Jeff, with all due respect, I do not believe I am trying to make it more complicated than it is. It is just that my approach is different than yours and I need to better understand the underlying logic to be able to better approach it.
If I were going directly to print or to a specific size on the screen the LR/PKS method might work fine for me....and it is quite good for those doing this and who do not want to think about it. However, my workflow for output is different.
If I were using Photoshop for final output, I would sharpen the file so it looked sharp, but not over sharp, as iewed on the screen. I would do this at 100% and do it the same whether it was a 6K image or a 2K cropped one. There is a reason for this and it worked successfully before using LR.
Today, most high end photo hosting sites, such as Smugmug and Zenfolio, allow you to upload full size images. These images are then downsized, behind the covers, using Lanczos, or better, interpolation, to progressive smaller sizes so that the image the viewer looks at appears to be automatically resized to the monitor or window the user has. If one were doing this with simple resizing, even with bicubic, you would normally get jaggies or halos....I know, I have done it...but with these services you do not.
Qimage does similar for printing. I will take a properly sharpened, but not resized for printing image. Based on my selection of print size, it will resize the image for the size selected and the ppi the printer wants (for Epson 4900 usually 720 or 360). It does the resizing with high end interpolation and applies "smart sharpening" so the sharpening is properly adjusted for the actual output to the printer.
I find this workflow to provide great quality output images without the need to provide multiple output image files for each screen size or print size that I may want or need.
The problem (but for some people, the strength) with LR's output sharpening is that you cannot view the output on the screen to evaluate it. This is why I am Tediously, to you) asking you to provide better insight into the differences in what is done when selecting 'screen' or print and how the ppi designation changes what the sharpening routines do.
Thanks, John