Daniel Lemire's blog

, 5 min read

Science and Technology links (April 29th, 2018)

  1. Our heart regenerates very poorly. That is why many of us will die of a heart condition. Harvard researchers find the mice that exercise generate many more new heart cells. The researchers hint at the fact that you might be able to rejuvenate your heart by exercising.
  2. Cable TV is losing subscribers faster than anticipated. In related news, cable prices are rising at 4 times the rate of inflation. That is, just as cable TV is facing stiffer competition from Netflix and others, it is also raising its prices.
  3. The rich get richer while the poor get poorer. That’s called the Matthew effect. It exists in science: young scientists who get lucky early on tend to do much better funding-wise. But why is that? It seems to be mostly because people who get bad news early on give up or become less aggressive:

Our results show that winners just above the funding threshold accumulate more than twice as much funding during the subsequent eight years as nonwinners with near-identical review scores that fall just below the threshold. This effect is partly caused by nonwinners ceasing to compete for other funding opportunities, revealing a participation mechanism driving the Matthew effect.

It follows that a key to success is to take failure well. It is like a superpower.

  1. A dietary supplement, MitoQ, is believed to rejuvenate blood vessels in mice. A new study finds that the effect is similar in human beings. (I am not taking this supplement.)
  2. When health workers go on strike, mortality stays level or decrease.
  3. Student strikes negatively affect performance on mathematics tests, but not on language tests.
  4. Naked mole rats do not age biologically. That is, their mortality rate does not increase with age. It is a rather unique feature among mammals and we do not know exactly what makes it work. Lewis et al. found that naked mole rats have relatively low levels of some amino acids in their blood, levels resembling those of hibernating quirrels.
  5. Should you be eating more fibers? Maybe not:

increasing fibre in a Western diet for two to eight years did not lower the risk of bowel cancer (…) after four years participants receiving dietary fibre had higher rates of bowel cancer.

In concrete terms, this makes me question the belief that whole wheat bread is better for your health than white bread.

  1. It seems that Neandertals had boats of some sort, going from island to island in the Mediterranean.
  2. Years ago I produced a small podcast for my students, in the pre-iPhone era. Feedback from my students was not great: they complained that they simply did not have time for such things. I used to listen to podcasts myself, but the hassle of selecting, downloading and managing audio files was too annoying. I had since assumed that podcasts had gone out of favor. I have recently rediscovered podcasts. Apple reports having half a million shows on offer, and they have counted 50 billion downloads. It seems that podcasts are all the rage right now. I suspect that this has to do with the high quality of cellular network connections. I can listen to podcasts down in Montreal’s subways. Though the monthly fees for my smartphone are high, the service is great… usually superior to what I get with wifi at the office. This almost makes me want to resume my own podcasting, but I’d have to give up other activities to make time.
  3. Progeria is a terrible disease that appears like accelerated aging. Many kids suffering from progeria won’t live to become adults. In a small trial, a protein inhibitor was found to drastically reduce the mortality rate of patients affected with progeria. Is a cure on the way?
  4. My colleagues from other scientific fields often assume that software is just a form of applied mathematics. Surely we can automatically reason over software implementations and prove them to be correct. This turns out to be remarkably difficult. It is hard to prove as correct even silly small functions, to say nothing of actual software. The fact is, building software is very much an empirical endeavour… which is why people like me always spend at least as much time testing the software as writing it. Moreover, I always assume that my software is, at best, empirically correct. There are always cases where it will fail, even if the hardware runs perfectly (which it does not).Hillel Wayne checks whether a particular programming style that is fashionable right now (functional programming) makes software correctness easier to verify. He finds the evidence weak. To me, the real take-away should be that it is ridiculously hard to prove that code is real-world correct, irrespective of the programming style.

What is also interesting is the reaction of the functional-programming community to his analysis…

A common critique of functional-programming communities is they have very aggressive people and this held true here.

I have been on the receiving end of the functional-programming communities. A mere joke at functional programming expense is enough to get hate mail. That is, software is nothing like objective truth and pure ideas. It is filled with deeply held ideologies.

  1. In our society is widely accepted that men are freer than women sexually, even though we have contraception that should act as an equalizer. That is, a woman that openly likes sex is a “slut” or a “prostitute” whereas a men that likes sex merely has a healthy libido. People who care about gender fairness should be concerned with this obvious bias. But what is the cause? Baumeister and Twenge find that the view that men suppress female sexuality is flatly contradicted and the the evidence favors the view that women have worked to stifle each other’s sexuality because sex is a limited resource that women use to negotiate with men, and scarcity gives women an advantage.