[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: VMs: Italian phrases



Nick Pelling incoming@xxxxxxxxxxxxxxxxx

> Hi Jeff,
>
> At 20:59 12/11/2003 +0000, Jeff wrote:
> >I have got so far:
> >
> >so il leda
> >
> >accosti deio nelle
> >
> >elle mio dicchi
> >
> >soille michi
> >
> >Any ideas on whether they mean anything?
>
> If you statistically examine at a glyph transcription of the VMs (let
alone
> an EVA transcription, which is even worse), you'll find that different
> letters are strongly predictable from their context.
>
> Basic computer science (or what one might call "foundational computer
> science") says that these  letters therefore contain (or express) very
> little intrinsic information. If you're suggesting a method by which you
> specify a different table for each possible preceding letter (as I believe
> you are), what you're effectively doing is encoding all that missing
> information in those tables.
>
> The problem then isn't making 11 words in a row look plausibly like a
given
> language - depending on the tables and methodology you're using, that (as
> you have found) is quite manageable. Instead, the problem is making *220
> (or so) pages* work in the same kind of way (which I'd guess is roughly
> 2500x more difficult).
>
> I'd also say: double letters are problematic - in ciphers from the same
> time frame, doubled-letters were thought to be the second easiest way
> (after vowel-frequency analysis) for code-breakers to "get in", so these
> were usually just converted to single letters (or extra symbols were added
> to code for particular doubled letters). The presence of doubled letters
in
> your plaintext alone is sufficient to make me strongly doubt its
> correctness, never mind any other aspects.
>
> All the same, the real test isn't what I happen to think, but what the
rest
> of the deciphered text looks like - I'm sure this would give you the real
> result, rather than just 11 words :-o
>
> Cheers, .....Nick Pelling.....
>

The words persist through out the length of the sample. BTW all the words in
the line above ran concurrently as separate phrases and not just individual
words.

Don't make assumptions Nick. This stuff is making sense. Look at the
babelfish translation I posted.

Jeff


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