That was a funny coincidence. I was just looking for an article (might have been a letter) they did on glue joints, specifically, jointing with bare equipment vs. waxed and buffed equipment.
I couldn't find what I was looking for.
I remember reading the article. IMHO, the article was intended to be a glue strength article, the methodology used tested the physical structure of the joints and the gap filling abilities of glue.
In my shop using observations of having tearing apart some joints, there isn't much difference in the strength of modern glues. When following the directions of the glue, most glues are stronger than the wood that they are gluing together. (Using tight fitting joints and clamping for the required time seldom have the glue fail.) When I have forced a glue joint to fail it is almost always the wood that fails and not the glue. What I have noticed is that polyurethane glue is very strong in large areas but as the glued surface area decreases the likelihood of glue failure increases. This may be due to inadequate clamping pressure???
I use Titebond and Titebond III mostly in my shop. TB for mostly melamine panels and TB III for most wood uses. (TB III seems to be easier to remove after curing from corners using a chisel. TB III is a real bear to remove from melamine and Formica surfaces.)
If I'm gluing a non-structural component into an area that can not be sanded after the glue cures, my choice is polyurethane. The foam comes off easily with a chisel or Shur-Form scraper.
There is one interesting note about TB III from my experience. I was making a face frame out of Poplar. The joint structure was pocket screws with TB III. A joint was incorrectly marked and assembled in the wrong spot. The joint was a typical end grain to edge (long) grain like almost all joints in face frame construction. After about 45 minutes I took the joint apart. After I removed the pocket screws, the amazing part is that the glue did not fail in some places but rather the end grain pulled some of the edge grain loose.