[Matroska-devel] Re: Matroska Specs Notes

Lhomme Steve S.Lhomme at NEOPOST.FR
Fri Nov 28 09:44:04 CET 2003

(I'll try to send this from work, not sure if it will work)

> 2. This line, "A Matroska file contains one or more segments. It can
> be safely started with the old 160-octets type header without the EOF
> And it has a link to the old MCF header page.  Is there really any reason
> support this?

The type header is not needed. But a Matroska file can have many segments...
It can actually even have more than one EBML head. That's needed if you want
to be able to use the "cat" UNIX function. But of course there are some
strict conditions for this to work. It should actually be only usable when
the 2 original files were linked. Otherwise playback is "uncertain"...
Anyway I'm not sure any player support this "cat" feature yet.

> 3.  This line, "There is currently no encryption defined. It will be
> when an open DRM system will exist. It should work at Block level and/or
> level and/or Track level and/or Segment level."  Encryption was just
defined, so
> shouldn't this be removed?

Not removed, but changed to say that one possible type of encryption is
supported under one particular OS...

> Is this really used this way?  What if the first file is
> damaged and missing some data, wouldn't that screw up the offsets for the
> files?  

Used, I'm not sure but IMO that's the way to do it. But right now I think
that each part of a link set only contain the seek positions of its own
data, not the entire. Am I write Mosu ?

> Also, "multiple" is misspelled.

You don't need anyone's approval to correct misspelling.

> First, is it
> possible to list more than one stream that the track should be the overlay

I don't have the specs here, but I don't think so.

> Second, would the codec EVER not be identified?  That sounds a bit
dangerous to
> me.

Mmm, I'll have to check that. I don't remember how it's supposed to work.

More information about the Matroska-devel mailing list