Re: Johnny Mnemonic
From: G. Orme (newsgroups_at_harmakhiss.org)
Date: 10/31/03
- Next message: John Hairell: "Re: NSA chooses ECC"
- Previous message: Colin Andrew Percival: "Re: Factoring Integers "near" a power"
- In reply to: Rob Warnock: "Re: Johnny Mnemonic"
- Messages sorted by: [ date ] [ thread ] [ subject ] [ author ] [ attachment ]
Date: Sat, 1 Nov 2003 02:25:33 +1000
Rob Warnock wrote:
> G. Orme <newsgroups@harmakhiss.org> wrote:
> +---------------
>> I just read the story and screenplay, that was made into the well
>> known movie with Keanu Reeves. In the story Johnny takes 3 snapshots
>> of a TV signal each from 3 different channels and uses that as a key
>> to encode the data he carries, 320 gig of it.
>> The same 3 snapshots are sent to the receiver separately for
>> decryption later. Is this a sensible idea, I assume it is something
>> the author William Gibson thought up.
> +---------------
>
> Maybe, but he's hardly the only one to use the idea. Vernor Vinge's
> excellent SF novel "A Fire Upon The Deep" contains the same concept
> as an oft-mentioned plot element, namely, that a simple way to improve
> the security of physically distributing OTP material is to generate
> *lots* more pad material than you really need, break it up, and send
> it via several *different* transportation paths [different ships,
> routes, couriers, etc.]. Alice & Bob then communicate by XOR'ing
> sections of pads sent by different routes, and using *that* as the
> actual pad for their OTP communications. Even if Eve intercepts &
> copies one or more of the different versions, as long as she doesn't
> get *all* of them the communication between Alice & Bob will still be
> secure.
>
> Vinge suggests that "3" is a good number of different couriers/routes
> to use. That seems like a minimally-adequate number. More would be
> better, of course, but the costs go up proportionately, and some
> pairs of sources and destinations may not *have* a plethora of
> alternate routes available between them for the couriers/cargo to
> take.
>
> [Another poster replied that Gibson had his character take all three
> TV signals *himself*, which is of course a flagrant violation of the
> "independent couriers & routes" principle Vinge espouses. I'll go with
> Vinge's method, myself...]
>
>
> -Rob
>
He did take the images himself but this was by clicking on a remote.
Since this was capturing a single frame, or perhaps parts of two frames on
the screen he couldn't have been able to see that data well enough to
remember it.
Also though someone wanting to break the code would simply look through
all the frames broadcast on all channels around that time and try them. That
depends though on whether the remote would have captured a discrete frame or
part of two frames (in the period of transition from one image frame to
another).
- Next message: John Hairell: "Re: NSA chooses ECC"
- Previous message: Colin Andrew Percival: "Re: Factoring Integers "near" a power"
- In reply to: Rob Warnock: "Re: Johnny Mnemonic"
- Messages sorted by: [ date ] [ thread ] [ subject ] [ author ] [ attachment ]
Relevant Pages
|
|
