Hacker Newsnew | past | comments | ask | show | jobs | submitlogin

I think it would actually be a lot harder to apply something like EURion to video or audio, because of user expectations. Video and audio recorders are not expected to duplicate an existing artifact but to make a record, something you usually only get one chance to make. A EURion measure would be prone to abuse, breaking the user expectation that the given event had been correctly recorded.

The obvious example: What if a guy robbed a bank wearing a EURion t-shirt? None of the security cameras would record him, unless you had some exception for security cameras. And once you have an exception for security cameras, do you have an exception for, say, home security cameras? If so, anyone will be able to get a circumventing camera; if not, a lot of people will be upset.



I don't know how varied the possible DRM policies are, but it would seem possible to implement a 'generation counter' for DRM'd media which would restrict the number of copies.

e.g. the output of a video camera could be tagged with "allow 3 copies only". Each DRM-aware copy produced would decrement the # of allowed copies. The copied file would have an allowed-count of 0.

If a EURion-like image triggered this mode on the camera then that would probably be sufficient for manufacturers to avoid liability, but still serve to effectively close the "analog hole" (not really analog, but hey).

Yes, it all does seem unlikely. But so does tagging of printer output with yellow dots and EURion, and DRM in general.

http://en.wikipedia.org/wiki/Printer_steganography


It wouldn't work. This has all the problems of DRM in addition to all the problems of pattern recognition. Not a chance.


The problems of DRM are the ones facing us right now. Sure, individual implementations are buggy. If there's pressure, that'll improve over time.

The basic technical approach of the current DRM systems is sound, which is why they're a problem. High quality implementations would restrict people.

In terms of pattern recognition, you're not exactly doing facial recognitiion. The hard part would be putting a signal in there which was:

- distributed through the whole picture (so you can't just mask off the logo in the corner)

- sufficiently unobvious to humans threshold so that quality doesn't go down too far. Dunno about this one, apart from the fact that we have a lot of bandwidth here to hide a signal

- could survive basic disruption attempts by 'hackers' (colour filters etc)

It is a fairly tall order, but so is convincing the entire PC parts industry to put crypto in all their components against the interests of their customers.




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: