Here’s a Pop Quiz. Maui is 2,336 miles away from San Francisco and Montgomery, Vermont is 2,067 miles away. Which is closer?
A map might suggest that they are both about the same distance away. But if you want to go to either, it turns out that Montgomery is almost twice as far away—ten to twelve hours of total travel time versus 6 to 7, factoring in flights, cars, etc.
In our modern world traditional geospatial maps aren’t useful for practically understanding how far away specific points are from each other or for getting a sense of how long it takes to get from point to point. They are great for the “how” of navigation, which roads lead where. But for what is really useful—travel times and making actual navigation decisions—we need something beyond the digital maps to which we have become accustomed and which are dominated by Google. Now is the time for someone to build them.
This is a harder problem than it sounds.
When humans navigated by ship or cart, speeds over land and sea were somewhat constant. In the modern world true travel time comes down to whether or not there are direct flights between places, or whether you need to deal with time consuming and risky connections. In many places, traffic is an enormous practical factor. Timing is a big issue. Tokyo is much farther away from San Francisco in the middle of the day when you have to wait for a flight. And the Lyft or Uber that is four blocks away and in the right direction is actually much closer than the one that is currently two blocks away but the wrong way down one way streets.
What it boils down to is that to understand relative place in our modern world, distance isn’t really the key factor, and cardinal direction is irrelevant. What matters are time and money. If you are willing to spend infinite money, the world might almost look as it does on a physical map today. You could fly privately into regional airports anywhere, and helicopter last legs at more-or-less constant speeds with no changeover time. But on a reasonable budget, the distance in time between places is very different than the distance in space.
Why It Is Time for Someone To Address This
This all might seem a bit academic, but there are a handful of practical reasons that maps based on time and financial means are increasingly important—far more so than spatial maps.
First, the fast-proliferating on-demand services offering ride-sharing, food, handymen or goods (Amazon, Google) within cities need this data and access to it through APIs to manage their services.
This isn’t just a convenience issue about accurate ETAs for deliveries. These maps are critical for businesses to manage routing and distribution to make their services work and cost effective. I have to assume that FedEx and UPS already have versions of this map using their own internal resources. To be competitive, others will need them too.
Second, people are getting more adventurous in their travel decisions, seeking out new places instead of always going to the same one. This is in part because mobile phones and technology make it a lot easier to learn about a new place. As they do so, the demand for accurate travel information rises.
Another reason: the coming of self-driving cars. Soon enough, driving directions will be fully automated. Unless a person is running or biking, he won’t care about how to get from point to point; he will trust cars to do that for him. This means that the focus and usefulness of maps will not be just about how to get somewhere, but about where you are going and when you will arrive. Customers of Uber and Lyft already experience this to some degree, but they’ll be even less conscious of how they are getting to a place when there is no driver to communicate with.
How it Could Work
Imagine yourself as a user opening a mapping app and seeing a series of concentric circles representing travel times and points along those concentric circles representing places the person is likely to visit, like their home, office or gym. Each major point would be surrounded by a constellation of other points near those locations, which would appear closer to you as you moved to that point. You could search for any point of interest or activity, like skiing, or good beaches.
You could change your price threshold for travel, and the points would re-organize themselves accordingly. If you drop your price lower then maybe that direct flight is no longer an option and New York pops further away. If you drop your price even lower, the points would reorganize based on only using buses and trains and the world gets much larger. If you pull your price to the maximum, it assumes you are willing to take helicopters.
You could also change the time you wanted to travel and fast forward the map to the future or rewind it to the past. As you moved through weekends and weekdays, rush hours, the relative distance of the places you are interested in would collapse towards you and move away as travel options changed.
Of course, the real value here would likely be the API, which various services could submit requests to in the form of options and get back expected travel times to each, forming a decisioning engine for where to go. Instead of having an API which gives you back directions to a single given place, these maps could have an API which gives you back decisions about where to go from a set of options.
Such a map would require a lot of data and a lot of computing power. Forget the hardest versions of general routing problems (which NASA literally has quantum-ish computers working on). Simply being able to generate a personal map of the world relative to where a given person is in space is a lot harder than mapping streets. Layering in all flights, public transit, traffic patterns, layovers and delays makes this an extremely intensive data acquisition and business problem before it is even an extremely hard engineering problem. All that said, it is now more possible than it was even a handful of years ago, as costs for compute decline and the volume of information that can be processed to detect patterns increases.
If you think about it, maps have always changed with the times.
In the 16th century, when few people traveled beyond their village and religion was central, the most useful map of the world to most people was more akin to Bunting’s very famous 1580 map, which depicts Jerusalem at the center of the world.
The builders of the New York and London subways decided to depict underground routes practically. They show stops not real space. But they are extremely useful ways to visualize how to get around practical systems of worm-holes built in the last two hundred years.
Now the world we are rapidly entering needs maps to do more than provide directions. They need to be decision engines for where to go oriented around time and price, not space.