Barry Fox

The movie industry, like the music industry, is its own worst enemy. Studios, like record companies, see only what they want to see. That way they can blame whoever they want for lost profits.

AT&T's research lab in New Jersey showed that 77 per cent of all movies available illegally on the internet came from sources inside the industry.

But at the PEVE European video and movie industry conference in Marseille recently, the studios were still not willing to face the fact.

"The Motion Picture Association [MPA] does not accept the 77 per cent figure," said Dara MacGreevy, one of many lawyers employed by the MPA.

I was at the conference to suggest that the MPA look closer at the new European copyright laws, and learn about technology. The laws make it an offence to circumvent copy protection. But they introduce a new qualification: the protection technology must be 'effective'.

321 Studios, which sells the DVDXCopy program that lets a PC make digital clone copies of a DVD movie disc, has said it will use the 'effective' clause as a defence against the MPA's lawsuit.

If the CSS encryption used to protect DVDs is so weak that the unlock keys are available on the internet, how can it be 'effective'? What about the gaping analogue hole left by a DVD's unclear standards?

Alistair Kelman, a barrister with 25 years' experience of copyright disputes, said: "The courts might have sympathy with the argument that protection is not effective if a schoolboy can defeat it."

Movie industry executives and lawyers pretend to know it all, but don't understand the difference between digital cloning and recoding analogue feeds to make a digital copy.

So at Marseille I showed them a copy of The Matrix Reloaded made by connecting the analogue output of an off-the-shelf DVD player to the analogue input of an off-the-shelf DVD recorder.

The movie plus some extras fits onto a single DVD blank costing 50p, with picture quality good enough to fill a large projection screen. Suddenly, some of them were interested in what I'd long been telling them.

Analogue copy protection (CGMS-A, the Analogue Copy Management Generation System) is described by the Procedural Specification for the CSS licence (Content Scrambling System for digital copy protection). Section 6.2.1.1.(1)(a)(i) says NTSC DVD players should put the CGMS-A copy-protection 'flags' in unused picture lines 20 and 21 of the analogue output signal.

But Section 6.2.1.1.(a)(ii) is less clear on where Pal/Secam players should put the flags. Instead of defining the flagging lines for Pal/Secam, the standard just waves the reader off to a different standard ETSI/ETC 300294.

This is a Wide Screen Signalling standard used by broadcasters to control widescreen picture display. There are sub-sections on subtitling and analogue copy control, one of which says that anti-copy signals should be in Line 23.

Nowhere does the DVD standard specify that Line 23 is where the copy control signals should be for Pal/Secam territories.

Philips, Pioneer and Panasonic DVD recorders on the market today correctly look in Line 23 when asked to copy a Pal/Secam disc. For NTSC they look in Line 20 and perhaps also Line 21.

But DVD players are mass-produced on Chinese factory lines which serve both NTSC and Pal/Secam markets. Players are putting copy control signals in the wrong lines.

I first raised the issue with the DVD Forum a year ago. It passed the buck to Licence Management International, which represents the DVD Copy Control Association.

LMI agreed in December 2002 that "no specific line is defined for Pal or Secam except by reference to ETC 300294". But LMI said it was up to me to talk to the studios.

So in February 2003 I told the MPA: "Some Pal/Secam DVDs can be routinely copied to consumer DVD recorders, despite the fact that CGMS-A should be preventing it."

The MPA first tried to fob me off by saying the standard is "quite clear and can be viewed by going to the DVD Copy Control Association website". But later it owned up. In September 2003 the CCA amended the spec, but left the CGMS section as unclear as ever.

The LMI stated in November 2003: "The CSS procedural specifications were modified recently. They did not have the specific modification that you were looking for."

So I went back to the MPA. Brad Hunt, chief technology officer, has gone back to maintaining that the "spec clearly states that CGMS-A must be generated in Line 23 as defined in ETS 300 294 standard".

Not so. The DVD standard still does not mention Line 23. Hunt added: "Some DVD player manufacturers do not appear to be properly generating CGMS-A signalling on the analogue video output ... the MPA has begun sending letters ... requesting how [manufacturers] plan to remedy it."

How many manufacturers has the MPA approached since February with a recommendation for corrective action to make DVD copy protection 'effective', and which of them have taken action?

Surprise, surprise, the MPA is not saying. But perhaps it will have to when 321 argues in court that DVD protection is not 'effective'.

Like this story? Spread the news by clicking below:

 del.icio.us     Digg this     reddit!