Hacker News Reader

Comment by exegete | Hacker News Reader

The dotted line is not directions. It just indicates that you can’t get to the destination by car so it brought you to the closest place and you’ll have to walk the rest of the way. The dotted line is not walking directions but I guess I can see the confusion.

View on HN

Most UX design makes lots of assumptions. Someone who is not familiar with Google maps could very well interpret the dotted grey arc as a real path. It's almost identical to the dotted blue path.

How? I really do not see it. Like, I'm glad they are taking feedback and acting on it, but this seems like an overabundance of caution

If they routinely use maps for transit or walk any significant distance from a parking spot, they have a completely different expectation about what the dotted lines mean.

Actually I do think the expectation for the navigation to bring you to the point from where you can actually reach your target is justified. Like when you want to drive to say Disneyland it shd send you to a parking near the entrance not to a parking at the other side from where you cannot enter (but might be closer to the castle).

Huh

This looks to me like an entirely valid hiking route that says to use actual marked foot trails on the ridge.

If you have configured it for walking it definitely uses real trails from topographical maps. It wouldn't be outrageous to assume that Google will automatically switch from driving to walking mode at the best location to exit your vehicle.

I would personally be double checking if I was confused but Google does do both things so it's easy to imagine Google being smart enough to combine the two now that it obscures so much of the interface.

https://www.google.com/maps/dir/43.3518109,-71.7913173/43.38...

I think there's a couple things here:

A dotted blue path indicates a path to walk on. This shows up in public transit or walking directions where part of the trip actually involves walking.

A dashed grey arc appears when Google Maps can't actually get you all the way to the destination but can get you close. This arc indicates distance, but not a particular route. This is what the screenshot in the article shows (except the screenshot shows light blue instead of grey for some reason).

In particular note how the arc in the screenshot is perfectly smooth instead of following a path, a lighter color than the rest of the route, and dashed instead of dotted.

And unless the user is really used to distinguishing between those two, having that with no further explanation is a terrible way to design a UI.

"It wouldn't be outrageous to assume that Google will automatically switch from driving to walking mode at the best location to exit your vehicle."

On the other hand, it also wouldn't be outrageous to assume that if an app is giving you "walking directions" that the route would be walkable for an average user rather than "highly dangerous, even for experienced climbers".

Sure, I think that would also be a reasonable guess, without a warning. Both are reasonable things for someone to think a company that shields itself in a cloak of magic were to do.

Honestly I stopped using USGS maps for moderate hiking and just started using google topo maps because they show trails, I'm not going to assume anything about a trail though just because google says it'll get me from point A to point B. I also actually know how to read a topo map so I know the kinds of things to look out for.

My point is that Google maps pitches itself as amazingly clever about giving directions. Expecting it to suggest when to change modes of transportation is totally within the realms of what Google certainly is capable of doing; they used to do that for that meme where they say to drive to some location in New Jersey and then swim the Atlantic to get to France. It still said where to start swimming from and where to dry off. And that was like 15 years ago.

Calling out warnings for dangerous trails like they do for traffic jams seems like an obvious thing to do for both situations, just do the same for when it's not actually suggesting a path at all. That'd be great!

I've been misled a number of times by Google Maps driving directions to places that can only be ultimately reached on foot, because in general it tends to route to the nearest drivable location by straight line distance and not necessarily the starting point of a usable/preferred walking route. Which seems to be what happened here. I don't usually attempt walking regardless, though!

Almost worse is that there is a walking path from the car directions end point to the top but it’s a difficult trail suggested only for “experienced hikers.”

Yeah, I find that Google Maps is pretty bad at knowing where you should be parking to access something by foot.

I'm honestly surprised because I'd think anyone who wants to hike a mountain knows no mountain trail goes straight up like that. A little bit of logic applied and you'd just follow the signs from the parking lot.

The signs from the parking lot are for a trail designed for experienced hikers only. It’s the wrong parking lot entirely to send people to who lack experience and are casually googling access to Ben Nevis.

How does it make sense to say “you have to walk the rest of the way” by drawing a line on a map, unless you intended it to imply that that’s the path to walk?

Here's what that line looks like currently: https://i.imgur.com/tougs6p.png (note: this is with the updated end-point at visitor center, probably won't have gone through loch previously)

It's de-emphasised, doesn't show up in directions, is a perfectly smooth curve rather than conforming to paths/roads, and doesn't have the label saying to walk it.

I think the vast majority of people would correctly interpret it as just showing the distance between the end-point and the destination set and I don't think it's necessarily bad UI. But still, with the popularity of Google maps and potential danger in misinterpreting it, it's probably worth considering ways to make it even clearer.

In order to look like it's indicating distance, it has to be a straight line, not a curve, and moreover actually have the distance written alongside it. That's at a minimum, I'd say; putting arrowheads on the ends might also help, e.g.

I can only tell it’s de-emphasized because I’ve been told it is. And while I have vision correction it’s reasonably up to date, and I’ve never suspected I have any kind of color blindness until reading this thread.

> I think the vast majority of people would correctly interpret it as just showing the distance between the end-point and the destination set and I don't think it's necessarily bad UI.

If this thread is anything to go by, slim majority at best.

> But still, with the popularity of Google maps and potential danger in misinterpreting it, it's probably worth considering ways to make it even clearer.

This is the really important point, and thank you for making it. Google Maps is used by 100 of millions (if not billions) of people. If these issues only impact 1% of users, that's still millions of people negatively impacted.

At scale, even "niche" issues matter.

If you use Google Maps on the regular, you'll be used to a solid blue line (yellow or red if there's traffic) on the car's route. I don't remember the first time I came across the dotted line, but I certainly don't remember having a hard time figuring out what that meant. Apparently this isn't the case for everyone, though.

Also, I'm a nove hiker at best, but even I know that there's essentially zero hikes where the shape of the trail is a nice arc across the map like the one in the story. I'm legitimately baffled that anyone would have thought that was a suggested path to the peak.

The dotted arc could easily be interpreted as “this isn’t the exact path but there is a navigable walking path in the general direction of this line” which is just as unhelpful and potentially dangerous.

You run into the dotted line a lot if you use mass transit directions.

How about a dotted line where the dots are question marks.

A ? ? ? ? ? ? ? ? ? B

A - ? - ? - ? - ? - B

A - - ? -- ? -- ? -- B

Agreed, this is extremely confusing and a pretty big design flaw. Google Maps pops up a warning if the business you are traveling to is closing soon, or if there are tolls. It should show a warning in these situations where the road ends far from the destination. The heuristics are less obvious, but I think you could catch most cases. You could even prompt to add another section to the trip in walking mode.

Yes! I see this all the time when setting directions for provincial parks and other "remote" destinations. It never occurred to me that someone would blindly follow them.

I think part of the possible confusion is that "cutesy" curved line from the parking area to the peak. It shouldn't draw anything between where the actual directions end and the map pin, just show the pin.

All that being said, changing the directions so it ends up at the visitor center is probably a good choice, regardless of how few people actually were or would be confused.

There are better ways to communicate this though.

I'd consider a shaded-in area with wiggly outlines, big enough to be path shaped, that just says "walk". That conveys "you need to get yourself to this pin, and we don't know how to help you".

I disagree. If Google doesn’t know enough about the walking path to even be able to plot it on the map, then they shouldn’t represent to users that there is a navigable path at all.

Yes, but if they know there's a navigable path but they don't know what it is this is a reasonable approach.

Obviously, mapping a destination on an island with wiggly lines through the water which say "walk" would not be useful, but wasn't the situation I was envisioning.

It never occurred to you that somebody would "blindly" (what does that mean in this context?) follow a dotted line drawn on Google Maps? That's pretty unreasonable, to put it lightly. Blaming users for the consequences of bad UI is a huge problem in software.

But I’m also confused that people will blindly follow turn-by-turn road directions too. Even when road signs disagree, even past road closed signs, even on gravel roads right before the first snowfall. People assume the computer knows the map and that the map is the territory. Yes, it’s terrible UI, but people don’t ask themselves if they interpret it wrong. At least you can fix the map; not sure how to fix the blind trust in humans.

While you and I and many a HN reader will understand this, there is something to be said that people will, at some point, mistake this for an actual route and try to follow it.

And that's on them isn't it? Is learning in how to read a map (Google or otherwise) seriously a too hard of an expectation from a person that goes mountain climbing?

I can see the confusion but users also should be mindful of the context of using an app. Often times there is no exact location for example so “routes” should be taken as a general guide rather than specific path.

If I use google map to navigate to a lake for example, I wouldn’t expect swimming instructions to the middle of the lake. In the photo in the article it looks similar. It’s just navigating to the rough vicinity, instead of an exact location.

> It’s just navigating to the rough vicinity, instead of an exact location.

Even that is problematic if they don’t know enough about the route to gauge whether it is viably navigable or hazardous.

It's not just the dotted line that is being criticized. The driving directions end at the physically closest car park, but not the most appropriate for most people.

Great that you know that. I didn’t know there were different kinds of dotted lines on Google maps until reading this thread. I bet a lot of other people don’t either. I get dotted lines every time I need to change transportation (walk to/from/between transit). If I saw the screenshot in the article the only clue I’d have that it’s different is the smooth curve, but that’s just because I’ve spent a lot of time looking at maps. I wouldn’t expect people using their phone for directions to know that’s unusual.

I don't think the article is suggesting people actually literally followed that dotted arc - the mountaineering trust chap's quote is that all the routes from that car park are trepidatious even for experienced people.

i.e. their concern is that GMaps users are being directed to that car park, (the highest one, hence closest to 'Ben Nevis' marker) not what the dotted line does or doesn't show from there on. So the solution is to direct them to the (lower-down/further away from destination) visitors' centre car park instead.

The article is suggesting both.

> "And the trust said that a dotted line leading from the car park to the summit could lead walkers into danger."

Yes but it continues:

> Mountaineering groups said the dotted line crossed "potentially fatal" steep, rocky and pathless terrain, while a suggested walking route for a different mountain, An Teallach, would lead people over a cliff.

i.e. you can wiggle it all you want, it's what it crosses that's the problem, you'd have to walk around and go up a different face?

Reply on news.ycombinator.com