Humanist Discussion Group

Humanist Archives: Jan. 10, 2024, 7:17 a.m. Humanist 37.381 - flip/flop into 2024

				
              Humanist Discussion Group, Vol. 37, No. 381.
        Department of Digital Humanities, University of Cologne
                      Hosted by DH-Cologne
                       www.dhhumanist.org
                Submit to: humanist@dhhumanist.org




        Date: 2024-01-09 06:17:56+00:00
        From: Michael Falk <michaelgfalk@gmail.com>
        Subject: Re: [Humanist] 37.365: flip/flop into 2024

On the issue of ‘when binary matters’…

I think the relevant distinction is not so much binary vs. non-binary – it’s
discrete vs. continuous. I don’t think there is much difference between a system
that forces things into 2, 5, 10 or 1000 categories, if those categories are
distinct. I recall watching a video on Computerphile once where David Brailsford
said that the only reasons computers are binary is that it is easier to
stabilise the voltage at two levels (high/low) than at 10.

A binary scheme can of course sort things into an arbitrary number of distinct
groups:

00 – group 0
01 – group 1
10 – group 2
11 – group 3

Anyone who codes will know that one of the places where the binary/discrete
aspect of the computer matters is when you’re simulating a continuous value. If
you have Python installed on your system, try typing the following into the
Python REPL:


     *   + 0.1 + 0.1 == 0.3

You may (or may not!) be surprised to see that the answer is ‘False’!

Michael Falk


_______________________________________________
Unsubscribe at: http://dhhumanist.org/Restricted
List posts to: humanist@dhhumanist.org
List info and archives at at: http://dhhumanist.org
Listmember interface at: http://dhhumanist.org/Restricted/
Subscribe at: http://dhhumanist.org/membership_form.php