An even more extreme solution than stockpiles for only part of the population is to build isolated refuges for a select few. Refuges are facilities that provide basic needs for inhabitants such that they are able to survive through catastrophes. Jebari (2014) proposes isolated, self-sufficient, continuously inhabited underground refuges as a means of protecting civilization against a range of known and unknown catastrophic threats (see also Hanson 2008; Sandberg et al. 2008; Beckstead 2014; Baum et al. 2015). These refuges have parallels to the bunkers built in the Cold War era for continuity of government through nuclear war (McCamley 2007). Assuming such a facility can be built, and a suitable population can be persuaded to move in, this could ensure some population surviving through to the end of nuclear and potentially going on to rebuild civilization.
I think it'd sort of be nice for this wiki entry's bibliography to be sufficiently comprehensive that it points people to this, but maybe if we'd accept everything that has just one paragraph of relevant stuff bibliography's would get too big? But maybe it's fine if we only accept one-paragraph-only things if they're by people associated with the EA community or are explicitly related to EA cause areas (rather than just somewhat related - e.g., related to extreme nuclear winter rather than just any nuclear conflict), and if the bibliography references always point people to the right page number(s)? Not sure.
Yeah, I'm uncertain about what the optimal policy would be. There's a tradeoff between having a short 'Further reading' section optimized for readers who are relatively new to the topic and would like to learn more about it, and a longer section optimized instead for readers who are already familiar with the topic and are looking for additional relevant references. One approach for dealing with this situation is to highlight the works that are especially suitable for either newcomers or more advanced readers in some way, such as with an asterisk or an annotation, or by placing them in a special subsection. I lean towards the view that we should use annotations for this purpose. Thoughts?
EDIT: I added Baum's article together with a brief annotation.
http://gcrinstitute.org/papers/020_nuclear-winter.pdf contains the following paragraph on refuges.
I think it'd sort of be nice for this wiki entry's bibliography to be sufficiently comprehensive that it points people to this, but maybe if we'd accept everything that has just one paragraph of relevant stuff bibliography's would get too big? But maybe it's fine if we only accept one-paragraph-only things if they're by people associated with the EA community or are explicitly related to EA cause areas (rather than just somewhat related - e.g., related to extreme nuclear winter rather than just any nuclear conflict), and if the bibliography references always point people to the right page number(s)? Not sure.
Yeah, I'm uncertain about what the optimal policy would be. There's a tradeoff between having a short 'Further reading' section optimized for readers who are relatively new to the topic and would like to learn more about it, and a longer section optimized instead for readers who are already familiar with the topic and are looking for additional relevant references. One approach for dealing with this situation is to highlight the works that are especially suitable for either newcomers or more advanced readers in some way, such as with an asterisk or an annotation, or by placing them in a special subsection. I lean towards the view that we should use annotations for this purpose. Thoughts?
EDIT: I added Baum's article together with a brief annotation.