Comparing City Street Orientations

I remember this from 2014:

http://www.datapointed.net/2014/10/maps-of-street-grids-by-o...

Where streets are coloured on the map by orientation.

While most US cities are a wash of one colour, Europe is much more chromatic!

Tacoma has a bizarre direction pattern for its streets. Every location east of A Street is "East", and west of Division is "West". Between these 2 (and including A and Division streets), everything north of 6th Ave is "North" and south of 6th Ave is "South".

6th Ave itself has no direction name, and oddly, streets increment in both directions. North of 6th Ave is 7th Ave North, while south of 6th Ave is 7th Ave South. There is no 1st Ave through 5th Ave.

These directional names persist through Pierce County, so the vast majority of roads in the county have "East" appended to them, all the way out to Mt. Rainier (which is also in Pierce County). North of Tacoma's Narrows Bridge lay Gig Harbor and the Key Peninsula, where streets prepend the direction name with "KP" - i.e. "KPN" for North - to be more clear that they're across Puget Sound from the rest of Pierce County.

https://en.wikipedia.org/wiki/Tacoma,_Washington#Transportat...

Charlotte's interesting because it grew fairly slowly until 1970, but has almost quadrupled in population since then. If you look at it on a map, the downtown core has a regular grid pattern, but the outer regions are a chaos where the city's expansion presumably overran a bunch of small towns in the last 40 years.

This looks incorrect based on Denver's graph. The post claims its effectively 100% along cardinal directions but thats obviously not the case. All of downtown is NE/SW.

(Maybe Denver is including a lot of urban sprawl?)

https://www.google.com/maps/place/Denver,+CO/@39.7488106,-10...

The histogram for Seattle looks unconvincing for me, because the entire downtown-pike-denny area is slanted to align with the coastline ( https://www.google.com/maps/@47.6090857,-122.3407184,15z ). Possibly due to weighting?

I live near Montreal, while the street orientations are in a grid pattern, since there is a 45 degree orientation for most of them, "North" is by convention in fact NW.

It gets even more confusing on the NE part of the city, where Montreal's "East" is closer to true North and Montreal's North is closer to true West.

It'd be cool to see this weighted by street size (using, e.g., average daily traffic numbers). Granted that data is harder to come up with.

The DC plot would tell a different story then, because the avenues are all off-axis and few in number, but they're big streets with traffic light priority that handle a lot of car traffic.

While navigability of grids is nice, I kinda like it when cities have haphazard street layouts—older cities with chaotic street grids tend to be easier to walk around in (assuming you know where you're going) with slower traffic and smaller streets, not as loud since sound doesn't carry through all the buildings that block streets off (New York is loud almost everywhere, Tokyo has countless quiet neighborhoods seconds from even Shibuya Scramble), and as the author mentioned, instill a sense of adventure.

But maybe there's ways of getting these things while still maintaining street navigability—narrow streets with not necessarily rail straight, but still well-structured grids for the major streets, and maybe non-uniform/unstructred alleys could work to get both in one system.

I'd love to see one of these rose diagrams for Tokyo .. when I was there I found it very, very difficult to navigate even my local neighbourhoods, and remarked as much to my hosts - who explained to me that the reason Tokyo was laid out in what could only loosely be described as "chaotic crows feet on acid", is so as to make it very difficult for any invading armies to march directly into the city in a straight line - giving defenders lots of places to choke them off.

A comparison to old european cities would be interesting.

You would need different math for star forts such as Palmanova (https://en.m.wikipedia.org/wiki/Palmanova), or for 17th century Amsterdam, with its radial grid (https://www.google.com/maps/d/viewer?mid=1_XUR_LAcsSSE9Lmx2S...)

Such a circular grid has the advantage that it is easy to find its center.

Charlotte is really interesting. It's very evenly distributed, even compared to Boston. It's also not balanced on the north / south axis. Are there more north bound one way streets?

I much prefer older European cities street layout. It's usually a mix of circular and rectangular patterns, but it's still usually easy to navigate by roughly assuming each turn is 90 degree. And it's walkable.

In Poland especially interesting is Szczecin layout. It seems rectangular when you're there, but when you look at it from height it's actually made mostly out of triangles.

You turn right 2 times and you get to where you started :) Messed with my head so much when I was there.

The histogram for Charlotte is asymmetrical... what's up with that? One-way streets? (The same is true for some other cities in the follow-up also.) The article doesn't say.

I'll never forget the first time I drove in Boston as a teenager. I came up to a forest of stoplights, some red some green hanging haphazardly at an intersection of seven streets, and my burning question was: which one is mine? I went when the driver behind me started honking.

Manhattan burned to the ground so they got a do-over on their street grid. Any East Coast city that wasn't centrally planned (e.g. DC) or burned to the ground will resemble Boston.

Boston also more than tripled in size due to land reclamation, which included leveling some hills. The contours that the current roads follow may have made more sense when Boston Common was a muddy beach and everyone walked to where they were going.

https://www.reddit.com/r/boston/comments/1s92xa/cool_map_of_... https://imgur.com/gallery/0C349u9

you think boston is bad, try this: https://www.google.com/maps/place/Varanasi,+Uttar+Pradesh,+I...