In chapter 1 I listed a number of researchers who are well known for their “evolutionary” approach to information (notably Marcia Bates) but I have only recently encountered Gary Marchioni’s perspective on a current issue in education for LIS, which is clearly related to research on information evolution as well: that is, the relationship between data science and information science.
In his 2023 article “Information and data sciences: Context, units of analysis, meaning, and human impact” for Data and Information Management, he writes:
“Consider how data science and information science might address the important issue of representing complex and temporally unbounded knowledge. Each of the following phrases ignite ranges of mental activity and unlimited traces of evidence to capture, store, and analyze: ‘Human history’, ‘visible light from the sun or a distant star’, ‘the thoughts you have had over the past decade’, ‘all the queries ever posed to a search engine’, ‘the collected works of Shakespeare’, ‘the philosophy of Confucius’. Each of these phrases is a compact representation for a collection of ideas or events. One distinction between data science and information science is where the scientist puts focus. The data scientist will tend to focus on the word tokens, their volume, and distributions with respect to other tokens. The information scientist will tend to focus on the relationships within the phrases and to other potentially related phrases; their groupings or classifications; and the associations and meanings among classifications of ideas, events, and concepts. Both types of scientist are interested in classification with an eye toward indexing and potentially prediction, however the data scientist focuses on discrete and thus countable/computable items and symbols (data) that lend themselves to more deterministic analysis; whereas the information scientist focuses on items with less clearly bounded items and concepts (information) that require more context for analysis and that may be subjective or more probabilistic. Essentially, both are concerned with compression or dimensionality reduction to manage or reference complexity or continuity but tend to focus on different granularities that lend themselves to different analyses” (page 4).
The study of keywords lies directly at this intersection. Dr. Marchioni’s examples above illustrate various academic search phrases, both objective (“visible light from the sun or a distant star” and subjective (“the thoughts you have had over the past decade”) but the distinction he makes also applies to the study of both commercial and popular keywords, though I would say also that there are far more “data scientists” studying these for marketing purposes than “information scientists” studying these for academic purposes, though the study of marketing keywords can be carried out using either level of granularity (and often is).
One of the few researchers who has been tirelessly working at both granularities is Dirk Lewandowski; for instance, his recent publication “Google, data voids, and the dynamics of the politics of exclusion” in Big Data & Society, in which he analyzes the effects of search engine keyword optimization on social polarization in Sweden and Germany.
Keywords In and Out of Context 