I could have contact with
Mr. Sean Luke:
who once had shown the algorithm and result of digraph-statistics on some Brown corpus.
I'd used that, and till now it has not been available. I've known this is because the code is just for a class while 2009 Sean had.
Now he showed me the Lisp code and the Brown corpus text he treated.
So I can study and trace what he did by the code on the corpus. Thanks a lot! May God bless you!
- the Brown corpus
- http://minimalism.linguistics.arizona.edu/~hammond/ling696f-sp03/browncorpus.txt
What had been happened on upstrike typewriters inside typebasket before QWERTY?
Until QWERTY layout had been stable almost in 1882, and later on until early 20th, the upstrike typewriters are the main stream of those.
keywords: type bar,
up-strike,
front-strike,
type-basket, letter sequence on type-basket,
- type bar
-
- up-strike
-
- transformation from up-strike to front-strike
-
- front-strike
- http://qwerty-history-jp.g.hatena.ne.jp/raycy/20140310/1394408594
- type-basket
-
- letter sequence on type-basket
-
ABC・・・ to QWERTY
How I used the digraph-statistics
For example, "er" and "re" are accumulated in one and represented by "er"
The frequency the adjacent type-bars move sequentially while a Brown-corpus-text had been typed out.
If the letters of the type-bar are put sequentialy around the type-basket and if like ...VWXYZ23456789ABCD... then
| A | B |
C | D | E | F | G | H | I | J | K | L | M | N |
| | Z | Y | X | W | V | U | T | S | R | Q | P | O | |
https://spreadsheets.google.com/pub?key=0AqzxPHll8XJldDFDbFZqMFVzYmlHd293Y1dCYWg0QVE&hl=en&gid=0
adjacent typebars' letter pairs frequency
Plots of the frequency the adjacent typebar's sequential motion along time and innovation goes by,
http://www26.atwiki.jp/raycy/pages/216.html
among many of scattered key layouts ( among many of over sixty thousands of Monte Carlo trials )