Interesting twist on what other people have done, namely show a heat-map of where you can go using public transit + walking within 5 min, 15 min, 30 min.
Right, the travel-time metric is not compatible with a Euclidean R² metric. You can imagine three subway stations in a triangle loop, such that it's a shorter trip to do a full loop on the subway then to walk to a point in the interior.
There's no way to continuously deform a map so that it represents travel times as distance in a plane.
Oh yes, unfortunately, you can't do this perfectly. There are some graphs that cannot be embedded in Euclidean space in any number of dimensions, e.g. a 4-cycle with distance measured by path length. It's a good-enough approximation for visualization purposes, though.
Not really - you have to wait for the subway, it also takes a finite time to travel, and it frequently stops. It can avoid traffic, but the actual MPH can be slower than a car when you include both those things.
Isn’t the point that you have a way to get to a point far away faster than you can get to a point in between? The worm hole thing ist because you can only exit at discrete points so it pulls a single point far away, and it’s sorrounding, closer to the starting point. That’s probably hard to map to a 2D map because there would be some overlap between the different „islands“ starting from subway stations
This sounds conceptually difficult: Wouldn't the presence of one-way streets or airport security checkpoints, for example, mean that the travel time from A to B might not be the same as the travel time from B to A?
Right. There could be northbound traffic but not southbound for example. The answer is they must be symmetrizing the distances somehow. Alternatively, you could show the travel times _from a source location_ unambiguously, which might be more interesting for the average user (e.g. from home)
Yeah, this is what I was hoping for with the touch. Seems like where I touch should be the origin and distance in x/y should represent time to travel from that origin.
That still wouldn’t work in all cases. For instance if a ring of locations are all more accessible than a location they contain. Which applies to basically all suburbs.
Indeed, you can't do this perfectly – it's an approximation. I just use one of the two directions as the travel time because if I wanted to symmetrize I'd have to call the Google Maps API twice . Asymmetry aside, it's also not possible to perfectly embed metrics in 2D Euclidean space, or any-dimensional Euclidean space for that matter (see other comment)
Before trying the site I assumed the way they would resolve this is that the distance calculations would be made relative to the point at which you held the map, but what actually happens is that the map is drawn uniformly no matter where you hold it. and of course this is an impossible representation due to the reasons you describe as well as the differences in mappings between a path that is configured to be walked in one direction, but traveled by vehicle in the other.
This is what it is supposed to do. The coordinates in map from one location to another change from distance to how much time it would take to reach there.
Sorry, that must be a bug. On desktop the map is large enough to be visible in its entirety but it is supposed to pan on mobile... I'll try to fix that
It would be a nice effect if the pressed point on the map stayed at the same location on the screen (the pressed point) during and at the end of animation.
Yes! The fact that isochrones become circles is one of my favorite things about this. I also discuss this in the video (https://youtu.be/rC2VQ-oyDG0?t=134) I made about these.
I prefer the typical isochron presentation over the warping. It's unfortunately hard to find free, public isochron maps, though. Some examples I've found:
There have been so many interesting map sites posted to Hacker News lately and my favorite thing about them is how they're all (seemingly) hosted on a Linode Nanode...
So it would be like a 2D black hole, you could approach it from any side, and you could move along the edges a bit, but if you go too far into it, you’ll just find yourself moving along an infinite area toward a point you’ll never cross, forever.
I had this though many years ago, but unlike distance, travel time isn't necessarily symmetric, and so it isn't possible to represent time just by warping a conventional map as A->B and B->A will get the same warp.
And if the map allows for different travel modes, like train vs driving, I think it breaks entirely.
But this isn't to poop on the website -- I'm glad someone has taken the time to build a demo, even it it can't handle some cases.
You can turn on arrows that update as you move your mouse around (assumedly, signifying an origin). Not sure exactly what they do, but they do hint at potentially taking asymmetric distance into account.
time2reach https://map.henryn.ca/San%20Francisco
NYC Subwaysheds https://news.ycombinator.com/item?id=36240264