> I think I was wrong in using iron for the stiff pipe. It's ductile just
> like copper,

??? Copper and iron have very different material properties. That is why engine blocks are made of iron, and not copper.

> PVC pipe vs. copper pipe approximately the
> same size, designed for the same task, as in for home plumbing. Drop a
> brick on the copper pipe and it dents. Drop a brick on the plastic pipe
> and it cracks. Which has failed? At least the copper one doesn't leak.

However, a brick dropping on a pvc pipe will most likely simply bounce off. The copper pipe will be dented, resulting in a flow restriction, a hardened brittle spot, and cosmetic marring. A solder joint nearby will probably be damaged to failure by the impact, while a pvc fusion joint nearby will most likely not be affected by the impact.

> Cyclic loading would be a whole 'nother analysis, different from impact
> loading.

Impact loading also varies tremendously. There is quite a difference between bellying up on a smooth boulder and impaling the boat on a spike. That's why spears are pointy and knives are sharp.

> Just because a material yeilds first does not mean that it fails first.

If the object was to not yield, the material has failed. I really don't think you're going to want a kayak that yields readily in use. Even if it were to retain watertight integrity, the paddling characteristics would be most adversely affected.

> I'm not saying stiffer cannot mean stronger, I'm saying that stiffer is
> not necessarily stronger, and that stronger can mean weaker.

Ok. It was looking to me like you were saying it as an absolute. Remember too with the latter point, you can make the entire body as a whole stronger by weaking selected areas (I beams as an example), but you cannot make the entire body stronger by weakening it throughout. All over variables held constant of course.