09 February 2009

Murphy's Sausages

Still in the land of "technical difficulties": The dryer must have heard about all of the attention other things in the house were getting, because it decided to die.

  • Here's a vastly amusing set of Hermione Granger music mashups (including a really disturbing Hermione/Snape implied-'ship bit, just in time for Valentine's Day)... posted by an IP law professor. I've waited until after the first caffeine should have hit my Monday readers before posting, for precisely this reason.
  • Snurched from Professor Froomkin: One of Ed Felton's cabal tries desperately to explain to techies about being acquitted versus being searched, and simultaneously manages to imply — but not specifically discuss — two of the differences between, on one tentacle, lawyers, and on other tentacles, politicians, political scientists, businesspeople, and techies:
    1. There is a difference between "should be the law" and "is the law," particularly when it concerns any aspect of change or conflict. This is most frequently the problem for political scientists and techies, who have this inescapable urge to treat "my vision of what the law should be" as reality, without all that much reference to any other point of view. It's sort of ironic to accuse the community of political scientists — those arguing over the Second Amendment being a recent obvious example — of quasianarchism, but after all they are academics...
    2. Process and intermediate steps matter, particularly in the larger context of "what does this mean to an individual" instead of "what does this mean to me as an advocate of policy/technology/business plan x." There's a game-theory construct concerned with this sort of situation, sometimes called "maximum adverse excursion" (and often by other analogous names) that points out that a favorable end payoff may not be the best result for a player if the maximum adverse excursion along the path leading to that end payoff exceeds that player's risk tolerance — that is, the long-run payoff requires surviving short-run losses.
  • Jay Lake talks about backing up your writings. Jay makes some good points about getting the raw data out of a single location, but neglects perhaps the most annoying aspect: Data format. The one real advantage of printouts is that they are not coding-dependent... as is the .doc file format that Jay prefers. When I'm writing a long piece, I save everything as RTF, not in a proprietary format. Even in the worst case — that Microsoft were to suddenly go out of business (remember WordStar?) — just about any program can already read and write RTF, and one can write a quick macro or filter to convert an RTF file to plain text (at worst) and probably formatted text.

    About a decade ago, we had an excellent example of what happens with data-format lockin: Remember the Y2K phenomenon, which was primarily caused by COBOL's default of two-digit years back in the 1950s and 1960s? That's not the last time it's going to happen; witness the recent dropping of DRM authentication by some music services.

    There's an additional bonus to using RTF: It won't store macros, so the chance of accidental infection is substantially lower.

  • I am a combat-trained biochemist. It's February 2009. And, therefore, I celebrate a Darwinian bicentennary with some glee. It's sad that so many in England still reject evolution, despite virtually incontrovertible evidence that the broad outlines of the theory that evolved from Darwin's On the Origin of Species is a correct explanation that can be observed in action daily. It's even sadder that various surveys have demonstrated that acceptance of the theory of evolution is even lower on this side of the pond... but that is probably more an indictment of poor teaching and textbooks than anything else, because there's a big distinction in acceptance once one asks about the content of the theory instead of the labels for it.