Newsline, October 2018

Letter from the Executive Director

Standards can take many forms.  Last week, I was at an SSP educational program in Washington.  During the meeting, Craig Van Dyke from CLOCKSS spoke about the development of ORCID (a recently-joined Voting Member of NISO, so "Welcome!").  He described the formation, goals, and underpinnings of ORCID as a standard. In this description, Craig specifically mentioned the difference between "S"tandards and "s"tandards, noting the importance of the capital letter at the beginning of the word. I have also made the same distinction in my own presentations and writings over the years. Obviously, there is a difference between the two approaches to consensus and I don't want to minimize the value of a formal process.  These values include the right to review and comment on drafts, an open process that is balanced among stakeholders, documented development procedures, the right to appeal, gathering input from external perspectives, and a structure for periodic review and maintenance.  The "Capital 'S'" standards have stability, weight, and recognition, having gone through this process.

But in practical terms, does circumventing the formal "Standards" process make the resulting ORCID system, or NISO recommendations such as KBARTODI, or our work on privacy principles any less valuable or impactful?  Of course not. These systems and structures are valuable because they are adopted, they are used, and they are built into an organization's workflows. Many formal capital "S" standards began as de facto standards, such as what is now JATS and the Dublin Core Metadata elements within NISO, or the PDF and Office file formats that were formalized elsewhere.

The process of developing these and other de facto standards is one key element that differs from the formal standards development process. Although they sought input from external stakeholders in the process, it was ultimately the National Library of Medicine (NLM) that took decisions about how the NLM DTD should work and what elements it should include or exclude that eventually became JATS.  NLM took those decisions based on their own needs, systems development, and goals. It wasn't a concern for them how, as an example, comic book publishers wanted to use or engage with the NLM DTD. The National Library of Medicine was happy to allow others to use the structures and adapt them for their own purposes, but did not want to or feel the need to be involved.  Of course, opening up the NLM DTD to the outside community, unburdening the agency from the problems of maintaining the structure, and making JATS less domain specific (as the DTD was) were all motivators for NLM to want to turn JATS into a formal standard.  Similarly, there were benefits to Adobe and Microsoft to opening up PDF and Office file formats to standardization.

Interestingly, as NISO has engaged both in formal and less formal ANSI-accredited standards, NISO Recommended Practice consensus development processes have taken a stronger role in our community.  For example, we've recently received a proposal from the JATS4R community to bring their work into the NISO portfolio as a suite of NISO Recommended Practices.  This proposal is currently being considered by the Information Creation & Curation Topic Committee and, if approved, will go out for voting member approval as a project for consideration. The JATS4R recommendations provide implementation support and guidance for the JATS standard to support reusability and interoperability.  Having JATS4R and the JATS standard published by NISO should reinforce both projects and tie them more closely together. It is also a perfect illustration of the value of both normative standards and the industry guidance that NISO provides.  In this particular case, NISO will be providing direction -- the "Capital 'S' standard -- and guidance -- the "lower case 's' standard, a recommended practice.  The two are tied together, but serve different roles and for different purposes.

In practice, as I said in a tweet at the time, the difference between the two is more about the process and the approach than the output the majority of the time. It is also important to note that regardless of the formality of the process, if the result isn't adopted, the standard isn't as valuable as it could be.  And if the output is adopted broadly, the standards formality isn't necessarily the most critical question.


Todd Carpenter,
Executive Director, NISO