I found a high resolution scan of the first page of the Constitution. After deciding on my scale, I superimposed a 78x64 square grid over the scan, as well as a separate 65x64 rectangular grid for when I was building in the vertical dimension. After hemming and hawing for about year about over how to best to depict the paragraph text, I eventually decided that random shaping was the only viable option with the medium. However, using the grids as a reference allowed me to place the larger text, the indentations, new lines, and empty spaces in nearly their precise locations as on the original document.
Side note: the scale wasn’t determined by the text, but rather by needing the thickness of the stripes to be evenly divisible by the width of the brick separator. I opted to be 15% undersized rather than 13% oversized since it would be an easier task plus I thought it would look better scaled to the brick separator.




I was able to post the images on a new community on my host instance. It looks like there is ultimately always going to be some degradation of image quality if I use my instance as the host, but this workaround is a improvement over the images being posted onto the main post directly. Thanks for the help!