Many were thrilled earlier this spring when the FY 2018 Omnibus Appropriations Law included the “public access to all non-confidential CRS reports.”
Well, not so fast it seems. Daniel Schuman, Kevin Kosar, and Josh Tauberer (3 folks doing great work over the last several years on the CRS reports issue) have found that the “Library plan to publish CRS reports falls short of the law, and is unduly expensive.” LOC plan “does not comport with the law or best practices for creating websites and is unusually expensive,” they wrote. By contrast, their own collection of 14,000 reports on everyCRSreport.com cost about $20,000. Another point of criticism is that the the library’s plans to publish the reports only as PDF files — rather than in both HTML and PDF formats — making them harder to access on mobile devices and potentially inaccessible to people with visual impairments. The plan also apparently ignores a directive to publish a separate index of all the reports published by CRS, they said, which would make it easier for laypeople to see all available documents at once.
the group makes several important recommendations. To comply with the law, the Library should:
- Update its implementation plan to ensure that it publishes all CRS reports — we believe there are many more than the 2,900 the Implementation Plan says will be published by Spring 2019 — by the statutory deadline of September 19 of this year. We request it aim for September 17th, which is Constitution Day. The Library’s implementation extends beyond April of next year;
- Update its implementation plan to include all CRS Reports, including insights, infographics, sidebars/legal sidebars, in focus, and testimony;
- Revise its implementation plan to ensure that HTML versions of the reports are available to the public just as they are already available to Congressional staff — this would help the visually impaired read the reports as well as allow reports to be read on mobile devices;
- Revise its implementation plan to include an index of CRS reports, in accordance with the law’s requirements; and
- Review the code we published to see whether it would help the Library meet its obligations, in particular our automated author information redaction functionality, or whether the Library could develop an automated tool that would enable it to comply with the timeline.
With respect to the website design, the Library should:
- Consult with the Government Publishing Office and the public on how best to implement bulk access;
- Develop a plan to respond to any initial heavy loads on the website;
- Implement a robust website search capability and develop a plan to do so;
- Create predictable URLs for CRS reports and a landing page for a report series, and set forth a plan to do so;
- Keep down costs by examining our approach to see whether it can use some of our techniques to save money; and
- Consider engaging an entity like the General Services Administration’s 18F to help keep down costs and ensure a quality product.
So far, attempts to communicate with the Library of Congress have fallen on deaf ears. So if any of our readers have connections to Carla Hayden’s office, please forward this on to her.
This work is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License.