Skip to content
 

XXX

You gotta read this, including all the comments. It’s fascinating. (Link from Jkrideau.)

11 Comments

  1. Anoneuoid says:

    I always thought it interesting that early civs used different numerical systems for different purposes. For example, originally, if you were dealing with slaves/cattle you would use decimal, but talking about the kings family you would use sexagesimal:

    This decimal system, employing signs borrowed from the bisexagesimal system, qualifies what apparently are domestic animals, but also what we believe are lower-status humans. It appears that high-status humans-foremen and high officials-were, as all humans in Babylonia, qualified sexagesimally.

    http://cdli.ucla.edu/staff/englund/publications/englund2004a.pdf

  2. Jonathan (another one) says:

    The fact that what is represented in a computer as 01000001110010000000000000000000 is what we would call 25 (see https://en.wikipedia.org/wiki/Single-precision_floating-point_format ) shows the gigantic disconnect between ease of computation and ease of reading. Stan is an example of the giant disconnect between ease of ideating a model and the gritty hidden details of solving it.

    • Jonathan (another one) says:

      And the fact that 25.1 is 01000001110010001100110011001101 makes the disconnect even clearer

    • Dalton Hance says:

      My brother taught himself to count on his fingers in binary. It’s actually a handy trick because you can get a lot further than with one digit per increment. There can be a bit of confusion when he asks for four of something though…

  3. Alex Gamma says:

    It’s like seeing a movie being rated VIII, then you watch it and it’s a V ;) But I found this bit interesting:

    “There was no context to the tweet, a reply or whatever, so I can only assume that… “
    “…the implied sneer in Professor Evans’ tweet…”

    Unprecedentedly, in this day and age we have unlimited digital space to add all the context we want. Yet, we build a system of online communication with self-imposed restrictions that force us to do semantic archaeology to try to understand each other’s utterances. I wouldn’t be so bad if at least scientists didn’t use it to discuss matters of science. But by doing so, they’re deliberately creating a breeding ground for misunderstandings, oversimplification, false accusations, hysteria, moral outrage, etc.

    I find the use of twitter by scientists to communicate about science truly paradoxical, given today’s emphasis on transparency and the sharing of all relevant data, also the increasing emphasis on communicating variability and uncertainty in one’s data and analysis.

    Question: is tweeting about a study much better than giving a p-value as a summary for a whole study?

  4. Alex Gamma says:

    *It* wouldn’t be so bad…

Leave a Reply