Actually, if you read a few history of science books what you'll see is that a lot of what we call "the scientific method" came out of arguments just like this. If you look at organizations such as The Royal Society in London in the 1600s they started out as glorified debate clubs arguing the new ideas of the day. Do you want to convince others you are right and win the debate? Better show evidence. Even better, show evidence that they can replicate for themselves. From this came the notion of more formal evidence which later evolved into "the scientific method". Optimistically, widened public debate via blogs might be the best thing that could happen to science, leading to more rigor on how best to validate and confirm new ideas.

1. If you read the article they do actually discuss events like the Cambrian Explosion (punctuated equilibrium). On p. 5 they point out that:
    a. Many of these "explosions" came from existing species expanding into available niches and exploiting existing gene expression plasticity which does not necessarily increase complexity.
    b. There is no reason to think that we don't see significant complexity growth during "stable" periods where there is a large pool of living organisms which can evolve and share new genes.

2. Also, it is worth noting that even if there are periods where genetic change is slower or faster, if this change is stable on AVERAGE over billions of years then this approach may still give correct results.

So, yes, they did consider events like the Cambrian Explosion and I don't think their theory is invalidated by such events.

In fact you can quantify the effect that the rich tend to stay rich in the US and the poor are stuck in poverty using something called the "intergenerational mobility index". Basically, the US is one of the worst countries in this regard and it has gotten that way mostly over the last few decades. Here is a summary from "The Price of Inequality," by J. Stiglitz, p. 18.

"It is at the bottom and the top where the United States performs especially badly: those at the bottom have a good chance of staying there, as do those at the top, and much more so than in other countries. With full equality of opportunity, 20 percent of those in the bottom fifth would see their childen in the bottom fifth. Denmark almost achieves that - 25% are stuck there. Britain, supposedly notorious for its class divisions, does only a little worse (30%). That means they have a 70% chance of moving up. The chances of moving up in America, though, are markedly smaller (only 58% of the children born to the bottom group make it out)."

Data on these claims:
"Some 62% of the children of those in the top quintile wind up in the top 40%"
from "Getting Ahead or Losing Ground: Economic Mobility in America:" http://www.brookings.edu/research/reports/2008/02/economic-mobility-sawhill

Why do you say that thread synchronization doesn't work with exceptions? Seems to me that this isn't that much different than database transactions. Grab the mutexes you need, perform the operations, commit on success, rollback / go to known state on failure. Release / destruct mutexes at end of transaction. Just like with normal exception programming the main point is to check that exceptions go to a known state.

I'll put in a vote for "Great Sky River" by Gregory Benford. This was a nebula award nominee I picked up about 10 years ago and gave me nightmares ever since. The background is a small group of humans on a planet which has gone "mech". To try and survive, they have increasingly replaced their bodies with robot parts. It's not enough: the machines are faster, smarter, and their motivations largely incomprehensible as they hunt the survivors down. Eventually, they come to understand part of the reason they are being hunted, cut to pieces, and brain dumped. It's not because humans even matter anymore, they are shown the museum full of deconstructed and "reinterpreted" human components they are to become a part of. This was a very believable "hard sci-fi" novel which laid out a quite convincing and bleak possible future.