|
|
 | | From: | Apokrif | | Subject: | Exotic scripts as paper and pencil ciphers (was: A general question on learning languages) | | Date: | 19 Jan 2005 17:19:01 +0100 |
|
|
 | [n-th Supersedes]
Geoff :
> > I even write secret messages in English with the Arabic script.
> I used to use hangul/hankul for secret messages, but now that I'm
> back in the US there are a lot of Korean-Americans to look over my
> shoulder, so now I use the Tibetan or Manchu scripts. Both of them
> are pretty straightforward for speling English.
A French ambassador is reported to use "birgien", a mix of Burmese and
Georgian characters:
http://ocsena.ouvaton.org/journal/article.php3?idarticle=174
Rather than using a well-known alphabet or syllabary, a good method
(not really secure against a skilled attacker) would be to design your
own writing system (which could include abbreviations and ideograms,
depending on your specific needs), resembling already existing scripts
so as to delure the attacker.
[fu2 sci.crypt]
--
`cat ~/.signature`
|
|
| | From: | Jacques Guy | | Subject: | Re: Exotic scripts as paper and pencil ciphers (was: A general question on learning languages) | | Date: | Thu, 20 Jan 2005 05:34:09 +1000 |
|
|
| (I put sci.lang back into the newsgroup line
Apokrif wrote:
> A French ambassador is reported to use "birgien", a mix of Burmese and
> Georgian characters:
> http://ocsena.ouvaton.org/journal/article.php3?idarticle=174
Alas, it does not say in what language, or languages he writes
his notes. Pity. If he'd been born (much) earlier, he would
have been a prime suspect for the authorship of the Voynich
Manuscript (no smilie, I'm being serious).
> Rather than using a well-known alphabet or syllabary, a good method
> (not really secure against a skilled attacker) would be to design your
> own writing system (which could include abbreviations and ideograms,
> depending on your specific needs), resembling already existing scripts
> so as to delure the attacker.
Using one of those ciphers often called "Papal ciphers" would
make it a bit safer. You write the alphabet (reduced to 25
letters) in a 5x5 square. You assign each row a number from
0 to 4, each column a number from 5 to 9. Each letter is now
represented by a two-digit code: a = 05, b = 06, and so on.
And "cab" becomes... (think, think, think) 070506
The trick is: omit the "row digit" when it is the same
as the previous "row digit". So: "cab" -> 070506 -> 0756
but "cat" -> 070639 -> 07639
With minimal training you can write in that as fluently
as in any simple-substitution cipher. You can also make
the "delete" rule optional, which amounts to inserting
nulls and makes it just a little bit more difficult to
crack.
And yes, I do think that that might well be how the Voynich
Manuscript was written.
And no, I do NOT think that that is how it was written, because
such a cipher is easy to crack, and so the Voynich would
have been cracked long ago. But it still isn't. QED.
|
|
| | From: | Paul J Kriha | | Subject: | Re: Exotic scripts as paper and pencil ciphers (was: A general question on learning languages) | | Date: | Thu, 20 Jan 2005 20:47:55 +1300 |
|
|
|
Jacques Guy wrote in message news:41EEB631.4096@alphalink.com.au...
> (I put sci.lang back into the newsgroup line
>
> Apokrif wrote:
>
> > A French ambassador is reported to use "birgien", a mix of Burmese and
> > Georgian characters:
> > http://ocsena.ouvaton.org/journal/article.php3?idarticle=174
>
> Alas, it does not say in what language, or languages he writes
> his notes. Pity. If he'd been born (much) earlier, he would
> have been a prime suspect for the authorship of the Voynich
> Manuscript (no smilie, I'm being serious).
>
> > Rather than using a well-known alphabet or syllabary, a good method
> > (not really secure against a skilled attacker) would be to design your
> > own writing system (which could include abbreviations and ideograms,
> > depending on your specific needs), resembling already existing scripts
> > so as to delure the attacker.
>
> Using one of those ciphers often called "Papal ciphers" would
> make it a bit safer. You write the alphabet (reduced to 25
> letters) in a 5x5 square. You assign each row a number from
> 0 to 4, each column a number from 5 to 9. Each letter is now
> represented by a two-digit code: a = 05, b = 06, and so on.
>
> And "cab" becomes... (think, think, think) 070506
>
> The trick is: omit the "row digit" when it is the same
> as the previous "row digit". So: "cab" -> 070506 -> 0756
> but "cat" -> 070639 -> 07639
>
> With minimal training you can write in that as fluently
> as in any simple-substitution cipher. You can also make
> the "delete" rule optional, which amounts to inserting
> nulls and makes it just a little bit more difficult to
> crack.
>
> And yes, I do think that that might well be how the Voynich
> Manuscript was written.
>
> And no, I do NOT think that that is how it was written, because
> such a cipher is easy to crack, and so the Voynich would
> have been cracked long ago. But it still isn't. QED.
Assuming, of course, that there is anything to be cracked. :-)
There is some real chance of Voynich text having been
generated only to _look_ like an encrypted secret message
while in fact there is no meaningful message contained in it
at all. But how could one conclusively prove that? :-)
Paul JK
|
|
|
