Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

City street orientations around the world

geoffboeing.com

Post reply on HN
Print view View original

41–50 of 89 posts

Re: City street orientations around the world

#41

Post by junke »

hoorayimhelping wrote: post #35
noelwelsh wrote: post #2

Funnily enough I find organic street layouts easier to navigate as a tourist than grid systems. In an organic layout there are central roads that go to places of interest and the smaller streets flow to the central roads. In a grids all the roads look the same (they have the same size for example) so I often can't work out how I'm oriented and wandering around doesn't necessarily lead to anywhere interesting. I also…

> In an organic layout there are central roads that go to places of interest and the smaller streets flow to the central roads That doesn't sound organic, it sounds like basic transportation planning and street hierarchy: https://en.wikipedia.org/wiki/Street_hierarchy . Smaller roads flow into bigger roads which flow into main roads which flow into highways. "By contrast, in many regular, traditional grid plans, as l…

Organic does not mean random.

Re: City street orientations around the world

#42

Post by sjwright »

noelwelsh wrote: post #2

Funnily enough I find organic street layouts easier to navigate as a tourist than grid systems. In an organic layout there are central roads that go to places of interest and the smaller streets flow to the central roads. In a grids all the roads look the same (they have the same size for example) so I often can't work out how I'm oriented and wandering around doesn't necessarily lead to anywhere interesting. I also…

This is one thing that Google Maps now solves very well, by highlighting popular areas in pale orange. While I prefer Apple Maps for walking and public transit directions, I must say I love Google Maps for providing the data to achieve strategic serendipity. Thirty minute walk back to the hotel? Let's detour down this parallel road, it looks like there's a lot going on there!

Re: City street orientations around the world

#43

Post by victormier »

seer wrote: post #37
noelwelsh wrote: post #2

Funnily enough I find organic street layouts easier to navigate as a tourist than grid systems. In an organic layout there are central roads that go to places of interest and the smaller streets flow to the central roads. In a grids all the roads look the same (they have the same size for example) so I often can't work out how I'm oriented and wandering around doesn't necessarily lead to anywhere interesting. I also…

I think the same way,however there was one city I did enjoy navigating as a pedestrian, and it was a grid layout - Barcelona. I think its because the grid is just for the cars, as a pedestrian you can walk “in between” the grid parallels and there were strange walkways and “diagonal” streets etc. Such a unique place

That's just one of the neighborhoods though. But Barcelona's grid neighborhood Eixample it's a very interesting design: https://en.wikipedia.org/wiki/Eixample

Re: City street orientations around the world

#44

Post by sjwright »

fredley wrote: post #22
panic wrote: post #20

Earlier quoted context omitted.

Especially now that we all use navigation apps in our cars.

To be fair, even with navigation apps central London is a nightmare...

It is a nightmare for cars, but navigation apps do make it an absolute joy for tourists. I can hop around with busses and the the tube like an expert. When on foot I can even find the most interesting detours off the most efficient route.

Cities like London prove that a city designed for pedestrians is better than a city designed for cars, for the simple reason that cars don't scale nearly as well as pedestrians and public transport.

Somewhat ironically, successful future cities are going to look more like London than your typical large US city. More walkable areas. More car restrictions. More car exclusion.

Re: City street orientations around the world

#46

Post by Stratoscope »

Redwood City, California has a strange mix of street grids. There's a roughly NSEW grid downtown, with separate grids to the northwest and southeast angled about 45 degrees. The grids meet in confusing jumbles of triangles:

https://www.google.com/maps/@37.4861576,-122.2356476,15z

It's easy to see where the 45 degree grids came from: one parallels El Camino and the railroad tracks, the other parallels Middlefield Road. But why the abrupt change to a NSEW grid in between?

Downtown Redwood City was once a shipping harbor and a center of the redwood logging trade (thus the name). Redwood Creek ran N-S through this area, and downtown was built up around it. If you're familiar with the area, the multistory parking garage between Broadway/Marshall/Main/Jefferson is located where the "turning basin" was: a wide part of the creek where ships could turn around, with businesses along the banks.

Broadway was originally named Bridge Street, for its drawbridge that connected the two sides of the creek. There's an interesting historical marker on Broadway across from the driveway into the garage, and here are a few pages with maps and photos:

https://1084.myt.li/tours/45512040/stops/1980869008/index.ht...

http://www.redwoodcityport.com/p7iq/Assets/port_history_for_...

http://www.redwoodcity.org/Home/ShowDocument?id=1038 [PDF]

Re: City street orientations around the world

#47

Post by lotsofpulp »

ghaff wrote: post #34
bfirsh wrote: post #29

Earlier quoted context omitted.

I am from London and lived in the US for a bit. I struggled to navigate. I think it is because I navigate by modelling the shape of the streets. When all the streets are the same shape, that subconscious system stops working. I also tend to find I underestimate distances in grid systems. I think it’s because in European cities I estimate distance by the number of distinct shapes I can remember on the way there. In a…

You sort of need to know the [scale of] grid in question. In a similar vein, I find the Las Vegas Strip is incredibly disorienting with respect to distance. You're used to buildings being of size approximately X and in Vegas the casinos are 10X. So you think--oh, that's just a couple of buildings away. 30 minutes later you're still walking (in the sweltering heat).

I found this out the hard way before google maps walking time calculator when I decided to suggest walking a group from Mirage to Rio to watch Penn and Teller. I was not popular that night, especially for those in heels.

Re: City street orientations around the world

#48

Post by JorgeGT »

It would be interesting to add a color dimension with the average age of buildings registered to each street, I'm thinking about cities that Madrid where a combination of "omnidirectional" and "4-direction grid" patterns is clearly visible.

Re: City street orientations around the world

#49

Post by 24gttghh »

nautilus12 wrote: post #45

Could someone explain some actionable insights that could possibly come from these visualization?

One I pondered is how some cities have a primary axis offset a few degrees from N/S, either more to the East or more to the West and how that might be related to their location in a given time zone with respect to where the sun rises/sets to shine directly down a majority of streets. I think of this in terms of perhaps not placing most streets in a grid where you would have to drive directly into the sun around dawn/dusk for most of the year, but it probably has more to do with the local topography; bodies of water or hills in the way, that kinda thing.

Re: City street orientations around the world

#50

Post by rmc »

n4r9 wrote: post #38

I wonder how roundabouts in European cities would affect these plots, since in OSM they are often represented as a single way. The original post about this says: "OSMnx automatically calculates all of the streets’ bearings. Specifically it calculates the compass bearing from each directed edge’s origin node u to its destination node v." I suppose it depends what they mean by "directed edge". I'd expect it's what most…

If OSMnx splits each way into sequences of "node N to node N+1", then the results are based on the number of node-node connections, not the number of ways.
Post reply on HN