[Matroska-devel] Re: Re: Re: Compromise Encryption Proposal

Mike Matsnev mike at po.cs.msu.su
Tue Jan 31 08:35:48 CET 2006


As I understand these proposals, they represent a radical departure
from current Matroska syntax. So radical that no current software can
be reused at all for reading and writing encrypted files (even at source
code level). Given this I see no point in putting it in Matroska specs
and calling it Matroska, as we'll end up having two different specs
under the same name. Why not create another ebml based container
with a different name and use that for encrypted content?

On Mon, Jan 30, 2006 at 05:13:59PM -0800, Joseph Ashwood wrote:
> ----- Original Message ----- 
> From: "Paul Bryson" <paul at msn.com>
> Subject: [Matroska-devel] Re: Re: Re: Compromise Encryption Proposal
> >The current Matroska signing element allows signing only certain parts. 
> >You could for instance, sign every cluster individually.  The length of 
> >time in a cluster can vary to the creator's settings, but typical might be 
> >5 seconds.
> It does, but I'd like to unify everything under a single heading. This will 
> simplify things and remove the potential for a problem determining which to 
> do first. This becomes particularly problematic when you consider that the 
> different orders actually mean different things cryptographically. Pushing 
> the ordering data into the codec removes this headache from the 
> specification. With my recommendation that allows the muxer to determine 
> whether to sign each part inidividually, or all parts at once, or some 
> combination of the two.
>                        Joe 
> _______________________________________________
> Matroska-devel mailing list
> Matroska-devel at lists.matroska.org
> http://lists.matroska.org/cgi-bin/mailman/listinfo/matroska-devel
> Read Matroska-Devel on GMane: 
> http://dir.gmane.org/gmane.comp.multimedia.matroska.devel

More information about the Matroska-devel mailing list