4 ms·
In most DRM systems the information is distributed encrypted. The decryption keys are given to technology developers who have specifically promised to obey the
by schoen 13y ago
In most DRM systems the information is distributed encrypted. The decryption keys are given to technology developers who have specifically promised to obey the DRM rules, as well as to make their technology hard for users to understand or modify so that the users can't easily undo the restrictions or extract the decryption keys.
Hence a browser developer or OS developer or developer of whatever software is in question wouldn't be permitted (by the DRM system's inventor or administrator) to get decryption keys if they didn't promise to implement these restrictions.
Some of the people who invented the modern DRM business ecology called this "the intersection of technology, law, and commercial licensing" (the title of a 1996 article by Dean Marks and Bruce Turnbull). Here, the "technology" is DRM implementations -- including software obfuscation and other measures; the "law" is anticircumvention laws like the DMCA §1201 that make it risky for people to use the decryption keys in ways that industry dislikes; and "commercial licensing" is the permission from a DRM developer to interoperate with that DRM, including "compliance" rules (about the functionality of the technology product) and "robustness" rules (about tamper-resistance), that result in the licensee being issued decryption keys.
In my view (I worked on EFF's objection) this is a deliberate attack on software interoperability: the whole point is to allow someone to try to prevent interoperability with software that hasn't been "approved". And it's also in extreme tension with the idea of having browsers that end-users can modify (their individual instances of).
- PeterisP 13y agoI see no way how an open source system can implement any effective DRM standard while staying open source. If a proper open source system has a component that enforces DRM, and is functional when I download it, then it includes those keys; but gives me an unconditional right to use and modify it. And I am physically able to modify it, un-implementing those restrictions. If part of the system cannot be modified by me, then the whole is not open source, and any open source system such as Firefox shouldn't include that part or standard.
- drdaeman 13y ago> If part of the system cannot be modified by me, then the whole is not open source "Open Source" definition does not include any clauses that require hardware manufacturers to provide you encryption and/or signing keys, so you could run your code. GPLv3 and "Free Software" are what you're looking for.
- schoen 13y agoThe Free Software Definition doesn't require free software to be copylefted or to include measures against TiVoization or against proprietary or restrictive downstream products. GPLv3 does this and the BSD license doesn't, and both are free software licenses. As I said in another comment, people who usually say "free software" are more likely to think that preventing restrictive downstream products is an important goal than people who usually say "open source". But that doesn't mean it's part of the definition of what it means to be free software. EDIT: I also think the comment the parent replied to was right to say "then the whole is not open source". BusyBox is both free and open source even though its license allows it to be included in the locked-down TiVo -- but the TiVo as a whole is not open source.
- schoen 13y agoI agree. In previous discussions about this, some people emphasized the idea of open source development, as opposed to giving an open source software to an end user. For example, you could run a binary through an obfuscator after compiling in a decryption key -- from source code that had been published and distributed under an open source license. If the license isn't a reciprocal/copyleft license, this is probably not a license violation, but it seems wrong to say that the user who receives the binary is being given open source software. This issue reflects the way that people have had very different ideas about what the point or purpose of free and open source software is (in some ways, reflecting the split between people who preferred to say "free software" over "open source" and vice versa). It's also a very concrete issue today in whether people call, say, the Chrome browser "open source". Most of their source code is downloadable, derived from the fully open-source Chromium project, but in Google's current practice, users never get the complete source code to the Chrome binaries that they run. If you're focused on the development process, it might almost make sense to call Chrome "open source" because almost all of its source code is distributed, licensed, and developed in an open source manner -- but if you're focused on what users can do with the software, it's obviously just a proprietary application (with a proprietary EULA, to boot).
- makomk 13y ago> For example, you could run a binary through an obfuscator after compiling in a decryption key -- from source code that had been published and distributed under an open source license. Of course, if you do that you may as well just not release the source code at all, because there's no way for the user to tell that the binary does actually correspond to the source without also being able to extract the encryption key and break the DRM.
- schoen 13y agoI think that verifiability is an important security benefit from publishing source code and I hope to talk more about that soon. Some people think that publishing source code is first and foremost a way to get other people to collaborate on its development, not to ensure any particular rights or knowledge or safety for people who end up using the software. For example, you could imagine a consortium of people who each make a super-proprietary locked-down thing and they publish and collaborate on the code of some libraries that their respective locked-down things need. They actively do want other locked-down thing makers to comment on how to make the code better and contribute patches, but they actively don't want customers to use that knowledge to make the thing less locked-down (or to be able to verify what it does or doesn't do). This is a situation that we often encounter in the real world, and in fact some of the locked-down thing makers are even surprised when people say the contrast in their behavior with respect to these audiences is strange or hypocritical, because they didn't know or didn't remember that other people think software freedom is partly or mainly meant to benefit users.
- jcromartie 13y agoContent keys do not have to be shipped with the browser. They are provisioned on a per-user and possibly even per-use basis for the specific content.
- alextingle 13y agoBut if the content keys work with actual open source software, then that software can always be modified to decode the content and remove any restrictions that the publisher might wish to apply. That's why they have to obfuscate things at the code level too.
- IanCal 13y ago> If part of the system cannot be modified by me, then the whole is not open source So would you say that firefox is open source now? If they implement this as a plugin, what makes it technically different from using Flash?
- chris_mahan 13y agoDoes it mean that Firefox could not both implement DRM and remain open source? Excellent!
- excitom 13y agoIn past years in my career I worked in game development. I can say without hesitation that anything client side can be reverse engineered and hacked. Any limitation built into a browser can be circumvented, and it will happen in a matter of days from release.