DLD09 Joi Ito has a recipe for a world without DRM - cunningly, it's software that stops you doing stuff you're not allowed to. Simple, yes? In the future world according to Ito, every object on the Internet will have licensing and copyright information attached to it, machine will talk to machine, and machine will overrule you if you're accidentally about to swipe something.
But it's still not DRM. We suppose it might be helpful in setting Arianna Huffington back on the straight and narrow, but it might also be more broadly unwelcome than Ito anticipates.
Speaking at the DLD09 new media conference in Munich earlier this week, Ito was pitching a machine-readable licensing system which he hopes (actually, claims, but his spiel is a little ahead of the game here) will become an Internet standard with the next generation of HTML. Ito stepped into Larry Lessig's shoes as CEO of Creative Commons two years ago, and describes the organisation as being in the process of converting from a political movement to a technical standard. In the context this is significant - pseudo revolutionary posturing out, beavering away in the standards organisations, in. And he's now pitching the fruits of that beavering.
In the future according to Ito: "Every object on the Internet will have a licence and copyright information and the author and the owner attached to the object, and if it's a derivative work, where it's a derivative work of." The licence Ito has in mind will be a Creative Commons one, but there seems no reason why other classes of licence couldn't use similar mechanisms.
And, "what will happen is, once we start building it into all the tools, into your camera, into Adobe Acrobat, into Google, you don't need DRM and watermarks. As long as it's built into the HTML, most of the people who matter will follow it."
In what way will they do this? "You downloaded some music, and say, 'I want to use this in my YouTube video,' it [the software] will say, 'Bap-bap! You can't do that because the copyright says you can't.'" Which does kind of look and feel like DRM, but as Ito says, it's not, it's a way to get away from DRM.
In some senses. "As long as the copyright information gets attached to all of the content on the Internet [a big ask, perhaps?], only the people who want to be shared will be shared, and all the people who want to be protected will be protected. In a year or two it'll be built into most of the tools so you don't really have to think about it."
Up to a point. Ito's presentation then skips merrily into visions of the sunny uplands of millions happily sharing, free of the expensive friction of confusing copyright Ts & Cs, and averts its gaze from the sewers where bad people continue to subvert, spoof and steal. Even if everything Ito says about next generation HTML standards is true - which it is not - he's talking about a system that presumes most people will be honest, and where the dishonest people will not be able to wreck the whole thing.
Creative Commons, he says, has "worked together" with W3C "to create RDFa [Resource Description Framework - attributes], which will be [in] the next version of HTML, but which is already being deployed by Yahoo! and Google and Microsoft and everybody else." There are however a couple of imprecisions here.
RDF isn't specifically about copyright, but is a framework for describing objects on the Internet. Creative Commons' efforts to create machine readable licences have been based on RDF since 2001, so "worked together" arguably overstates CC's influence in the development of the standard. Nor is RDFa "in" the next generation of HTML as such. RDFa is a W3C recommendation for XTML, but it is not as yet in the spec for HTML5.
W3C sources tell The Register that there is a section within W3C that would like RDFa included in HTML5, but that there is also strong opposition to the notion. This opposition has doubts over both the point and viability of RDF, and it may turn out to have been rash for Ito to align Creative Commons too closely with RDF. By talking up the pervasiveness of RDFa now, Ito may help it win, but there may be a price to pay for Creative Commons, and that price will be greater if it is closely identified with RDFa, and RDFa loses.
But if we imagine for the moment that it wins, that every object on the Internet does indeed get licensing information attached to it, and that a broad range of applications and devices acquires the ability to understand it - what then?
The system will certainly be a boon to the people Creative Commons is aiming it at, the honest ones who wish to enhance documents by using objects - audio, video, pictures - that they are permitted to use. Currently it can be difficult to nail down licensing Ts & Cs even when one specifically wishes to adhere to them, and this is the copyright "friction" that the system Creative Commons is pushing is intended to overcome.
But although recalcitrant software might interdict your attempts to use an object without a licence, it won't represent an absolute roadblock, because there's no DRM, no watermarking, and the copyright attributes aren't nailed absolutely to the object. Nor, if you think of it merely as an alert system or a convenience for people who want to use material in accordance with the wishes of the owner, does it necessarily do that, either.
Screen grab a picture or record an audio track, and unless your screen grab or recording software is primed to figure out what you're about to do and suggest that you might like not to do it, you could still lift a restricted document by accident. And figuring out whether or not use is permitted is complicated, too, given that rights can vary depending on the nature of the product, medium, and user. Is it non-commercial? What's the circulation? Even if you built a nightmare copyright bureaucracy into every single object and piece of software on the Internet, it's hard to see how that would work.
And fairly easy to see that RDF's opponents might have a point. The problem might however be that Ito is overselling automation. The Creative Commons system is certainly about machine readable licences, and in production circumstances where the producers have close control over processes and have standardised them to be aware of RDFa licensing, it could be useful. But pitched as a more general system, and/or as a reason why there's no need for DRM, it doesn't look credible.
Or, given that you've got all of the licensing and usage Ts & Cs associated with every object on the Internet, it could even look like a reason for DRM. If you have copyright attributes associated with everything, mightn't it make sense for you to want to protect that association? ®