Four short links: 12 November 2010

Experimental Design, Distributed Microsatire, Annotations, and Machine Learning

  1. Evaluating Extraordinary Claims — science isn’t easy, and the most difficult part of science is experimental design. Peter Norvig takes us through a handful of studies that claim to have found prayers to be effective, and critiques their experimental design. Lest you think scientists only criticize prayer experiments, read the Atlantic’s profile of John Ioannidis.
  2. Thimbl — decentralized microblogging using SSH and finger with an hilarious manifesto. Performance art at its finest. (via Jason Ryan)
  3. Annotator — open source toolkit for annotating
    text, written for The Open Shakespeare project. (via Rufus Pollock‘s blog about it)
  4. Machine Learning: A Love Story — presentation on the history of machine learning, by Hilary Mason of bit.ly. (via Julie Steele)
tags: , , , , ,
  • http://http:/twitter.com/lnorvig Laura Norvig

    Ha, I just noticed in Peter’s evaluation he points to sloppiness and typos as possible indications of flawed studies. However, he misspells it “sloppyness”.

  • buzzstatik

    Anyone who wants to be informed about current work in scientific research on prayer needs to read “Nonlocality, Intention and Observer Effects in Healing Studies: Laying a Foundation for the Future” by Stephan A. Schwartz and Larry Dossey, MD, in the Sept/Oct 2010 issue of Explore Journal (explorejournal.com) — considerably newer than ther 2007 book Norvig cites.

    And if he thinks that nonlocal functioning “would be the kind of thing that would make the news, and I might have heard about it,” he doesn’t seem to know much about either modern parapsychological research or journalism.