It is finished.
I ended up doing all of the 12-tablet charts in the 3-hole chapter of Tablets at Work. The photos show the entire length of the band, with a lot of overlap from photo to photo to make sure I didn't miss anything. I'm not showing the backside because it's kind of irrelevant, even though parts of it actually look kind of cool.
As you can see, I used 2 strands of white and 1 strand of teal per tablet. There are 12 pattern tablets (3-threaded) and 2 edge tablets (4-threaded) per side, for 16 tablets total. All the pattern tablets had a / tablet orientation (S-threaded) except for the edge cards, which were / \. The weft was the same color and thickness as the white background threads of the warp.
In no particular order, I did charts from the following pages of the 3-hole chapter (which has a dark blue heading and is on p.476-523 of Tablets at Work by Claudia Wollny):
My conclusions and thoughts and what-not:
My favorite patterns and tablet manipulations are the very textural ones -- waffling holes plus or minus floats of the threads. These are what I had thought of as the Icelandic Missed Hole technique.
The ones that are more like traditional diagonals patterns are nice enough, with the little line of holes adding some textural interest.
A lot of the patterns use ideas about modern diagonals -- tablets moving in sets of two, two tablets and two rows at a time.
Three hole feels very different from two hole, though some of that might be because I like weaving the Latvian two hole designs and these three hole designs feel like they are coming from a different designing/weaving tradition.
This technique has some similarities with Sulawesi as well as, of course, diagonals. And float patterns. And double-face. Heck, the more tablet weaving I do, the more connected everything feels to everything else. All of the techniques are variations of other techniques. Or so it sometimes feels like.
Wollny gave a very simple/universal/general tablet setup, suitable for anything. Depending on the design, it might make more sense to have the tablets in a long V formation (the tablets on the left half of the pattern set in one orientation, either / or \, and the other half set the other way) for symmetric designs. Or set them Sulawesi style, with two tablets set / alternating with two tablets set \. Or maybe this way ends up working best for bands that will have floats and hole waffling.
I'm glad I did the sampler. I have some ideas for my own designs, for the next time I do this technique.
And at some point, I'll start adding brocade, since that seems to have been a fairly common element of these bands in the archaeological record.
All of the above are subject to change as I think about it more and gain more experience and knowledge.
What's next? I'm not sure.