regular, scheduled train traffic. If any one of these three elements is missing, adding a junction node for the grade crossing will not benefit routing and may in some cases degrade routing.

How does this sound?

Kartografer wrote:I like the criteria, although to me the second is a clarification of the first and makes it unnecessary. Also, it's hard to know whether a given rail line has scheduled traffic on it. Because of that I wonder if we can just use the second criterion.
I support simply leaving unjunctioined railroads at ground with elevation conflicts. Gaps look weird or will give warnings in regression checker and Magic. Lowering isn't a problem now but will also look weird once the app shows elevation.

This is great and I'm up with all of it, except I do think some language about the rail traffic is needed. So many new junction nodes have been added to the map for rail lines that haven't had traffic in years, or even decades. I agree it can be hard for any non-local to know if regular scheduled traffic uses a given rail segment, but I wouldn't think it's prohibitively difficult to determine if the segment is nearly or entirely abandoned.

I have attempted to combine the essence of former points 1 and 2 into a new point 1, and have given editors some wiggle room with evaluating "regular, scheduled train traffic". Hope this looks OK. Proposed Draft #3:

Proposed article, draft #3 wrote:• In specific circumstances, create junctions between drivable roads and railroads at grade crossings.(rr) Otherwise, do not junction grade crossings. The resulting elevation conflicts are acceptable.

(rr) Adding a junction node for a grade crossing only benefits routing if (1) between the existing junction nodes on either side of the grade crossing, there is at least one destination on one side of the crossing and room for cars to back up waiting on the other; and (2) the tracks support regular, scheduled train traffic. If either of these elements is missing, adding a junction node for the grade crossing will not benefit routing and may in some cases degrade routing. If it is difficult to determine whether regular, scheduled trains use the tracks, you may assume they do if the tracks appear active and well maintained.

The conversations here and on the True Elevation thread appear to be converging on a need for guidance regarding junction nodes and turn pockets etc. Rather than repeat that discussion everywhere it comes up, it may be that separate guidance is best. To that end I've suggested something, please see:

viewtopic.php?f=1636&t=261298