Please leave your sense of logic at the door, thanks!

Archive for September, 2016

DRM and Web security

Wednesday, September 21st, 2016

For a few years now, the W3C has been working on a specification that extends the HTML standard to add a feature that literally, and intentionally, does nothing but limit the potential of the Web. They call this specification "Encrypted Media Extensions" (EME). It's essentially a plug-in mechanism for proprietary DRM modules.

Much has been written on how DRM is bad for users because it prevents fair use, on how it is technically impossible to ever actually implement, on how it's actually a tool for controlling distributors, a purpose for which it is working well (as opposed to being to prevent copyright violations, a purpose for which it isn't working at all), and on how it is literally an anti-accessibility technology (it is designed to make content less accessible, to prevent users from using the content as they see fit, even preventing them from using the content in ways that are otherwise legally permissible, e.g. in the US, for parody or criticism). Much has also been written about the W3C's hypocrisy in supporting DRM, and on how it is a betrayal to all Web users. It is clear that the W3C allowing DRM technologies to be developed at the W3C is just a naked ploy for the W3C to get more (paying) member companies to join. These issues all remain. Let's ignore them for the rest of post, though.

One of the other problems with DRM is that, since it can't work technically, DRM supporters have managed to get the laws in many jurisdictions changed to make it illegal to even attempt to break DRM. For example, in the US, there's the DMCA clauses 17 U.S.C. § 1201 and 1203: "No person shall circumvent a technological measure that effectively controls access to a work protected under this title", and "Any person injured by a violation of section 1201 or 1202 may bring a civil action in an appropriate United States district court for such violation".

This has led to a chilling effect in the security research community, with scientists avoiding studying anything that might relate to a DRM scheme, lest they be sued. The more technology embeds DRM, therefore, the less secure our technology stack will be, with each DRM-impacted layer getting fewer and fewer eyeballs looking for problems.

We can ill afford a chilling effect on Web browser security research. Browsers are continually attacked. Everyone who uses the Web uses a browser, and everyone would therefore be vulnerable if security research on browsers were to stop.

Since EME introduces DRM to browsers, it introduces this risk.

A proposal was made to avoid this problem. It would simply require each company working on the EME specification to sign an agreement that they would not sue security researchers studying EME. The W3C already requires that members sign a similar agreement relating to patents, so this is a simple extension. Such an agreement wouldn't prevent members from suing for copyright infringement, it wouldn't reduce the influence of content producers over content distributors; all it does is attempt to address this even more critical issue that would lead to a reduction in security research on browsers.

The W3C is refusing to require this. We call on the W3C to change their mind on this. The security of the Web technology stack is critical to the health of the Web as a whole.

- Ian Hickson, Simon Pieters, Anne van Kesteren

Posted in Multimedia, W3C | 7 Comments »