I have thought about the joint I had presented earlier, and have actually made a few modifications to it. The reason is that, as designed, it relies exclusively on the shear strength of the pegs or bolts. It seems to me that it is necessary to have an interlocking joint that resists shear by its own virtue, with pegs/bolts only adding to this quality. It's hard to explain what I am taking about, when I have the opportunity I might put up some drawings to show what I am talking about.

two things I notice about your design are ingenious:
1: The use of the lancet arch. This arch is the strongest form of the common arch. The Romans would have used it if they could have figured out how to make it. This is one of the greatest achievements of the Middle Ages.
2: The arch continues to the ground. If the arch is fastened directly to the foundation, the problem of outward thrust is totally eliminated. The ground itself will handle this for you.

The concept with my original design was that the interior faces of the timbers would form a continuous, uninterrupted arch. This was accomplished by applying two octagons on top of each other, 1 rotated in relation to the other. The reduced ends made a similar figure on the exterior and helped direct load forces through the arch. This design was adapted to a simple arch, and would require a lot of revision to suit a lancet arch.

Just a suggestion, but it might be interesting if you designed your structure with cross vaulting, which would give you a very interesting interior



The advantage of the groin vault is that the loads are 100% transferred to 4 spots, and the two crossed vaults oppose the other's outward forces. The result is you have almost purely vertical forces. The arches will hold each other together. This fact is the real reason why medieval churches were built cruciform, it reinforced the otherwise unstable long vaulted halls.

The disadvantage is that the framing would be quite complex.