Rightsholders, Tech Companies Debate ‘Diligent Search’ Requirement in Orphan-Works Comments

It’s unreasonable to ask potential users, especially mass digitization projects, to identify a copyright owner, which can be costly and time-consuming and does not protect the user from copyright holders that appear later, Microsoft wrote. “Where mass digitization is involved, the cost of such searching and uncertainty is multiplied by a factor of thousands, if not millions.” To prevent mass digitization processes from bearing the “impractical, if not impossible” burden of diligently searching for copyright owners, orphan works legislation could create a safe harbor for mass digitization projects that make their digital copies available to the public, Microsoft said. Additionally, registries of works should be created to best gauge how many works are not claimed by authors or publishers, Microsoft said, encouraging that Google release to the Copyright Office the data it collected as part of its Google Books program.

Mass digitization has benefits that go beyond preserving works, Microsoft wrote: “Providing the public with the ability to search and analyze large volumes of digital works enables scholars, scientists, journalists, educators, students and others to gain new insight and knowledge not possible when forced to track down and view each work individually.” Microsoft cited an English/Haitian Creole translator it developed after the Haiti earthquake in 2010 that required “parallel data between English and Haitian Creole for training the engine.” The translator would not have been possible without the efforts of those “who made data and dictionaries available with minimal license restrictions,” Vikram Dendi, Microsoft senior product manager, wrote in a blog post excerpted in the comments.

Mass digitization projects that involve orphan works -- including U.S. and non-U.S. works -- should have liability limitations for for-profit and non-profit organizations that use copyrighted works after unsuccessfully conducting diligent searches for the works’ copyright owners, Google said. The definition of a diligent search “would have to be objective,” the company continued: The definition should include “a specific database or set of databases” -- which “must support automated searching” -- and the “user needs to have a clear and definite stopping point for the searches.” Google is developing its own tools that “help people who are trying to find copyright owners in order to get permission to make uses of their works,” the company said, pointing to Google Images search-by-image tool and the Google Books project that “digitized and indexed many volumes of the U.S. Copyright Office’s Catalog of Copyright Entries."

Google also suggested amending the federal criminal code to say that a copyright owner cannot seek statutory damages and attorneys’ fees from infringers if the copyright owner’s contact information was not accurate in the Copyright Office’s registration records at the time of the infringement. Additionally, the Copyright Office should “make the contact information of owners accessible publicly, online, in a machine-readable format,” Google said: Though the Copyright Office would “have to shoulder the logistical burdens of digitizing the registration records for all in-copyright works and of building a database around them that could handle automated mass queries ... a big part of that work is already being done” through the office’s current digitization projects.

Mass digitization projects are not necessarily dependent on a solution to the orphan works problem, Google wrote: The company’s Google Books Library Project “displays only short snippets of the books that are in copyright (including orphan books)” to index the books. Google said it would welcome a separate inquiry into mass digitization, which could include the issue of disputed rights that occurs when authors and publishers disagree over “the interpretation of old book contracts."

Mass digitization projects should not be exempted from needing to conduct diligent searches to locate copyright owners, the MPAA wrote. The fact that digitizing a large number of works means undertaking the time-intensive and costly process of locating copyright owners for all of those works “does not mean the user should be free to make the use anyway, on the basis of some ‘mass digitization’ exception,” the comments said. Recent mass digitization cases, including Authors Guild v. Hathitrust, did not focus on the reasonability of the burden to locate copyright owners, the group continued: The cases focused on the users’ belief “that, because of the social benefit their massive copying would assertively produce, the copyright law made their conduct non-infringing, and thus no authorization was necessary” from the copyright owners. The MPAA encouraged the Copyright Office to consider mass digitization a separate issue from the orphan works problem.

The orphan works problem may encompass fewer works than it has in the past, thanks to technological advances that allow users to find works’ copyright owners, including “the greater accessibility of U.S. Copyright Office data; the growth of voluntary registries and other data sources ... ; and the overall acceleration in quality and ubiquity of online tools for finding individuals and companies,” the MPAA wrote. The goal of any orphan legislation “should be to reduce the frequency of situations in which copyright owners cannot be found” and then to find ways to let individuals use works whose copyright owners can’t be found, the group said. In its comments, the Authors Guild said the orphan works problem is “vastly overstated.” A 2005 survey of the group’s members “found that 85 percent had ‘never’ or ‘rarely’ been unable to reach a copyright owner to request permission to use a copyrighted work,” the guild wrote.

Diligent searches are not the solution to the orphan works problem, though it may be a part, the Authors Guild wrote. Diligent search requirements have incorrect incentives, “rewarding failed searches with uncompensated use of copyright protected materials,” the group said. Instead, the Copyright Office should consider an orphan work licensing system, like “foreign licensing and collecting organizations,” the guild wrote. “Collective licensing for a well-defined, limited set of uses may be the only means of addressing the complex compensation, control, and security issues raised by the mass digitization of books."

"Orphan works legislation is not the right way to address” mass digitization, the RIAA wrote. Mass digitization projects should be required to perform a diligent search for what is usually “a readily-identifiable copyright owner,” the group said. If a project has searched but cannot find the owner, “it may treat the work as orphaned under the [Copyright Act] and proceed to use it.” Orphan works legislation should not be written to permit mass digitization projects to avoid “the burden of having to locate owners and seek their permission to digitize,” the RIAA said.

"It is unusual to encounter an orphan problem for” musical recordings because recordings from major music companies since 1972 have “almost invariably” been registered for copyright, and the copyright information can be found in “widely available and reasonably comprehensive” databases, the RIAA wrote. (Federal law first covered copyrights in 1972.) The group said it supports the formation of an “intent to use” database, which would require users to identify the potentially orphaned works they want to use, allowing “copyright owners to exercise diligence to ensure that their works are not erroneously treated as orphaned.” Orphan works legislation should specifically describe requirements for a diligent search to find a copyright owner, which will vary by industry, the comments continued. Additionally, the legislation should limit liabilities for all entities -- including commercial enterprises as well as nonprofit institutions -- if the infringer made such “a good faith, diligent search to find the owner.”