[Discuss-gnuradio] CPTWG / BPDG / "Compliance & Robustness Rules"

[Eric Blossom]

On Thu, Feb 28, 2002 at 12:25:17AM -0800, Seth David Schoen wrote:
> I've gotten some other documents since then, and I'm hoping to produce
> good HTML versions of them.  In particular, we have the BPDG's "Draft
> Compliance and Robustness Rules" which explain what BPDG expects will
> be required of devices which can do ATSC demodulation.  (They don't,
> thus far, address the issue of how those requirements would be
> enforced.)  Currently, the section about DRM requirements is
> unfinished (and there was a long discussion at the BPDG meeting today
> about the difficulty in completing that section), but there was a
> consensus of BPDG members that ATSC-demodulating devices would be
> required to output certain content only by "protected digital outputs"
> and "approved recording mechanisms".

Seth, I sugggest that in the interest of time, you just convert them
to pdf's and put them up.  They really do spell out the thinking of
the group.  

> Eric Blossom gave a great technical presentation at CPTWG in conjunction
> with BPDG; maybe he'd like to talk about that.

I spoke to them about these topics:

  * Software defined radio
  * Free software
  * GNU Radio in general
  * Our work on an all software ATSC receiver & why it was possible

People paid close attention, took lots of notes, and asked a few
questions.  I believe they actually got what I was talking about.  
A person from the MPAA specifically asked if we were doing the 8-VSB
demodulation in software.  I answered in the affirmative.  I believe
that up to that point they figured it couldn't be done.

I presented a couple of slides about what's possible with a dual
processor Athlon (one person suggested we call them "biathlons")
such as total number of ops/second, FIR taps/sec, Moore's law etc.

Prior to my talk, Fred von Lohmann of the EFF asked several pointed
questions about the group's intentions with their "Compliance and
Robustness Rules" vis-a-vis free software & the ability to innovate.
This provoked a lively debate that had no clear conclusion.  Fred also
pointed out that what they were up to was effectively the creation of
private law.

Also, in watching a few of the other presentations and talking to some
folks from the cable industry, it became obvious that exactly the same
issues are occuring with regard to digital cable, and broadcast HDTV
that is relayed on to cable.

As I understand it, you can either now, or soon will be able to go to
your electronics retailer (circuit city, fry's etc) and buy an "Open
Cable" (DOCSIS I think is the technical spec) compatible set top box
that you own.  It will demod the digital stuff and depending on the
model provide either analog or digital outputs of the content.  The
digital outputs apparently are supposed to have IEEE-1394 (Firewire)
5C protection or equivalent, which is supposed to ensure that the
digital output only goes to devices that implement the "Compliance and
Robustness Rules".  They aren't worried about the analog outputs.
Also, to receive the encrypted traffic that the cable guys are
broadcasting you need a "POD" which is effectively a crypto dongle
that you rent (subscribe) from your cable operator.  Unlike
conventional smart cards, the whole 45 Mbit/s data stream is passed
through the POD.

Note that the optionally encrypted stream from the cable head end is
orthogonal to the whether or not the ATSC stuff is carrying the
"Broadcast Flag" (the do-not-output-me-in-digital-cleartext bit).
Apparently according to FCC regs, at least the bottom tier of cable
must be unencrypted so that cable ready TV's will work.

FWIW, it would be technically possible to build an open source cable
box.  Parts of it would be in software.  Parts of the QAM demod would
probably take place in an FPGA on the board containing the RF tuner
module.

It was also apparent that certain of the CE representatives are not
inclined to upset their customers in order to give the content
producers what they want, particularly when it comes to DVD players,
DVD burners, etc.

Although there are a multitude of points-of-view represented at these
meetings, the fundamental concern of the movie folks is how to ensure
that their content doesn't end up on the internet.

Eric