4 ms·
When you see an article on HN about how someone took some code, profiled it, and increased its speed by 10 times, I think it's tempting to credit virtue to the
by jerf 7y ago
When you see an article on HN about how someone took some code, profiled it, and increased its speed by 10 times, I think it's tempting to credit virtue to the person who sped the code up. But I think you also have to remember that the only reason you could speed up the code by 10 times in the first place is that the original code left that much on the table. It isn't just that there is a person who was skilled enough to speed the code up, it is that somebody else left the opportunity behind in the first place.
For similar reasons, I find it quite likely to be engineering-meaningless [1] when someone takes a life form that lives for a 2-3 weeks and extends that by a factor of 5. We humans have the "live for fifteen weeks" problem solved. Optimization being what it is, it is very likely that whatever that improvement is already effectively exists in us. (Possibly not directly, possibly through some other mechanism that doesn't completely match, but it's probably done.) It is, unfortunately, much less impressive to "optimize" an organism that has left that much on the table.
Improve an elephant's lifespan by 50% with an intervention when they're already fully-grown and I'll really take notice. Unfortunately, there's all kinds of reasons why that's hard....
[1]: Contrast with meaningful to science. There's nothing wrong with studying this sort of thing as steps to both eventually treating aging, and just plain scientific understanding of the world, in the mean time. But something can be scientifically useful, and even perhaps someday eventually be useful to engineering through its contribution to a dozen other discoveries, while being "engineering-useless" today. Remember the context we're speaking in is in being excited about all the longevity news in the media and to hope that maybe it'll be ready by the time we personally need it.
- reubenswartz 7y agoThis is a very important point. Human lifespan is already an outlier compared to our primate cousins and other creatures of similar (or even larger size). I'm sure there's plenty of room for "optimization", but it's not as easy as it is with nematodes or mice, which also get a lot of attention in the general science media.
- HSO 7y agoBy the same token, there are animals who have the „live for 400 years“ problem solved so there must be sth left on the table yes? Other animals have the „regenerate this body part“ problem solved, same.
- gitanovic 7y agoSome have the never get cancer solved (I read that about chickens... but I can't remember where)
- D-Coder 7y agoElephants get much less cancer than their body size suggests. Lots of articles but here's one: https://www.sciencealert.com/elephant-cancer-resistance-noncoding-dna-accelerated-regions-human-health https://www.sciencealert.com/elephant-cancer-resistance-nonc...
- jerf 7y ago"By the same token, there are animals who have the „live for 400 years“ problem solved so there must be sth left on the table yes?" Note my claim wasn't that optimization isn't possible, just that the impressiveness of the optimization is inversely proportional to the amount of performance left on the table. In terms of that being a promising avenue, I'd observe all the animals with long lifespans I'm aware of seem to achieve that with a quality of life that would in humans be considered very elderly behavior, due to low energy consumption. People generally want eternal youth, rather than eternal elderliness. I don't know of any 200-year-old animal leaping about from tree to tree or anything like that. Body part regeneration seems a lot more like a solvable problem to me than generalized "fix aging". While we know we're lacking spontaneous body part regeneration, we also know the "programs" for growing parts are in there, since we used them once. (As I've observed before on HN, "solving aging" would be a lot easier if we just had to produce organs with the same DNA or something. It's a bit body-horror-esque, but if we could just grow clone bodies and transplant things in we'd probably be pretty close. But there's that whole problem where we'd be happy with a new heart, we'd be happy with a new liver, we'd be happy with a new left arm, but we stubbornly want our old brain....)
- 1996 7y agoI don't want my old brain - just the information stored inside.
- rsynnott 7y ago> Optimization being what it is, it is very likely that whatever that improvement is already effectively exists in us. This only really works if there's evolutionary pressure on humans to live forever, though. And there probably isn't. We live longer than most mammals (even without good nutrition and modern medicine) but that's arguably because we're somewhat social animals; we provide a useful function after we stop reproducing. But there are limits. While having grandparents still around might be an advantage (and likely to make a prehistoric human more likely to pass on genes for long life), there are diminishing returns; having great-great-grandparents around is much less interesting. Not to say massive human life extension is practical or even possible, but we shouldn't necessarily assume that our lifespans have already been fully optimised.