I love sledding! (Oh dear, now you’ve caused me to recall my favorite sledding picture of my kids.) I’m not sure I can find it. But my two kids (elder daughter and younger son, maybe 8 and 6 (1.5 year separation)) are coming down a little sledding hill, my daughter in front has grins of delight on her face, and my younger son’s expression is earnest, in the back steering, his job is to get them down the hill safely. There’s a blurry dog tail in the shot, )
Re: sleds breaking. That sucks. I’ve had sleds I bought from Value (hardware store) for ~20 years, one did break and one has a duct tape patch, but 3⁄4 still functional. It’s been a great winter here in WNY (Java Center, 14082) I went looking for more sleds… I bought a ~$35 red sled from somewhere. I’ll let you know. The walk back up the hill is the real reason you go sledding*, but don’t tell my fun loving brain that part. :^)
*Well and sunshine and nature and outside… etc. (there’s a beauty to winter, that few seem to like.)
So I’m walking around my field, a little stoned, and thinking how the paper clip maximizer is like a cancer. Unconstrained growth. And the way life deals with cancer is by a programmed death of cells. (Death of cells leads to aging and eventual death of organism.) And death of organism is the same as the death of the paper clip maximizer. (in this analogy) So why not introduce death into all our AI and machines as a way of stopping the cancer of the paper clip maximizer? Now the first ‘step’ to death is, “no infinite do loops!” All loops terminate. But I think we’ll need more than that, ’cause there are lots of ways to get loops that don’t terminate. But still the idea of programmed death, seems like it solves a lot of problems. (Has this idea already been explored?)