• I believe that functionally k=p=t=f
  
• I believe that ch=Sh=e=ee=eee; there is no difference between them. The encoder used different amounts of "c" shapes to mask the code, but they are functionally equivalent.
  
• Same with n/in/iin etc.
  

(19-04-2020, 03:38 PM)-JKP- Wrote: You are not allowed to view links. Register or Login to view.

(19-04-2020, 12:24 AM)ThomasCoon Wrote: You are not allowed to view links. Register or Login to view.
To be more explicit on a few points:

• I believe that functionally k=p=t=f
  
• I believe that ch=Sh=e=ee=eee; there is no difference between them. The encoder used different amounts of "c" shapes to mask the code, but they are functionally equivalent.
  
• Same with n/in/iin etc.
  

I have different ways of looking at the VMS text. In fact, every couple of months I force/challenge myself to find a new way of interpreting it (after all these years, I have zillions of versions). But... one of the ways I've looked at it is this (see the top left of f67r ):

                      cheodaiin aekeey s ar air s oar cheey dair cThey

Possibillity 1:     cccccciii cciiccc c ci cii c cci ccccc cciii iiccc  (simplest form)

Possibility 2:      cccoociiii ciciiccci c cii ciii c o cii ccccci ciciii ciiccci (one of many possible ways to interpret the outliers like d and a or d and o as separate chars)

Possibility 3:      cccoociiii ciciiccci c cii ciii c o cii ccccci ciciii c||ccci. (gallows could be stop char, signal char, or modifier char) 

(I have about a dozen other possible "simple" interpretations, for example, "a" could be split into c + i, but the two examples is enough to get the idea across. I also have versions where I have collapsed similar patterns into one, similar to what Thomas is describing. I also have more complex interpretations. I even have a color-coded version that I created to see if the VMS text could be interpreted as embroidery patterns.)


One way to look at it is as a numeric code (several different kinds of numeric codes, in fact). There are many possibilities... binary, Roman numerals, Indic-Arabic numerals interpreted into letters, and on and on...


This does not necessarily interpret into an alphabet (it could be other things... references, symbols, etc.) but... if it does, it might have to be configured in a way that creates about 20 variations. This could easily be done with the simplest form above (I'm not going to tell you how, I'm curious as to who can see a straightforward way to do it, and possibly one that is different from the one I use).