Re: VMs: Word Length Distribution

[knoxmix@xxxxxxxxxxxxx]

    ----------------------------------------------------------
    On Tue, 13 Apr 2004 at 10:29:45 +0100
    Nick Pelling wrote (in part)
    However, this is still only part of the story: one thing I've meant 
    to do (but never got around to) is to test whether verbose ciphers 
    reduce or enhance this binomial-like distribution effect. Could I 
    suggest you try this out? 
----------------------------------------------------------

Column 4 might be optimal. Additional conversions later.
Maybe more of this in time with methods and details on a webpage. 

I do not remember where I read it but here is an approximate quote by 
a critic: One is certain that profound ideas are expressed in this 
volume. One is damned if one knows what they are.


Each word was included once regardless how many times it might 
appear.
column 1 -- word length
column 2 -- original selection from Takahashi's transcription
(generally, lines with unreadable characters and many (not all) lines 
with few words were omitted)
column 3 -- cFh, cKh, cPh, cTh  are single glyphs
column 4 -- cFh, cKh, cPh, cTh, ch, sh  are single glyphs
(For the last column, the 3 v's and 2 z's were converted to x's.)

If the columns touch one another, let me know so I can better 
separate them. With a little juggling they should copy to a 
spreadsheet for graphs and statistics.


1		15			16			17
2		70			85			110
3		237		297		440
4		641		736		1138
5		1276		1388		1798
6		1701		1739		1847
7		1645		1565		1267
8		1096		993		670
9		626		554		294
10		261		231		142
11		122		111		77
12		91			84			50
13		50			48			22
14		27			23			13
15		18			12			9
16		16			13			9
17		4			4			2
18		3			2			2	
19		6			4			1
20		2			2			0
21		2			2			0

......... Knox
______________________________________________________________________
To unsubscribe, send mail to majordomo@xxxxxxxxxxx with a body saying:
unsubscribe vms-list