Please stop guessing about highway/waterway crossings
Posted by valhikes on 14 October 2024 in English.You know what can really mess up carefully laid plans? What can even lead to a dangerous situation? When you get to the crossing of the South Fork Trinity River Trail and the East Fork South Fork Trinity River, a waterway that very much is still a river, and that bridge that someone has assured you they know is there is not there. There’s not even the evidence of once having footings for a bridge.
Apparently the mapper knew based on Bing aerial photography. Through thick trees. I can make guesses about how that crossing goes, too. At least when I guess wrong, I know it was a guess and might have made contingency plans around it.
Fortunately in the early days of Autumn, that ford I was LIED to about is just 4m of knee high wading without much current. That river comes out of some mountains that collect snow. It isn’t always so low.
The thing is, if it is important, it HAS to be correct. Your guess isn’t good enough. If it isn’t important, it can be left until someone who knows feels like doing it. There’s no renderer having trouble deciding how to render this crossing. It’s quite common that they aren’t specified.
I’ve been coming across a lot of guesses lately. I know they are guesses because they are wrong.
I was looking over an area I’d been working on when something started feeling wrong. It took me a couple minutes to notice that one of the roads I had aligned recently had sprouted four bridges. What? It was the unimportant spur of an unimportant low standard Forest Service road. They also get referred to as “unimproved” roads. We’re talking a road that might be produced simply by running a blade across the dirt. The whole thing probably cost a fraction of the cost of a bridge and suddenly it had four. Based on Bing imagery again, in which you can clearly see it is just a dirt road.
Why did this person do this? Well, they don’t like to see errors on the map. IT’S NOT AN ERROR. They were using iD, which uses yellow colors and the term “issue” for this sort of thing. It uses red colors and the term “error” for errors. You know what is an error? Fake bridges. But no automated error catcher is going to see those.
They added that “we can’t have” people just drawing roads over waterways without saying how they cross. Well, the road is there. The waterway might be. I’ve seen these NHD lines climb over ridges, so it’s debatable. But let’s assume the waterway is there. Seems more important that they are on the map than that we know how they cross. We can’t just leave roads off because we don’t know how they cross the waterway.
I was adding in details along a trail I’d just hiked in Prairie Creek Redwoods State Park when I noticed the beautiful curving footbridge across Hope Creek was marked as a ford. What? Then I found some more fords marked where I knew there are permanent bridges. Most were on inconsequential crossings, but 3 were on Prairie Creek where I know there aren’t fords. Prairie Creek swells in the winter to something that would be chest high at least to cross. Telling people this is a ford could change their plans in ways that are unnecessary. Where exactly those bridges are, I can’t quite remember. Unfortunately, the creek isn’t aligned well. The trail should be.
Moving over to Redwood National Park, I found bridges on the abandoned logging roads that predate the park. Popular way for loggers to cross waterways: dump dirt on it and let it wash out next winter. These are only summer roads anyway and there’s more dirt where the first came from. It’s not good for the salmon downstream, but loggers aren’t known for caring about salmon. Anyway, some of these roads still exist, some don’t because they’ve been actively removed, some don’t because the forest is good at erasing them. None of them beyond what’s already been marked as trail is in legal public access areas, but for emergency purposes it would be nice to know what’s there. I’d actually stuck a question asking if we should keep these tracks tracing USGS lines and the person adding the bridges had answered (in a changelog), of course, we wouldn’t want people to have to get out the old USGS map to follow them. This is a person who clearly trusts the USGS map even when the latest one is from 1970 and many of the quads don’t even know there’s a National Park there. Frankly, if you want to follow an overgrown, vanishing road, you better not be too lazy to get out the USGS to do it.
So anyway, when you’re looking at iD nag you about crossings and making it so easy to add in a bridge or a culvert or a ford and even getting ready to pat you on the back to say how many crossings you’ve resolved, please remember that fourth option: “ignore this issue”. It’s a very valid response! When these are important, they have to be right! If they aren’t important, they can wait. There can be consequences to guessing wrong. Don’t guess. Please please please don’t guess.
And if you “don’t like seeing errors on the map”, well, getting it wrong is an error. Leaving it is just an issue.
Discussion
Comment from Xvtn on 14 October 2024 at 20:19
I strongly agree. It’s OK to allow warnings! I think iD’s UX needs to be changed to not imply that lack of data equals an issue with your changes.
Comment from rphyrin on 16 October 2024 at 13:21
I wonder if there is even a surefire solution to fix the inherent problems that any ‘anyone can edit’ collaborative project (like this one) faces. Not everyone has enough resources to defend every inch of the content, whether against prank vandals or a very persistent good-faith editor (who was misled and actually caused significant damage to the project). The back-and-forth edit warring is exhausting and feels pointless. Also, not everyone has the mental stamina to prolong the conflict by reporting it to the available higher authority.
Or maybe we could look up to more mature projects out there (like Wikipedia). Over time, they made the project less and less inclusive (only the trusted elites have the privilege to edit the content), all for the sake of quality assurance.
Comment from valhikes on 16 October 2024 at 15:34
Honestly, these are 100% well intentioned edits. The ones I’ve seen are using iD, so they’re operating under the pressure of a nag that inserts itself above the tag fields. After so long with sophisticated editors, they’re not going to see a lot of errors, so it’s quite understandable to interpret these “issues” as “errors”.
I live in the wild west and travel the wilder west, so there’s too much needing done to bother checking what is hopefully as good or better than I left it. A couple months ago I turned my attention to a Congressionally designated wilderness and booted the single “residential” road crossing it for three trails. Two of them do happen to follow within half a mile of where the road was drawn. For Trinity County, within half a mile can be pretty good. (Oh, the things I’ve seen in TIGER data…)
I just want these editors to consider “ignore” as a legitimate answer when they can’t find evidence for how a crossing occurs. This isn’t a natural inclination, after all. But the TL;DR of it all is: If it is important, it has to be correct. A guess isn’t good enough. If it is not important, it can wait until someone knowledgeable does it. A guess isn’t needed. So please, just hit that “ignore” rather than guess.
Comment from Exe19 on 17 October 2024 at 18:51
Ignoring iD warning messages counts as editing error in How Did You Contribute Quality Assurance statistics, which does not help with your issue.
Comment from Minh Nguyen on 20 October 2024 at 23:20
Thank you for this important message. Validators work with very little context and even less intelligence. This goes for not only iD’s validator but also JOSM’s, Osmose, etc. “Ignore this issue” is absolutely a valid response. If something is so bad that you shouldn’t ignore it, it would be an error and iD would actively block you from saving your changeset.
I’ve been concerned for years about the tendency for these issues to become “gamified” because mappers perceive HDYC as their permanent record, as if editcountitis wasn’t enough of a problem. (At this point, I wear my Osmose issue count as a badge of pride.)
Aside from possible UX improvements to iD, we should figure out how to identify any bridges or culverts that were created hastily or carelessly. By default, the “Add a bridge” and “Add a tunnel” suggestions create bridges and tunnels of a certain length based on factors such as the tags on the crossing way and the angle of the crossing. When iD introduced this feature, the developers expected the mapper to manually adjust the bridge to match the real-world length. However, this is unlikely to happen if the mapper can’t see the bridge in imagery. Maybe we can reverse-engineer these heuristics in an Overpass or QLever query.
Comment from GeeMaps! on 21 October 2024 at 02:04
A possible fix could be changing the iD warning message slightly:
“Possible error - road crosses waterway
DO YOU KNOW WHAT”S THERE? (in Bold!)
If so, please tag as either bridge, tunnel or ford.
If you’re not certain, please “ignore”.
Thanks!”
Comment from valhikes on 21 October 2024 at 06:23
Re automated changeset quality checkers: I believe I looked at one of these once and came away very dissatisfied with the whole business. When JOSM is telling me that there’s unnamed “residential” roads touching the slew of often unnamed residential roads I have just turned into named and numbered unclassified/track/service roads, I tend to say, “Yes, but this changeset is already obnoxiously large and I’m going to get those in the next one.” Even when I finish fixing those too, it still counts against me.
I’ve been known to add a ford where there is no waterway too, and that didn’t get fixed in the next go. Still seems important when an otherwise “good” road sets out across a river(!) without a bridge or even culvert. I tried to redirect one person’s attention to drawing the river to fix a ford issue instead, but I’m not diplomatic enough or they’re not interested in waterways. (If you’re a person interested in drawing waterways, here is a map of missing GNIS named waterways.)
Re improvements to the iD interface: The suggestion seems very verbose, which is rarely an effective strategy. Simply changing the order so that the “ignore” option comes first instead of last could be what someone needs to understand it is valid, not just tacked on as a last resort. The wording could be changed. “I don’t know” might work. Although the most honest, a lot of people are uncomfortable admitting it. “Cannot be determined at this time” might be effective. It’s getting wordy, but “cannot be determined” without the modifier is false.
Comment from Minh Nguyen on 21 October 2024 at 11:13
Good idea. “Ignore this issue” is an accurate description of what happens, but some users may perceive ignoring to be an act of laziness or even malice, whereas it can actually be more neutral than that.
“I don’t know” would be similar to the options available in StreetComplete and MapRoulette, two alternative editing environments that prioritize human factors in their workflow designs. “Can’t be determined at this time” isn’t terribly verbose. There’s already a “Not the same x” option on the warning about missing brand tags, where x is an arbitrarily long brand name. If it gets long enough, it simply wraps to the next line.
Please open an issue in the iD issue tracker so it can be triaged and given further consideration. Thanks!