[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
VMs: Word formation
When removing spaces from the VMS (or dots as in EVA) and performing a
sliding scan of triplet occurances it is found that a sizeable proportion
occur only once. This means that words can be formed from the patterns quite
easily. This may mean that an actual plaintext was used as the input or that
the author had analysed the structure of language and was able to mimic it.
Other languages have a similar property of once only occurance but are not
as pronounced as the VMS.
Take a look at the sample below to see what I mean. An english phrase was
inserted into the top line. Then the links were made to coincide with the
patterns found. Echoes of the top line can be seen in the lines following.
This makes it easier to form words initially. However as shown in my
previous posts it soon can go all too wrong.
I have said before that Italian fairs better than other languages. It does.
However I feel that even if this is the solution it will be an almost
impossible task to prove it, as one mistake in the path through the tables
and it goes completely astray.
The sample. Best viewed in courier as all the letters line up.
this book is giving many lines within a old many
s w ing ma g t
n n o ng man t wit g ng m
t ng ng
n y t e
ng man y lin
n n i an m ng
n o ng n t
g y ng t o
n y lin i ng man
P.S. Notice the top line has no repeating patterns until the very end. Could
this be a property of all paragraphs in the VMS? If this is the case then I
am convinced that these top lines are error correction.
To unsubscribe, send mail to majordomo@xxxxxxxxxxx with a body saying: