Terminology Is the Starting Point
“So difficult it is to show the various meanings and imperfections of words when we have nothing else but words to do it with.”
—John Locke
Technologies change how we work in and with the world around us. These advances have radically transformed how the information community functions. And with each advance in technology, new words, terms, or uses of old words are introduced. Many of the terms we use to describe our interactions, such as Wi-Fi, megabytes, podcasts, insta posts, bot, doomscrolling, blog, or meme, wouldn’t have made sense to a person living a century ago. Emojis and text speak have also developed to speed and simplify communication. Beyond slang and common usage, how technologies function requires specialized terms and definitions that must be agreed across parties, be they trading partners or communications systems protocols. This is where the standards process comes in.
Terminology is often the most difficult part of any standardization effort. The precise meaning of words used to describe how a process works, how a product should be built, or how a thing should be measured is critical to turning a document into a functional application. Terminology is the bedrock upon which all other things are built. We can’t fit the pieces together if we don’t first know how to describe what the pieces are. The finer points of what distinguishes one thing from another can often take us into the lands of philosophy and epistemology. But the classification of a very precise set of things and how they interact with other objects or are counted is among the most impactful decisions. It is the foundation of all the subsequent choices and affects how projects develop.
NISO has several descriptive standards that define terminology, from the Z39.7 Data Dictionary that describes the metrics by for assessing a library’s performance to the Standard Terminology for Peer Review, the Contributor Roles Taxonomy (CRediT), and the Journal Article Tag Suite, among others. Beyond simply defining terminologies, NISO even maintains standards on how controlled vocabularies are created and structured. At the international level, there is also work at ISO on how thesauri are developed and their interoperability with other vocabularies (ISO 25964). There is an entire ISO Technical Committee focused on language and terminology, along with groups that coordinate terminology maintenance across the information and documentation technical committee of ISO (TC46).
As new domains or product types coalesce, agreement upon definitions of terminology is one of the first things to establish. For example, last month at the ISO TC 46 plenary meeting in Delft, the Standards Administration of China (SAC) put forward an idea for an international standard describing the elements that should go into data papers. These relatively new forms of publication don’t include traditional narratives, but are more aptly described as publication wrappers for a data set, providing key interoperability and provenance information. Such data papers have only really been circulating for about fifteen years. Data sets had long been included as supplementary materials in publications, and their use grew significantly with online publication, leading to the 2013 publication of a NISO recommended practice on how these materials should be managed. In 2009, the International Journal of Robotics Research included two data papers in a special issue, launching their inclusion as a publication type for that journal. The number of publications accepting data papers grew slowly from 2009 to 2020, but accelerated from 56 in 2020 to 140 in 2023, according to data collected by Balsa-Sanchez and Loureiro at Universidade da Coruña, and this number has likely only increased over the past two years. Given their growing presence, we should start to clearly define what they are, what they should include, and how publication processes should deal with them. The first step in this process is to establish clear, agreed-upon definitions that are better defined than “I know it when I see it.”
Last month, NISO released a report on AI applications in scholarly publishing that includes potential new areas of focus for the NISO community. Everyone is encouraged to share feedback on which efforts should be prioritized (deadline is July 15). Several of the project ideas focus on terminology and consistent descriptions of elements of the AI interactions. In this nascent space, many of the interactions of AI systems lack consistent definitions or specific methods for things like training, tracking usage, and licensing. For these systems to gain broad marketplace adoption, the basics of describing what they do and how other systems interact with them are a missing foundational element.
Similarly, another project idea described in the report is to include information on content that will be ingested into an AI, or at least training those building AI systems about the structure of scholarly content. Participants suggested that AI tool developers lacked a clear understanding of the differences and distinctions between things like peer-reviewed content, preprints, and opinion pieces. Some argued that the level of rigor of the review process and the self-correction of retraction processes should be included in some manner in the training process of AI systems. Even this requires consistent use and application of the terminology around retractions and peer review. While NISO’s work on the Communication of Retractions and Expressions of Concern (CREC) and the Peer Review Terminology both provide a useful basis for greater understanding, work remains to be done around applying these terminologies. If the publishing community isn’t clear about what constitutes a peer-reviewed item or a retraction, how can we expect others to be clear in their understanding when they ingest data about them into their own systems?
Certainly, these and other topics will be fertile ground for discussion during the NISO Plus Global/Online event in September. We’ve received several dozen proposals for sessions and will announce the full program later this month. A preliminary glance at the topics tells me there will be lively conversations about how these issues will affect our community. Please save the dates: September 16–17. You can also register now!
In the meantime, I hope that you all can relax over the summer (in the Northern Hemisphere) and perhaps enjoy a good book.
Sincerely,
Todd A. Carpenter
