A couple of weeks ago I posted an explanation of how to get to Saint-Germain-Laval, France, the hometown of Daniel Greysolon Sieur du Luth, the person after whom the city of Duluth is named. I was not entirely sure if my own trip there would happen. It did, and that post has now been updated with more accurate information and a few explanatory photos for anyone in the future who might be interested in visiting.
Part 1: Background
About five years ago I came across the article Du Luth’s Birthplace: A Footnote to History. In 1966, the author visited the French village where Duluth’s namesake was born and documented the few traces of him that remained. The article is well researched (all the quotes from Sieur du Lhut used here are taken from it) but the images are low resolution scans of black and white photos taken more than 50 years ago. I wanted to see more of what the place actually looks like.
During the pandemic, I colorized six early Duluth photos, which was absurdly time consuming but seemed like as good of a way as any to spend some evenings inside. A friend of mine recently informed me that Adobe Photoshop now has a tool that will colorize photos automatically with far better results than my drawing over pixels method. He was somewhat correct.
Previously I posted some AI generated images that could plausibly be of Duluth but it was difficult to create images that were distinctly Duluth. The absurd specificity of Jim Richardson’s interactions with ChatGPT inspired me to try again but with a different process.
The Duluth News Tribune recently published an article about the Downtown Task Force’s recommendations to improve conditions in downtown. This summer, I spent some time walking through the Skywalk system and was a bit shocked by how empty it was. The summer might not be the most popular time to use the Skywalk, but it wasn’t just the absence of people. So many of the shops that I remembered were gone. I didn’t intend to make a themed photo series about this, but I had my camera and kept turning a corner to find another impossibly long, completely empty hallway.
A Van Gogh style painting of a 1920s cargo train traveling on a winter night though an evergreen forest next to a huge blue lake with the aurora borealis in the night sky.
DALL-E is an online tool that uses machine learning to generate digital images from plain English text descriptions. You type a description of something real or imaginary and the program does its best to create a unique image based on that description. After some time on a waitlist, I recently received an invite that allows me to create and download a limited number of artificial intelligence generated images per month. This came at a good time, as I recently found a watercolor print of the Duluth hillside in a Lincoln Park shop that I liked quite a bit but could not afford. I decided to use some of my AI image credits to see if I could get the automated system to produce Duluth art of at least somewhat comparable quality. In the examples that follow, I describe this process, showing what worked and what did not. The captions of each picture show the text query that generated the image.
Some years ago, I had a Duluth nameplate hanging from the back of my bicycle, which I suppose is why I impulsively bought another when I saw it recently in a Canal Park souvenir shop. I vaguely remember the one I had purchased in my younger years having a generic cityscape and not actually showing Duluth.
This one seemed no different, and might even be the same design, but this time I noticed that the building on the far left had a rather specific architecture. Checking with a group of skyscraper-obsessed friends led me to a conclusion that I should have reached myself: this Duluth souvenir depicts the skyline of St. Paul.
Sometime back, I included an aerial photo in a PDD comment and realized that because they are taken from straight overhead, the photos on Minnesota Historical Aerial Photographs Online can be matched up pretty easily with Google’s current aerial imagery. And then I put that thought aside for quite some months until I finally came back to it and put together this seven-part series of aerial photos showing places in Duluth that have changed somewhat dramatically over the past decades.
The fake internet court podcast Judge John Hodgman, where pressing issues are decided by Famous Minor Television Personality John Hodgman, Certified Judge, once again mentioned Duluth, this time at length in the concluding segment of its Great Lakes Beach report.
In the new Netflix series Colin in Black & White, former NFL quarterback Colin Kaepernick serves as the narrator for his own story of growing up in California’s Central Valley with white adoptive parents. At the very beginning of the third episode, a modified version of a photograph taken at the 1920 lynching of Elias Clayton, Elmer Jackson, and Isaac McGhie in Duluth is used as part of a montage explaining the concept of white privilege.
A friend of mine from Seattle was eating at Jimmy Mac’s Roadhouse in Renton, just south of “the Emerald City,” and sent the above photo to me. At first it seemed like a manufactured novelty sign because searches only returned other copies of the same sign. But then I found a Minnesota Historical Society listing for a paper crate label from the same company from around the 1950s, so it appears to have been a real business.
Microsoft’s Flight Simulator came out on the Xbox last week and allows players to fly anywhere in the world, including Duluth. It combines information from Bing maps with an algorithm that builds 3D representations of the landscape, creating shapes and textures when data is missing or incomplete.
The video below shows a simulated flight over Park Point.
As part of Perfect Duluth Day’s long-running quest to document every reference to the city, no matter how minor, I add this entry, discovered today while catching up on the fake internet court podcast Judge John Hodgman, where pressing issues are decided by Famous Minor Television Personality John Hodgman, Certified Judge.
Geoguessr is a website that features variations on a rather simple game: you are shown a location through a modified version of Google Streetview. You must guess where you are by marking the location on a map. The labels and location marker normally added by Google have been removed, so you must rely on a compass and clues from the environment. The closer your guess is to the correct location, the more points you get. Each game consists of five rounds. The tops scores appear on the main page for each map, with a tie going to the player who finishes the fastest.
