[clug] A most interesting read, most interesting

[Sam Couter]

Michael Cohen <michael.cohen at netspeed.com.au> wrote:
> Sam, this sort of thing has been talked about for yonks. Its not going to work
> any better than the current system, because somewhere, the data is present in
> an unencrypted form in order to be presented to a human neural sensor of some
> kind.

And this is one reason why DRM is fatally flawed, but they will still
try it. And it will still be a pain in the arse and illegal for us to
get unrestricted access to the media we've paid for.

> Encrypting the signal between the DVD rom and the monitor will simply
> move the clear text data to the monitor (i.e. you might need to patch signal
> lines inside the monitor). 

The signal lines will be cast in blobs of epoxy, or the electronics will
self-destruct if the case is opened, etc.

> Apart from this, This proposition is obviously ludicrous because you will be
> unable to watch any video on monitors which do not support such encryption.

Isn't there a big fuss going on at the moment about HDMI or something
for exactly that reason? People bought these high-def TV sets, but now
it looks like high-def equipment is going to refuse to send them a
high-def signal. Sorry, fuzzy on the details through lack of care.

> These were then so popular that today almost every dvd
> player sold is multi-region. In fact its a fact which most players advertise as
> a feature!!!

I think the DVD consortium may have relaxed this requirement because
some jurisdictions (including Australia) required that a DVD player be
multi-region capable.

> Sam, the fact that an OS is closed source only makes it a little more difficult
> to figure out where to tap the data. Recent advances in reverse engineering
> show that not having the source code is not that big a hurdle - see for example
> the book Rootkits by Greg Hoglund, James Butler which describes a technique
> called Direct Kernel Object Manipulation, where they are able to manipulate
> internal NT kernel data structures to an absolutely amazing degree using only
> reverse engineered knowledge about the internals of the windows kernel.

Vista (maybe only 64-bit versions) doesn't allow this. It detects kernel
object manipulation and shuts down. Only signed drivers are allowed, and
as a user you don't get a say in it.

> Rather than having to reverse engineer every DRM application's way of doing
> things, just reverse the windows API implementation, and in one swoop you can
> circumvent _all_ DRM applications using this API. Whats more is that the MS api
> will be well documented so you will have a great start.

Yes, just like having the GnuPG source code available makes it easy for
me to reverse engineer and decrypt any encrypted messages I
want.</sarcasm>

The important part of DRM isn't the software, it's the keys.

> How does this statement follow from the previous two? I dont quite understand
> what you are saying here. How does the truth of the previous two statements
> imply that DRM on linux can not work? It can work just the same as on windows
> (which we all agree is not very well), because on linux you are still able to
> release software without source code and not under the GPL. Just because the
> rest of the system is GPL doesnt mean every piece of software has to be.

Easier driver access, for one. Also, no secure media path, which is
coming with Windows.
-- 
Sam Couter         |  mailto:sam at couter.id.au
                   |  jabber:sam at teknohaus.dyndns.org
OpenPGP fingerprint:  A46B 9BB5 3148 7BEA 1F05  5BD5 8530 03AE DE89 C75C
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 189 bytes
Desc: Digital signature
Url : http://lists.samba.org/archive/linux/attachments/20061228/4738cf28/attachment-0001.bin