It was an interesting article, but the one thought I have is that this is one more argument for using epoxy fillets. Using the quick epoxies (5-min is common, and I use 90-second stuff all the time), one can get strong, "finger smooth" fillet joints (wear gloves!) in a very short amount of time and shrinkage is not a problem.
Interesting article - I can tell you from personal experience that his recommendation for making fillets with shrinking glues (like Elmer's wood glue - a common LPR building adhesive) is a good one.
For my Der Red Max clone, I did exactly as the article suggested: I laid on a *thick* line of yellow glue in the fillet joint area, and left the rocket in the horizontal position for that glue to dry and not disturb or "run my finger" in the glue. It shrunk tremendously into a very strong fillet. Noticeable in appearance, but very sturdy.
This did make for the application time for those fillets to be quite long - a (approx.) 1/8" thick bead of yellow glue takes a few hours to cure enough that it won't run or glob up when you move the rocket to apply the next set of fillets. But id *does* work very well.
Soon after that build, I discovered the Fix-It epoxy clay from Apogee, and have been using it ever since. It's comparable in finished appearance, and the body tube will fail before the fin/fillet/root joint does. I suspect that the yellow glue fillet on my DRM is comparably as strong (and a tad lighter,) but the epoxy clay is just easier to work with IMO.