I've been thinking about all the discussions that inevitably come up regarding what you should, and should not do, when rigging. Nobody ever seems to mention which link in the rigging chain you want to be the weakest one. They go on about which ones you can't do much about (the main anchor limb) or which one failed on them when they thought that 2,000 lb. chunk of wood was only 500 lbs. but they seldom express any real grasp of the notion that planning a rigging setup isn't just about choosing components that work well when everything goes as planned... it's also about trying to control the results when everything blows up in your face and the proverbial shit hits the fan.
I've always tried to pick the one component of each rigging system that I believe will cause the least damage when Murphy strikes. That can vary, depending on what it is you're doing that day, but there is almost always one part of the system that is more expendable than the others... and that will induce the least amount of damage when it breaks.
That seems to be, so much of the time, the point of attachment for the piece you're attempting to get to the ground. So, if you can buy a ten-pack of 40kN steel carabiners for a price that works out to be $11.20 each (I have gotten them that cheap) then this is a small price to pay. I don't want the rope, anchor, friction device or tree to fail, so the carabiner is a logical choice for the weak link. If the anchor point (limb) fails, then you have a lot more stuff than just the load coming down. If the rope fails, that's going to hurt First National of Billfold a whole lot more than a broken, $12 carabiner.
Of course, you still try to stay within the limits of your weakest link, but when that doesn't quite turn out the way you expected... it seems to me that having the results turn out pretty much as you would expect makes it a lot easier to focus on minimizing the damage. Like keeping anything/anyone from being under the load, which you should be doing, anyway. Having to worry about what's under the anchor, or a rigging line redirect, or who will get busted upside the head when a portawrap goes flying... none of that should be a constant distraction. There should be an expected, probable weak link that gives you a much better chance of predicting the results of a misjudged load size or unexpected shock load.
Personally, I like that weak link to be quite a bit under the rating of everything else in the system. I've even used cheap, $8 aluminum screw-link carabiners with my 1/2" systems... just to be sure that it breaks instead of something else.
