Our business world stands today at the threshold of a new industry, as “big data” gains value on a very big scale. Vast quantities of information are evolving rapidly into an asset class in its own right, akin to software and hardware with their own ecosystems and competitive dynamics and innovation cycles. And of course, legal issues.
There are troves of data being collected and stored and used by governments and companies globally. Complex algorithms are being developed to extract value from all this data. Retailers, for example, have a more detailed account of our lives than we ourselves can access. With their massive number of customer touch points, they’re so data-rich that they know what products and services customers want before consumers even know it themselves.
Some of this might sound a bit creepy. While retailers are not breaking any laws, there is much debate about the need for policies to harmonize privacy, prudence, social acceptance, and ownership with an undeniably massive business opportunity. But setting aside the privacy issues, what sort of legal regime do we need to ensure this new industry grows, and provides maximum private and public value? Do we leave the growing big data asset class to the law of the jungle? Or do we need new rules to foster growth and make our country the world’s most attractive home for the business of big data? These aren’t just abstract legal questions; they’re questions going directly to our patent, trademark, and copyright — intellectual property — laws.
When software was in its nascent stages, it was given away for free to sell hardware. Over the course of a few short decades an industry emerged. And along with it our legal system adapted to foster growth through new and evolving copyright, patent, and trade secret regimes. Now, software is a multi-hundred-billion dollar industry enjoying rapid growth and innovation, delivering bright new consumer benefits and life-saving breakthroughs at warp speed. And the U.S. leads the software industry practically across the board.
Many would say our country’s leadership is attributable in no small measure to our supportive intellectual property laws that have struck just the right balance between providing incentives for investment in innovation and providing access to third parties.
So taking software as a guide, it is fair to say the stakes are high, and policy matters. But no existing policy device within our current intellectual property system is tailored to mediate big data. However, it may be possible to interpret or re-fashion our IP laws so that the best protections and incentives are afforded to the data industry and the maximum social good is realized. While we do not yet have a complete roadmap for the interplay between big data and IP, we do have a few viable starting points. For one, our patent and copyright systems can continue to play their current roles – protecting inventive ways to draw value from data (through patents) and the creative aspects of data (through copyrights). There is certainly no evidence that a much greater or lesser level of protection is called for in these areas, and there is wisdom in the old saw: if it ain’t broke, don’t fix it.
The trademark system (think brands like Coke (KO) and McDonalds (MCD) ) may have an especially important role to play in accelerating the development of data as an asset class. Certification marks in particular may prove quite useful. Businesses and consumers alike benefit greatly from the vetting and standards compliance testing performed by certification organizations, such as the widely recognized and trusted UL (Underwriters Laboratories).
Once these organizations certify compliance by an applicable product or service, they permit the purveyor to affix a certification mark. When you purchase a lamp or a toaster, the “UL Certified” mark provides assurance that the appliance will plug into the socket in your wall and work with your home’s electrical system.
How would this work in the big data industry? A data certification mark would attest that the applicable data is accurate, properly formatted, and thoroughly covers the subject. In effect, the mark would certify that the data’s prongs will fit into the analytical software’s wall socket.
One can envision standards-setting organizations establishing norms and permitting use of applicable certification marks by those who collect, clean, organize, format, store, retain, curate, and provide data according to an agreed-upon level of quality and accuracy. Such standards in turn would enable just the kind of cross-use (between industries like retail and healthcare), follow-on use (beyond the purpose for which the data was originally collected, such as where soil composition data is used to understand moisture levels), and study (such as by academic, government, or industry researchers) that promise to make big data a huge creator of value.
At present, there is no reason to believe radical changes are needed in the IP system to render it safe for the advent of the data era. New opportunities for the trademark regime along the lines described above should be considered. In the end, the nations and regions that maintain a policy focus on fostering the growth of the data industry will be well positioned to lead into another promising field spinning out from information and computer technology.
David J. Kappos is a partner at New York City-based law firm, Cravath, Swaine & Moore, where he supports the firm’s clients with a wide range of intellectual property issues. From August 2009 to January 2013, Kappos served as director of the U.S. Patent and Trademark Office, advising the president, secretary of commerce and the administration on intellectual property policy matters.