[Cryptography] Licensing of cryptographic services in France

[Sebastian Stache]

On 2024-08-29 01:57, Phillip Hallam-Baker wrote:
> 
> 
> On Wed, Aug 28, 2024 at 3:23 PM efc--- via cryptography 
> <cryptography at metzdowd.com <mailto:cryptography at metzdowd.com>> wrote:
> 
> 
> 
>     On Tue, 27 Aug 2024, Phillip Hallam-Baker wrote:
> 
>      >       That will push solutions based on steganography, hiding
>     encrypted
>      >       information in non-encrypted streams. There will of course
>     be a black
>      >       market (the market always wins in the end, despite what
>     todays socialist
>      >       politicians want us to believe) for encrypted communication
>     through white
>      >       listed corporations. I read an article a long time ago,
>     that this is how
>      >       it works in china, were companies do sell internet
>     connections to the west
>      >       if you know who to ask through the right channels.
>      >
>      >
>      > Yup. Already designed in.
> 
>     Very interesting and thank you very much for the information Phillip. In
>     case of steganography based solution, what would you think the
>     performance impact would be?
> 
>     For regular text-only, chatting I would imagine that it would be close
>     to nothing.
> 
>     But when it comes to transfering files in the 10-100 MB range would
>     steganography make that too painful, and what about audio/video chats?
> 
> 
> Steganography just means a slight reduction in payload capacity. instead 
> of passing 1200 bytes of payload at a time, it would be maybe 1140. So 
> worst case 5%-10% reduction in bandwidth.
> 
> If you are going to do any form of traffic analysis defeat with onion 
> routing or the like, you are obviously going to increase latency 
> according to your path lengths. But no reason it shouldn't be acceptably 
> fast if you have a modern network and modern hardware.
> 
> 
> 
> _______________________________________________
> The cryptography mailing list
> cryptography at metzdowd.com
> https://www.metzdowd.com/mailman/listinfo/cryptography

That wouldn't be steganography, at least not in my terminology. For 
instance, assuming the real message is encrypted and that we're using 
sound as decoy medium, 95% payload would sound like white noise. The 5% 
Beyonce vocals wouldn't be discernible, rendereing the whole 
"steganography" excersise pointless, or worse, since you're wasting the 
5%. And if we don't encrypt, well, this last sentence is in fact mixed 
with 5% poetry, believe it or not.

Z