ARL Public Policy Briefing
October 2023
|
Katherine Klosek, Director of Information Policy
|
In October, ARL asked the US Department of Justice to consider regulations that advance born-accessible publishing standards. The Canadian Association of Research Libraries (CARL) is preparing a submission to the Government of Canada’s generative AI consultation. The Library Copyright Alliance (LCA) pushed back on rightsholders’ concerns about the Library of Congress’s proposed rule on e-deposits. In comments to the US Copyright Office, we restated our position that no amendment is needed to address issues of copyright and generative AI. And, LCA will continue to engage in the US Copyright Office Digital Millennium Copyright Act (DMCA) rulemaking.
In Canada, Liberal MP Greg Fergus was elected as Speaker of the House of Commons on October 3, making history as the first Black Canadian to hold the role and taking over the position from Anthony Rota who resigned from the role amid controversy.
The monthly Public Policy Briefing highlights developments in ARL’s public policy priorities, issues that are relevant to the research library community in the United States and Canada, and details on advocacy conducted by ARL and CARL. Please encourage your colleagues to sign up for the Public Policy Briefing.
If you have questions or suggestions, please email me at kklosek@arl.org.
|
Copyright and Fair Use/Fair Dealing
|
Association of Research Libraries Calls for Born-Accessible Publishing in Comments to US Department of Justice
|
In response to the US Department of Justice’s (DOJ) notice of proposed rulemaking “Nondiscrimination on the Basis of Disability; Accessibility of Web Information and Services of State and Local Government Entities,” the Association of Research Libraries (ARL) submitted comments supporting the federal government in codifying the Web Content Accessibility Guidelines (WCAG) into law for the first time, and highlighting strategies ARL member libraries are using to accelerate the adoption of born-accessible publishing.
The DOJ proposal would amend Title II of the Americans with Disabilities Act (ADA), which governs accessibility for state and local government, including public libraries and academic libraries that are part of institutions of higher education. The proposed rule would require state and local government entities to adhere to WCAG 2.1 Level AA, with exceptions for certain categories of web content. For instance, in the higher education context, class or course content posted to a learning management system (LMS) would not need to be accessible, unless and until a student with a disability who would be unable to access the course content posted on the LMS enrolls in a particular class or course.
The DOJ points out that this approach is consistent with obligations of public entities under Title II of the ADA. But this rulemaking governs digital accessibility, and the digital works acquired by libraries for use in course and class work are inherently more compatible with assistive technology. Today, it is far more efficient for a publisher to make accessibility improvements to the born-digital version than for an institution to convert an analog copy into an accessible digital copy, or to unlock a digital copy and convert it to a more accessible format. ARL’s comments recommend a born-accessible approach, in which vendors would design digital academic content with the needs of people with disabilities in mind.
The proposed rule reiterates that it is the public educational institutions that are responsible for meeting their legal obligations. But libraries of all sizes experience considerable challenges when negotiating for accessible digital scholarly works. Nevertheless, to address this barrier and meet the institutional legal obligations, libraries are working to move the scholarly publishing industry toward adoption of born-accessible standards through institutional and consortial negotiation strategies. For example, UC Berkeley Library has successfully proactively shifted the responsibility of producing accessible materials to vendors and publishers through its e-resources licensing process. University of Washington (UW) Libraries has elevated accessibility to a “deal-breaker” in its licensing requirements, and does not allow automatic renewal of subscriptions of inaccessible products. And, the Iowa State University Library’s Access and Acquisitions Department requests strong accessibility language in all new e-resources license agreements, in accordance with the university’s Digital Accessibility Policy. By creating loopholes such that WCAG does not need to be applied to password-protected course content, the DOJ’s proposed regulations have the potential to undermine library and university efforts to require vendors to provide born-accessible content and technology.
ARL is pleased that the federal government is adopting web-accessibility guidelines into law. We anticipate the US Department of Education will take the same approach of proposing exceptions to web-accessibility guidelines when it issues its notice of proposed rulemaking on Section 504 of the Rehabilitation Act. ARL and our members will engage with that rulemaking as well.
|
CARL to Prepare Submission on Government of Canada’s Generative AI Consultation
|
On October 12, the Government of Canada launched a consultation on generative AI tools and their implications for copyright holders, specifically designed to address the use of copyright-protected works in the training of AI systems; authorship and ownership rights related to AI-generated content; and liability, especially when AI-generated content could infringe existing copyright-protected works. CARL will be submitting a brief, which is due December 4.
|
Copyright and Fair Use/Fair Dealing
|
House of Commons Adopts Amendments to Canadian Copyright Act
|
LCA Pushes Back on Rightsholder Groups’ Comments on US Copyright Office Electronic-Deposit Proposal
|
Last month, LCA submitted comments in support of the US Copyright Office’s proposal to expand the categories of e-deposits that users may access on-site, subject to limitations; for instance, the works may only be accessed through computer terminals in the Library of Congress reading rooms, which are not connected to the Internet and are under the supervision of the library staff. Expanding the existing on-site access rules to a larger set of works submitted electronically for deposit is one way of ensuring that the public can continue to access works in the Library of Congress collections.
A rightsholder group, the Copyright Alliance, expressed concern that this expansion would grant the Library of Congress access to all electronic deposit copies of every published copyrighted work, and suggested the Copyright Office heighten security and protection of electronic deposit copies before instituting its proposed rule. In reply to the Copyright Alliance, LCA reiterated that the Copyright Office’s proposal has a limited scope—expanding the categories of e-deposits to which the public would have highly circumscribed access—and that it is highly unlikely that anyone would take advantage of this rule to make infringing copies.
|
LCA to US Copyright Office: New Legislation Is Not Needed to Address Copyright Issues Related to Generative AI
|
In response to the US Copyright Office notice of inquiry on generative AI, the Library Copyright Alliance reiterated our position that the existing US Copyright Act, as applied and interpreted by the Copyright Office and the courts, is fully capable at this time to address the intersection of copyright and AI without amendment.
LCA’s comments highlight ways that generative AI can support libraries in providing access to information, and caution against prematurely hindering the evolution of generative AI as a useful tool. LCA notes that the courts are the most appropriate venue to address issues like the lawfulness of ingesting training data, and whether the outputs of a generative AI system are infringing. While LCA objects to deepfakes that abuse an individual’s name, image, and likeness, we note that this is not a copyright issue, and that any legislation that would restrict the copying of “styles” would chill free expression.
In response to the question of whether copyright owners should have to opt in to the use of their works for training materials, or if they should be provided with the means to opt out, LCA makes the point that copyright owners already have the technical means to opt out of their works being used for training materials. However, LCA’s position is that browse-wrap license terms that prohibit the ingestion of materials should not be enforceable, and Congressional intervention may be necessary to clarify that exceptions and limitations in the US Copyright Act prevail over license terms that are inconsistent with those exceptions.
LCA reiterates that ingestion for training purposes is a fair use, and that collective licensing is not necessary because copyright owners’ consent is not required to train generative AI models. The Copyright Office should focus its report on preventing the generation of infringing derivative works, rather than relitigating the question of whether ingestion is fair use and contemplating new collective licensing schemes. Finally, LCA suggests that the office consider supporting voluntary measures such as licensing arrangements and codes of best practices.
The Copyright Office will issue a report on its AI initiative in 2024.
|
US Copyright Office to Renew Exemptions to Permit Circumvention of Access Controls on Copyrighted Works
|
As part of the Digital Millennium Copyright Act (DMCA) rulemaking, the US Copyright Office considers petitions for the renewal of current exemptions, and petitions for new exemptions. In October, the office announced its intention to recommend the existing exemptions for which petitions were submitted for renewal. The full list of renewals is in the announcement; LCA supported petitions for the following classes of works, which will be renewed:
- motion pictures for educational purposes by college and university faculty, students, or employees acting at the direction of faculty, or K–12 educators and students;
- motion pictures for educational uses in massive open online courses (MOOCs);
-
motion pictures for educational uses in nonprofit digital and media literacy programs offered by libraries, museums, and other nonprofits;
- motion pictures for the provision of captioning and/or audio description by disability services offices or similar units at educational institutions for students, faculty, or staff with disabilities;
-
motion pictures for preservation or the creation of a replacement copy by an eligible library, archives, or museum;
- text-and-data mining of motion pictures by researchers affiliated with a nonprofit institution of higher education, or at the direction of such researchers, for the purpose of scholarly research and teaching;
-
text-and-data mining of literary works that were distributed electronically by researchers affiliated with a nonprofit institution of higher education, or at the direction of such researchers, for the purpose of scholarly research and teaching;
-
literary works or previously published musical works that have been fixed in the form of text or notation, distributed electronically, whose technological measures interfere with assistive technologies;
-
individual play by gamers and preservation of video games by a library, archives, or museum for which outside server support has been discontinued, and preservation by a library, archives, and museum, of discontinued video games that never required server support; and
-
computer programs, other than video games, for the preservation of computer programs and computer program–dependent materials by libraries, archives, and museums.
The Copyright Office also initiated three rounds of public comment in the newly proposed exemptions. LCA will work with the Authors Alliance and the American Association of University Professors (AAUP) on expanding the exemptions for text-and-data mining to allow researchers to share corpora of motion pictures and literary works with researchers at other institutions. LCA will also collaborate with the Software Preservation Network (SPN) on expanding the exemption for preservation of software by libraries, archives, and museums by removing the requirement that electronic distribution, display, or performance of software be made to “only…one eligible user at a time.”
|
|
|
Manage your preferences | Opt Out using TrueRemove™
Got this as a forward? Sign up to receive our future emails.
View this email online.
|
21 Dupont Circle NW #800 | Washington, DC 20036 US
|
|
|
This email was sent to .
To continue receiving our emails, add us to your address book.
|
| |
|
|