Traffic Congestion, Mobile Apps and Game Theory

There have been numerous complaints that mobile map apps have been dumping traffic from freeways onto nearby residential streets (previously). But a team of researchers, including Berkeley’s Alexandre Bayen, have been looking at the problem from the perspective of game theory (see papers here and here). Basically, the apps operate selfishly, and lead to results for all drivers that are less good than if they had been working cooperatively. It’s an example of the Nash equilibrium. More at Boing Boing and CityLab.

A Google Maps Roundup

We’re almost at the end of the week of Mario on Google Maps. Announced for March 10 (“MAR10” Day), the temporary feature changed the navigator arrow into Mario driving his cart. Announced for both Android and iOS, but for some reason it never turned up in Google Maps on either my iPhone or my iPad, so I didn’t rush to post. [Business Insider]

Something that is turning up on my iPhone: plus codes, which appear to be Google’s homegrown solution to location codes, map codes and the like: a short string of characters that indicate a specific location on the globe. They were announced back in August 2015, but last month Geospatial World made note of their rollout.

Public transit navigation now includes wheelchair accessible routes, as of yesterday: “this feature is rolling out in major metropolitan transit centers around the world, starting with London, New York, Tokyo, Mexico City, Boston, and Sydney. We’re looking forward to working with additional transit agencies in the coming months to bring more wheelchair accessible routes to Google Maps.”

Slashgear looks at the new Google Maps APIs for gaming, which, I guess, enable developers to build real-world games on top of Google Maps. Note that Pokémon Go is not built on Google Maps: I suspect this outcome means that Google has noticed that.

Inevitable, and surprisingly not before now: Disney’s parks in Street View.

Apple, Google and Waze: Which Is Most Accurate?

Artur Grabowski spent most of 2017 testing three mapping apps—Apple Maps, Google Maps and Waze—to see which app was the most accurate in terms of travel time to destination. His questions: which app estimated the shortest travel times, which app actually got him to his destination in the least amount of time, and how much did each app over- or underestimate travel times? In the end, based on 120 trips in the Bay Area, roughly 40 using each service, Artur found that Apple’s estimates were the most reliable (indeed, Apple underpromised and overdelivered), but while Waze promised the shortest travel times, those promises were usually overly optimistic; it was Google Maps that provided the shortest travel times.

Why does Apple underpromise and overdeliver, while Waze does the opposite? Artur suspects it’s because Waze needs to monetize its app with ads, and Apple doesn’t:

For Apple, Maps is a basic solution for its average user who wants a maps solution out of the box. Apple Maps does not directly drive ad or subscription revenue for Apple so there is less reason for Apple to incentivize iOS users to use Apple Maps over other solutions. However, Apple does care about user experience, and sandbagging trip time estimates so that users arrive at their destination on time results in a great user experience. Hence, I believe that Apple is intentionally conservative with estimated arrival times.

At the other extreme, Waze (Alphabet) makes money through ads when you use their app. What better way to get people to use your navigation app than by over-promising short trip times when no one takes the time to record data and realize that you under-deliver? If an unsuspecting user opens Apple Maps and sees a 34-minute route and compares that to 30 minutes in Waze, the deed is done. Now Waze has a life-long customer who doesn’t realize they’ve been hoodwinked and Waze can throw at them stupidly annoying ads.

[Daring Fireball/Kottke]

Strava, Responding to Security Concerns, Disables Features

Strava has reportedly disabled certain features in the wake of the privacy and security issues raised last month, with users reporting that they can no longer create workout segments. In a statement given to The Verge, Strava said: “We are reviewing features that were originally designed for athlete motivation and inspiration to ensure they cannot be compromised by people with bad intent.” [Canadian Cycling Magazine]

Previously: Strava Heat Map Reveals Soldiers’ LocationsNon-Anonymized Strava User Data Is Accessible.

Non-Anonymized Strava User Data Is Accessible

More on the privacy issues regarding Strava’s global heat map and its customer data. Now Wired UK is reporting that Strava’s data isn’t anonymous. Because you can compare your results with nearby users, all it takes is a local GPS tracklog—which can be created out of whole cloth, as Steve Loughran’s blog post demonstrates—to see detailed information about users. Wired UK:

By uploading an altered GPS file, it’s possible to de-anonymise the company’s data and show exactly who was exercising inside the walls of some of the world’s most top-secret facilities. Once someone makes a data request for a specific geographic location—a nuclear weapons facility, for example—it’s possible to view the names, running speeds, running routes and heart rates of anyone who shared their fitness data within that area.

The leaderboard for an area, the Guardian reports, can be extremely revealing. “The leaderboard for one 600m stretch outside an airbase in Afghanistan, for instance, reveals the full names of more than 50 service members who were stationed there, and the date they ran that stretch. One of the runners set his personal best on 20 January this year, meaning he is almost certainly still stationed there.”

Which makes the security issue regarding military personnel using fitness trackers even worse than simply the anonymous aggregate of the routes they take. Yes, this is very much an unintended and unforseen consequence of relatively innocuous social sharing bumping up against operational and personal security protocols; and it’s as much on military personnel to, you know, not use GPS-enabled devices that upload your location to a third-party server as it is on companies to have clear and effective privacy controls. This is very much the result of a whole lot of people not thinking things through.

Previously: Strava Heat Map Reveals Soldiers’ Locations.

Strava Heat Map Reveals Soldiers’ Locations

Strava is a mobile fitness tracking app that uses GPS data from phones and watches. It has access to a lot of data, and has been using that data to create a global heat map showing the paths taken by its cycling and running customers. The map’s most recent update, last November, aggregates user data through September 2017. But analyst Nathan Ruser noticed a problem: in places where local Strava use is low, the map can reveal the paths of people from wealthy western countries—for example, soldiers at U.S. military bases overseas, whether they’re patrolling or simply exercising. (U.S. troops are encouraged to use fitness trackers.) Which is to say, suddenly Strava is a security problem. Details at BBC News and the Washington Post.

New Jersey Borough to Close Streets to Congestion-Rerouted Traffic

When your navigation app (e.g. Waze) suggests an alternate route to avoid congestion, that has knock-on effects on the communities you’re routed through, particularly when a lot of traffic gets pushed onto quiet residential streets.  That’s the situation in Leonia, New Jersey, the New York Times reports, where later this month the police will be closing some 60 streets to non-local traffic in hopes of routing all that Wazer traffic somewhere else. Some of the somewhere elses aren’t happy with this move, naturally. [Engadget]

Google’s Map Data Alchemy

Justin O’Beirne’s lengthy analyses of Google Maps and Apple Maps are always worth reading,1 and his latest is no exception. Looking at the rapid proliferation of buildings, areas of interest and other examples of Google’s Ground Truthing program, Justin discovers that Google’s buildings data are a product of its satellite imagery, its places of interest are a product of its Street View data, and its areas of interest (the orange-shaded areas that indicate business districts) are the result of combining those stores of data.

…so this makes AOIs a byproduct of byproducts[.]

This is bonkers, isn’t it?

Google is creating data out of data.

This is slightly more than Google’s competitors are able to match. As always, Justin’s analysis is worth reading in full, and comes complete with before/after animations that make his point visually clear.

Google Maps Abandons Experimental Calorie Counter

An experimental feature in the iPhone version of Google Maps that measured the calories burned (and equivalent in mini-cupcakes) when walking a route instead has been pulled due to complaints, TechCrunch reports: the feature couldn’t be disabled, the calorie counts were vague and unhelpful, and it could be actively harmful to users with eating disorders. More at BuzzfeedSlate and The Verge.

Augmented Reality Comes to OS Maps

The Ordnance Survey’s OS Maps mobile app now has a new augmented reality mode. “Using the phone or tablet’s camera view, hills, mountains, coastal features, lakes, settlements, transport hubs and woodland in the vicinity are identified and labelled. If a label is pressed and there is a data connection, a page of useful information about that location is displayed, including nearby walks, photos and places to stay.” AR is very neat but battery-intensive; nevertheless this strikes me as a very useful application of the technology. [iOS App Store, Google Play]

Canada’s Indigenous Communities on Google Maps

CBC News reports that more than 3,000 indigenous communities in Canada—traditional First Nations reserves as well as treaty settlement lands and urban reserves—have finally been added to Google Maps. For some reason I thought they already were—U.S. Indian reservations have been on Google Maps for some time, after all (their visibility, or lack thereof, was commented on in 2011: here, here and here).