16 March 2023

More New Suspish on Stark Street, or, It Comes From Eugene


When I saw SUSPISH on the back of the PPD Sunshine Division building on SE Stark just east of 122nd, I was delighted. I love the style of this artist and remembered, as I pointed out here, that we've seen them before.

But there's more delightfulness ... they've returned to the original building we first saw them on. 14410 SE Stark is a kind of a warehous-ey building that at one time, more than a decade ago, held a Hotel/Motel Furniture Liquidators. It closed around a decade ago and has stood vacant ever since. Now, the first time we saw Suspish, it was on the east wall of the building, the one that faces the Franz Bread Thrift Store next door. It was painted over in due time.

Now, it's on the buildings facade:

This differs from the other Suspish, in that we have the angler-fish catching its prey but, moreover, it's more or less the same as the first Suspish that appeared in April 2021, which I include here for comparison:

The addition of sparkles in the first image gives it a little more visual delight.

But there's more. That nifty web thing, the hashtag, accidentally showed me that Suspish isn't just a couple of fun graffiti on the outer east side of Portland. 

Suspish is a Eugene thing. perhaps come to visit, to stop by and leave something in a Banksy-esque way. Eugene Weekly spoke with The Artist back on 2022. They're something of a street-art city mascot down that way. 

04 March 2023

How To Have A Logical Address System In An Illogical Street Pattern


Georgia has an amazingly-adaptable system for figuring out a regular system of addresses in counties that have irregularly grown road systems. It's impressive and works very well.

I'd like to introduce you to a document titled Street Addressing Standards and Guidelines for the State of Georgia, created by the Georgia Spatial Data Infrastructure GIS Coordinating Committee, Framework Transportation Technical Working Group, and released in August 2000. It really is an amazing document, and you can get your own copy of the PDF here:

It's axiomatic by now that a consistent, logical system of addressing promotes various public and private goods, from predictable private wayfinding to ease of location for emergency first-responder services. The organic way many Georgia cities grow kind of defeats this, though; take a look at the city center of Atlanta - various sized grids at various angles, no smooth transitions between them - and one begins to see what the challenges of creating such a system might be.

To this end, the group with the long-winded name came up with a complex-yet-simple system. I'd ask anyone interested to read the document (which also goes into considerable depth with respect to establishing logical street-naming rules which promote setting up solid address databases) but it is kind of dry, so here's the TL; DR abstract. It has 2 main moving parts:

1. Identify two intersecting roads, or two intersecting alignments, that cross the county. They meet at the reference point, which is the originating point, from which all addresses radiate. This divides the county into rough quadrants, which can be designated NW, SW, NE and SE.

2. Determine addresses as a function of distance from that reference point; the rule they suggest is one house number per 20 feet of street length as measured from the centerlines of the street. 

The illustration right demonstrates the idea of finding two intersecting trans-county alignments which meet at the reference point and creating rough-but-roughly equivalent quadrants (the county used by the illustrator is Lowndes County, Georgia and the reference point is in city center Valdosta, for what it's worth ... and a look at the county shows they haven't chosen to implement this system, so it amounts to a demo). 

The real magic of this system comes in tying the house numbers to distance relative from the reference point. This is necessary because if a Georgia town starts off with a rectilinear grid, they typically do not hold; they devolve pretty quickly into a county road network that's based on paths that indigenous people took and that settlers created to get to their properties as they developed the lands. There isn't a regular gridded road pattern, and there isn't going to be one ... a regular amount of block numbers per 'block' is difficult if not next to impossbile.

One drawback is that even hundreds don't fall on street intersections, and that's something to learn and get used to, but any address system can be learned and they've been doing things this way in Atlanta since the 1920s, so it may be a stumbling block in the beginning but that wont' last for too long.

It occurs to me that bits and pieces of this logic can be applied to ordering other towns depending on the form of the street network. There's a thought experiment I plan on exploring quite soon, and I'll write about that here. But not quite yet.