Many of the almost 24,000 technical standards maintained by the International Standards Organization (ISO) are subject to copyright restrictions and are not freely available.
Two weeks ago, Jon Sneyers, senior image researcher at Cloudinary and co-chair of the JPEG XL (ISO/IEC 18181) adhoc group, invited fellow technical experts to collaborate on an open letter urging the ISO to set its standards free.
In an email to The Register, Sneyers explained that paywalled, copyrighted standards inhibit education and innovation.
"Specifically for JPEG XL (or codecs in general), a free spec makes it a lot easier for external enthusiasts to make an alternative implementation," he said "Often this would be done as a free-and-open-source software hobby project, not necessarily with the goal to make a better implementation, but just as a personal learning project. With a free spec, there might be a few of such attempts, some failing, some succeeding."
The value of alternative implementations, he contends, is that they help verify the correctness of the spec. He notes that
libjxl, the reference implementation of the JPEG XL spec, is the only such library at the moment, which makes it less likely discrepancies between the spec and the implementation will be found and more likely that
libjxl, compliant or not, will become the de facto standard, thereby diluting the authority of the ISO spec.
"With a paywalled spec, there will likely be no such attempts, because a hobbyist is unlikely to pay a big fee up front just to read the spec (which is something you'd do before you even decide to have a go at it and try to implement it)," he said.
Via Twitter, Ian Graham, senior lecturer in operations management at the University of Edinburgh, explained the problem with paywalled specifications succinctly. "It is very difficult to teach about ISO standards when the students can't access them," he said.
The collaborative open letter notes that in 2019 the Switzerland-based ISO, which through its national affiliates brought in CHF 6.6m ($7.28m) from publications sales and CHF 12.9m ($14.23m) in royalties from publication sales by national affiliates, earned almost as much in spec fees as the organization collected in membership fees that year (CHF 21.2m or $23.39m).
And it claims that the ISO Central Secretariat has "begun strictly and narrowly enforcing the criteria for Publicly Available Standards (PAS)" by paywalling new editions of standards that were formerly free, redefining Technical Reports so they can no longer be free under any circumstances, and "pressuring other standard organizations with which it collaborates and that have a policy of publicly available standards (e.g. ITU-T), to change their policy towards making joint documents non-publicly available."
Sneyers believes the ISO should be moving in the opposite direction by tearing down its paywall.
"I think international standards are great for worldwide interoperability and as a way to do knowledge transfer: collecting best practices and international expertise and condensing it into a standard," he said. "I see a paywall as a significant obstacle to that."
A handful of other technical experts have already signaled their agreement as co-signatories, including:
- Luca Versari, ISO/IEC JTC 1 / SC 29 / WG 1, editor of ISO/IEC 18181 (JPEG XL)
- Leonard Rosenthol, Chair of ISO TC 171/SC 2, member of various TC’s including ISO/IEC JTC 1 / SC 29 / WG 1 and project leader for over a dozen different standards.
- Bryce Adelstein Lelbach, ISO/IEC JTC 1 / SC 22 / WG21 (C++) / LEWG chair, INCITS (US) PL22 chair
- Jeff Hammond, ISO/IEC JTC 1 / SC22 / WG5 (Fortran) and WG21 (C++) contributor
- JF Bastien, SC22 / WG21 (C++) EWG Chair
- Eric Portis, W3C WICG & WHATWG participant
- Patrick H. Lauke, W3C PEWG chair, W3C AGWG member
The letter cites numerous arguments for a policy change. Among them: resentment from technical experts who write standards without compensation only to see the ISO charge for their free labor; the risk of irrelevancy if experts choose to pursue standardization through other organizations that don't charge; and the possibility that the ISO's reputation for neutrality will be tarnished if it's perceived as a for-profit entity.
The ISO did not immediately respond to a request for comment.
Sneyers's campaign echoes similar paywall demolition initiatives, such as the ongoing effort to make US court document service Pacer available at no charge, as well the late Aaron Schwartz's attempt to open up JSTOR, and, of course, Sci-Hub.
- Open standard but not open access: Schematron author complains about ISO paywall
- New ISO standard kind-of explains how to ignore standards
- So, you're 'ISO 27001 accredited', huh? Just saying so doesn't cut it
- ISO blocks NSA's latest IoT encryption systems amid murky tales of backdoors and bullying
Carl Malamud, a public domain advocate and president of public.resource.org, lauded Sneyer's effort to change the ISO.
"Standards that are not freely available lose their very reason for being," said Malamud, who has spent years working to make US government documents freely available. "Standards required by law that are not freely available are a violation of a key principle of the rule of law: that the law must be promulgated to be valid. This is particularly true for public safety standards. ISO will lose relevance in our modern world if they do not change their approach to this important issue. They are on the wrong side of justice with their current stance."
"This effort by so many senior ISO participants to make sure their important work remains relevant is very commendable," he added. "I hope they will have some impact on the organization."
Sneyers said he has no immediate plans to present this letter, which he hopes will spur discussion of reform within the national standardization bodies. He said he has been reaching out to other ISO members who share his view about the need to make standards freely available, but he is considering opening the movement up to the public and policymakers.
"The national bodies (ANSI, DIN, etc) are where this discussion will have to be had first, because they are the only ones that can really influence ISO," he said. "I am mostly just trying to give ammunition to those who want to start that discussion." ®