What's a "real encryptor"?

Typo correction from first post:

If PGP outputted such a hard-to-distinguish-from-random data format, it
opens up many different possibilities for sending your messages. Ideally, no
one would be able to tell if it was an encrypted message except by
successfully decrypting it. As it is now, such schemes have to rely on
"encrypting" an already encrypted PGP message to hide the fact that it IS a
PGP message! Many of us just want to be left alone and are tired of having
our files tagged as BEING encrypted. Personally, I suggest using PGP as a
Clipboard utility so I can cut a message out, encrypt it, paste it back in
and save it as a word processor file which I then Macintosh BinHex encode
as text, and e-mail off. Now I'm just sending a BinHexed word processor
file, just like thousands of other Macintosh e-mailers out there every day!
This isn't good enough since it is so easy to reverse, AND can be automated.
Honestly, I'm not doing this much yet with distant friends, but then there
are only two of them ;-), and they are still struggling with just e-mail.
PGP is still a program only used by those why really need it. It may remain
that way, so for those people, having a random data block output would mean 
they wont set off alarms and catch the attention of the government, just for
sending a love letter to their mistress ;-). It would also render the Clipper
issue moot.

-=Xenon, who never could type, and breaks things a lot still=-

Xenon, you could make your point a lot clearer if you wouldn't detweil
so much in your posts.  But let me propose a technique that I think
would be what you meant to define as a "real encryptor".

Take a file and encrypt it by taking the first block of data and using
it as a key to encrypt the rest of the file.  Than take the beginning of
the file and encrypt it with RSA.  Therefore, since you need to know the
first block of plaintext to decode the rest of the file, you could only
decode the file if you first decoded the RSA block.  Perhaps some random
padding could also be added, and a random session key inside the RSA. 
The file would have no identifying markers to show what key it was
encrypted with, or what key was needed to decrypt it.  Is this what you
wanted?  I think I could hack that...

And a little flame: Before you put down our software, try writing some
of your own.

Matthew J Ghio <mg5n+@andrew.cmu.edu>  writes:
>
> And a little flame: Before you put down our software, try writing some
> of your own.

  This is uncalled for.  Xenon has consistently admitted that he isn't a
  programmer and that he isn't *able* to write cryptography software.  
  In no way does that make his criticism any less valid.  Rather than wasting
  time flaming him (I know, you just flamed a little), perhaps *you* should
  write the code Xenon wants... or perhaps *I* should...

  That's like telling someone who can't read that they shouldn't criticize
  the technique that you are using to teach them to read until they have
  tried to develop their own technique to teach reading...

  Remember, no matter *how* the criticism is couched, the *real* criticism is
  that PGP doesn't have random block output mode...  All of the other verbage
  in which it was couched isn't relevant or important...  looking past the
  manner in which the complaint is presented and attempting to fix the problem
  which was the cause of the complaint is the key to good product maintainence.

  Let's help PGP become a product...

 Jon Boone | PSC Networking | boone@psc.edu | (412) 268-6959 | PGP Key # B75699
 PGP Public Key fingerprint =  23 59 EC 91 47 A6 E3 92  9E A8 96 6A D9 27 C9 6C

please take me off this list
Richard Savel
rsavel@welchlink.welch.jhu.edu