While Steve Jobs is a villain to some and a hero to many, his 2005 Commencement Address at Stanford University (see below) is often held up as a classic of the carpe diem genre. It is a wonderful feel-good speech but something always sticks in my throat about it: along with many other commencement addresses, I fear it may be terribly biased.
My concern is that if you only ask very successful people for advice, they will often make recommendations that try to make sense of their own, often unusual, experiences. In Jobs’ case he dropped out of college, had his company “stolen” from him and nearly died. Whatever the recommendations, it may be that 99% of people would regret following them by the time they get to 50. Of course, the failures will never be asked to give a commencement address at Stanford so the down sides of such a strategy are never heard.
What Jobs may have seen as a kind of divine hand (“gut, destiny, life, karma”) guiding him towards greatness, may simply have been luck. He was great because if he wasn’t he wouldn’t have been asked to talk to Stanford graduates.
This all reminds me a little of a scam where a con artist tells you he has a fool-proof method for guessing the winner of a horserace. As proof he correctly guesses three winners in a row. Then takes your large bet for the 4th race and disappears with your money. How does the scam work? The con artist simply does a Monte Carlo experiment.
They tell many people they will correctly guess three winners in a row, then randomly splits the favourites in each race between the people they are conning. The ones who are assigned horses that don’t win are simply discarded leaving only those people who he guessed correctly for and they are progressively more and more convinced the con artist has some real skill. In this analogy, the star giving the commencement address is the one for whom the strategy payed off big time. The voices of all the people who lost are never heard.
What has this got to do with astronomy and cosmology, you may ask? Well, we are often interested in how average properties of a population change with time or space, e.g. the average brightness of exploding stars. Looking nearby it is easy to ensure that one has a representative sample of a population, but at great distances this is very hard. All but the brightest supernovae (or whatever other object you are studying) are too dim to be detected by your telescope and if you don’t account for the missing “failures”, your estimate of the mean will be very wrong.
This is Malmquist bias and accounting for these “missing voices” is something we worry a lot about in doing cosmology with supernovae, and that is what Steve Jobs has got to do with the acceleration of the Universe.