5.0170 Etaoin Shrdlu Redux (4/81)

Elaine Brennan & Allen Renear (EDITORS@BROWNVM.BITNET)
Fri, 21 Jun 91 16:31:21 EDT

Humanist Discussion Group, Vol. 5, No. 0170. Friday, 21 Jun 1991.


(1) Date: Thu, 20 Jun 91 15:31 BST (14 lines)
From: DAVID BARRY <UBJV649@CU.BBK.AC.UK>
Subject: RE:5.0158

(2) Date: Thu, 20 Jun 91 16:25:01 CDT (31 lines)
From: James Marchand <marchand@ux1.cso.uiuc.edu>
Subject: Re: 5.0164 Rs: Etaoin Shrdlu

(3) Date: Thu, 20 Jun 91 19:10:56 CDT (23 lines)
From: James Marchand <marchand@ux1.cso.uiuc.edu>
Subject: Re: 5.0164 Rs: Etaoin Shrdlu

(4) Date: Fri, 21 Jun 91 11:32:58 MST (13 lines)
From: Skip <DUSKNOX@IDBSU>
Subject: Re: 5.0164 Rs: Etaoin Shrdlu; Qwerty; Dvorak

(1) --------------------------------------------------------------------
Date: Thu, 20 Jun 91 15:31 BST
From: DAVID BARRY <UBJV649@CU.BBK.AC.UK>
Subject: RE:5.0158

fURTHER TO THE CORRESPOMDENCE ABOUT QWERTY

There are some keyboards that are optimised for use on computers. They
are all CHORD typers. That is you press more than one key at a time
(like playing the piano) One I have heard of is the Philips Velotyper
another is the Moulton keyboard. The last is claimed to avoid
Repetitive strain injury. You can type on these things VERY fast if expert
(ie equivilent to touch typing in QWERTY) Yours with two fingers....
David Barry
JANET ubjv649@uk.ac.bbk.cu
(2) --------------------------------------------------------------36----
Date: Thu, 20 Jun 91 16:25:01 CDT
From: James Marchand <marchand@ux1.cso.uiuc.edu>
Subject: Re: 5.0164 Rs: Etaoin Shrdlu

I think it may have been Morse who thought up the etaoin shrdlump
frequency, based on the number of letters in the box in a typesetting
shop, though he may have had the devil (pun intended) of a time doing
it. At any rate, there are several counts in existence, based on any
number of words. You can validate the etaion shrdlump count by playing
the Vanna White game on TV (or on your computer). A good place to look
for letter frequencies is in cryptography manuals. Those which come to
hand:

Laurence Dwight Smith, Cryptography (NY: Dover, 1943), reveals himself
to be an adherent of Ohaver: p. 153: etoanirshdlcwumfygpbvkxqjz, as does:

Helen Fouche Gaines, Elementary Cryptanalysis (Boston: American
Photographic Publishing Co., 1942), p. 218.

Another book of the same vintage: Fletcher Pratt, Secret and Urgent
(Garden City: Blue Ribbon Books, 1942).

At any rate, we cryptanalists were always admonished to remember Etaion
(or oin) Shrdlump, and I pronounced his first name as if it were Irish
(thick with thick and thin with thin). The various counts differ but
little. For those interested in phoneme counts, patriotic pride leads
me to mention:

Tables of Transitional Frequencies of English Phonemes, ed. Lee S.
Hultzen, Joseph H. D. Allen Jr., Murray S. Miron (Urbana: University of
Illinois Press, 1964). With a discussion of other counts.
(3) --------------------------------------------------------------32----
Date: Thu, 20 Jun 91 19:10:56 CDT
From: James Marchand <marchand@ux1.cso.uiuc.edu>
Subject: Re: 5.0164 Rs: Etaoin Shrdlu

A rare trip home and away from my computer allows me to confirm my
previous report on the origin of Etaion Shrdlu. It is indeed from
Morse, who derived it from counting the pieces of type in a print-shop.
I quote from Colin Cherry, On Human Communication, 2d ed (MIT, 1957),
36: "Morse realized that the various letters of the English language
are not used equally often; a visit to a printer's office and a count of
the quantities of type used gave him an estimate of the relative
frequencies of the letters." On p. 37 he cites Morse's original code,
which is in the order etainoshrdlu, but note that a, i, n, o, and s have
the same frequency. P. 37: "The first table of letter frequencies was
probably that of Sicco Simonetta of Milan in the year 1380; another,
used by Porta in 1658, included digrams also (letter pairs, such as ed,
st, tr)." Cherry's is a splendid book and when he says something, "You
can book it, Dano." He recommends another splendid book, G. K. Zipf,
Human Behavior and the Principle of Least Effort (Cambridge:
Addison-Wesley, 1949). Zipf's book is a must for anybody interested in
language statistics.

Jim Marchand
(4) --------------------------------------------------------------19----
Date: Fri, 21 Jun 91 11:32:58 MST
From: Skip <DUSKNOX@IDBSU>
Subject: Re: 5.0164 Rs: Etaoin Shrdlu; Qwerty; Dvorak

James O'Donnell is still wondering about the frequency table. The one
I remember, having memorized it in my childhood, runs like this:
etaonrishdlfcmugyp (I never bothered with the rest)
I got this list from a book on cryptography. It has served me well over
the years, mostly when I play hangman.

ELLIS 'SKIP' KNOX
Historian, Data Center Associate
Boise State University DUSKNOX@IDBSU.IDBSU.EDU