Links: Probability and DNA Testing, 16th Century Executioners, Walmart as a Bank, Etc.

  • Jordan Ellenberg has an excellent article in Slate about probability and DNA testing for crimes. This is important to think about, especially as we consider the implications of the recent Supreme Court decision on such testing. (In particular, the larger the database, the bigger the problem.)

  • What was it like to be an executioner in the 16th century? An interesting read.

  • You should definitely check out the New York Times Magazine‘s Innovations Issue, which features short blurbs about the origins of dozens of inventions. It’s a nice thing to dip into for some interesting histories. (If you can get a hand on the paper copy of the issue, that might be easier to browse through. For the online version, click on the “navigation” button in the upper-left-hand corner.) It’s interesting to see several cases where the famous brand name that we use a generic was not the first; for example, liquid paper beat out Wite-Out by several years.1

  • For the academics out there, consider following David Gries’ approach to the annual report (via Chris Potts)

  • Is this for real? Elon Musk proposes a “hyperloop”, an insanely fast means of transportation.

  • Could Walmart Beat Payday Lenders? Matt Yglesias makes a good point. There are a lot of situations where the credit market is inefficient, and people either pay exorbitant interest or don’t get loans they need.

    One example I’ve thought about is lending for Ph.D. students. At most research universities, Ph.D. students should be very low credit risks, because they have reliable salaries.2 And yet there are at least two situations where students without their own savings need short-term credit: costs for moving and the first month of school (because the first paycheck arrives at the end of the month), and when they’re expected to pay for conference travel out of pocket, to be reimbursed only months later. This seems like an obviously place for universities to step in and provide very inexpensive credit.3 Given how low interest rates are—certainly for financially strong universities—it would be relatively inexpensive for universities to pay students at the beginning of the month, or to offer interest-free loans to incoming graduate students. Similarly, why not offer interest-free loans for reimbursements—or, better yet, for the NSF to allow for conference travel costs to be reimbursed on purchase?4

  • The Joke’s on All of Us, by Andrew Hudgins, in the Times.

  • Perhaps the first ebook: an edition of Paradise Lost made in 1964-1965 on IBM punchcards, which ended up getting into the Project Gutenberg library.

  • A good article in the Times about gay Supreme Court clerks. It’s often said that one of the most important determinants in reducing homophobia is when people know someone close who is gay. (We’ve certainly seen the positive effect of this in many conservative politicians with gay children.) Particularly interesting is the focus on Lewis Powell, who had a disproportionate number of gay clerks and yet upheld an anti-sodomy law in the 1980s.

  1. From the article on the history of bullets, we hear the following about the musket ball:

    They were accurate up to about 40 yards. One British colonel said of musket fire, “No man was ever killed at 200 yards…by the person who aimed at him.”

    Interesting probability question: if you look at the probability distribution of a shot from a musket, is it disproportionately less likely to go dead on than stray off? Let me state this question mathematically, so that it’s clear what I’m asking. Assume, for simplicity, that we’re shooting at a very large standard circular dartboard, with bullseye in the middle. Is the probability density function highest in the center of the dartboard? If we assume that the density function is radially symmetric, our first guess would probably be that the density is decreasing in the radius, and so, yes, the density would be highest in the middle. But I could easily imagine that the poor mechanics of the gun make it so that the density function is actually highest at a positive radius away. (The intuition in all this is a bit confused by whether or not we’re thinking about a two-dimensional model or a one-dimensional model. In a one-dimensional model, annular discs don’t grow in area, while in a two-dimensional model they do. It seems plausible—especially based on the colonel’s quote—that the one-dimensional model might be somewhat apt.) If that were the case, then the strategy of shooting dead-on for the target might not be ideal. (This would apply similarly for darts; what would matter, of course, is the payoff. If the bullseye is far more valuable than the outer rings, then it could make sense. This is analogous to whether one is shooting at a lone opponent, say in a bullseye, where only the bullseye is valuable, versus shooting at a line of soldiers in a battle, where anything on the dartboard is valuable.) 

  2. Obviously, there’s variation among Ph.D. students of how well their stipends cover their needs, depending on regional cost-of-living, family status, economic status of the university and the department, etc. 
  3. I did a back-of-the-envelope calculation to consider whether it would make sense for private businesses to offer these services; the margins didn’t look great because of the relatively small scale. Universities would have the structures in place to provide this, not to mention much cheaper mechanisms of enforcement (e.g., garnishing future wages) than the legal system private lenders would have to resort to. 
  4. Have any other graduate students wondered what happens if you buy a plane ticket to a conference, but then can’t go? Do you still get reimbursed?