VMs: RE: VMS numbering systems hypotheses...

[Nick Pelling <incoming@xxxxxxxxxxxxxxxxx>]

Hi everyone,

>         I was also going to broach the fact that a vast majority of 
> labels begin
> with 'o' - and if 'o' was a number I would strongly suggest o==1.

Because I suspect that the *encoding* was right-to-left, to my eyes these 
instances of "-o" and "-9" fall at the *end* of labels.

In this context, two of my non-numeric predictions are that...

        "-9" codes for...
                "-us",  or
                "-um"   (as in Tironian shorthand):

        "-o" codes for...
                "-i"/"-e"       or
                " ",    or
                "(null)",       or for
                "-(as you'd expect)".

This last would mean (in Shannon entropy terms) "the remainder of this word 
is as you'd predict it to be", ie "use the top-ranking candidate word in 
this context", like a word-level "..."

Fo... ex..., th... is th... ki... of thi... I'm talk... abo...

My guess is also that "9-" codes for "cum-" or "con-" at the 
(right-to-left) start of words - but that's another story entirely. :-)

> >If we are sure that the all vms words are numbers, shouldn't we forget
> >about cracking it, until the code book is found? :-/
>
> Not if it's a spelling code that uses various charts/columns/rows or the
> like?
> Something like - columnNR + rowNr from a basic default set for some words,
> and chartNR+columnNr+rowNR for others?

Given that I believe the VMS' underlying language is probably a mix of 
Latin and Italian, I suspect that verbs/nouns in the code book would need 
some form of conjugating/declining - and this is what I suspect much of the 
rest of the text is (though possibly in a obfuscated way).

In addition: common words (like "cum", "et", etc) could well have short 
codes of their own - "89" [EVA "dy"] is a strong candidate, for example.

Cheers, .....Nick Pelling.....