Adding a reflector to keep neutrons bouncing around longer, which I don't think was done in the original Little Boy weapon (the gun-barrel one), is a simple method of raising efficiency. The gun-type nuclear weapon is extremely simple to begin with; Manhattan Project researchers were so sure it would work they didn't test it, whereas they did feel compelled to try the implosion design, in the Trinity test. But for reasons I don't recall (which I think relate to neutron flux and neutron speed), the gun design works only with uranium, or at least is never used with plutonium. By the way, the boosting method you referred to brings a spot of fusion into the picture, by way of deuterium and/or tritium. A mixed fission-fusion reaction isn't something you can get with a quantity of enriched uranium alone, which is what we began by talking about.

You may have hit on something there, as 100 pounds of highly enriched uranium would constitute less than went into the Little Boy weapon that exploded over Hiroshima, Japan, in 1945; judging from various online sources, that weapon contained about 64 kilograms of HEU, and 100 pounds amounts to only 45 kg. On the other hand, the design of the Little Boy weapon was (again, judging from online sources) highly inefficient. My rough guess is that 100 pounds of HEU would be enough for two or three weapons that employed an efficient design.

Not reading the linked excerpt right now, because I've read reviews of the book that recount much of what it tells. None of the standout incidents that reviewers mentioned—the H-bomb dropped into a backyard, the H-bomb lost (for a while) off the coast of Spain, the ICBM that blew up in its silo—were new stories; they were things that happened some time ago, which attentive people have long known about. One of the incredible things about fission and fusion weapons is that the horror stories have never yet lost their power to unnerve us. Not sure what that means, but I think it's either (A) we won't get over nuke shivers until we've managed to come up with something even worse, if that's possible, or else (B) there's no way we'll ever get over the threat of being reduced to atoms.

Regarding Schlosser's book, I wonder what's the most recent we-almost-lost-it story the author relates. The weapons and related bad stuff (e.g., stockpiles of plutonium and highly enriched uranium) certainly haven't gone away, yet I don't recall hearing any near-miss tales lately. Maybe I'm not as attentive as I used to be; maybe we're overdue for another scare; or maybe the older types of accidents have finally become less likely and the newer problems are more diffuse, such as plutonium unaccounted for at a plant, which could be theft that'll lead to a mushroom cloud and could just be bad accounting.

Networks don't drive social change; people do. This is akin to saying "guns don't kill--people do," a position that some find objectionable. But in both cases, it can make a difference what instruments are available. Social change, protest movements, and other forms of rebellion may be facilitated by one's network, whether it's the telephone network or Twitter; similarly (though this analogy is getting strained), a murderous rage can be facilitated by a handgun in the desk drawer. Though I haven't read it yet, Malcolm Gladwell's article demands to be read--all of his articles do, in my experience--because he's probably saying something different, or at least more subtle, than that social media don't promote or drive change: that seems too obvious for him.