Re: Curious coincidence

[mskala@xxxxxxxxxxxxxxxxx]

On Mon, 12 Jun 2000, Jim Reeds wrote:
> Stolfi, I am struck by the almost exact 50-50% split into 0g and 1g
> words. Would it be possible to divide the text into a few large
> portions ("A" vs "B", say, or "Bio" vs "Non bio" or "front" vs
> "back) and see if the pheonomenom holds in the pieces, as well?

I haven't been very systematic about it yet, but I did write a script to
go through the manuscript line-by-line and, after each line, print the
difference between the number of 1g and 0g words encountered up to that
point.  I displayed it as a character-based plot so I could see general
trends.  That experiment suggested that the phenomenon does *not* hold in
smaller sections - there were short gallows-rich chunks at the start and
end, and a longer gallows-poor chunk in the middle.

I'd like to, but haven't yet, split the VMS into the standard "sections"
and run counts within each.

Matthew Skala, "the modern CEO's worst nightmare" (Macleans, 2000-04-10)
    My Internet doesn't include channels, commercials, or the V-chip.
       http://www.islandnet.com/~mskala/ mskala@xxxxxxxxxxxxxxxxx