In 1992 it was widely held that the internet was for academics and nerds, with no use case for ordinary people.

Judged by the historical record of 1992. True: contemporaneous sources (press coverage, industry commentary, histories of the internet) show the dominant view in 1992 was that the internet was mainly for academic, research and technical users and had little or no relevance to ordinary consumers. Mostly true: that view was dominant, but a notable minority saw consumer uses, or consumer online services (CompuServe, Prodigy, AOL) were already popular, so 'no use case' overstates a mainstream view. Mixed: opinion was clearly divided. Mostly false: widespread contemporaneous expectations of consumer internet use existed and the claim overstates a minority view. False: the consensus was clearly that the internet had broad consumer potential.

Set before the research began.

1. E1 Price War Erupts Among On-Line Computer Services Los Angeles Times · 21 April 1993

   “But the question that has plagued the developers of all this--whether consumers will care to use them--has proved a sticking point for the on-line industry as well.”

2. E2 Appraising the Value of the Internet for Ministry Ordained Servant (Orthodox Presbyterian Church) · 3 June 1996

   “By 1992, the total number of users, however, for all these combined services was less than 2 million (less than 1% of the U. S. population).”

3. E3 NSFNET — The Decade in Which a Research Network Became the Public Internet ITlibra history site

   “NSFNET Backbone services are provided to support open research and education in and among US research and instructional institutions, plus research arms of for-profit firms when engaged in open scholarly communication and research.”

4. E4 [PDF] The Market Prospects for Consumer Online Services in Europe. ERIC (U.S. Dept. of Education), from Online Information 96 proceedings · 1996

   “Certainly the Internet was widely used in the USA from 1992 onwards, but the proprietary interfaces that the consumer services offered were much easier to use than Mosaic and the other browser applications available at that time.”

E3 shows the NSFNET backbone's services were provided to support open research and education at research and teaching institutions, which is consistent with the view that the internet was mainly for academic and technical users. Consumer online services existed by 1992 but had fewer than 2 million users combined (E2), their proprietary interfaces were easier to use than early browsers (E4), and the industry was still unsure in 1993 whether consumers would use them (E1). This fits the criterion's "mostly true" case: the academic view was dominant, but consumer services already existed, so "no use case" overstates it. None of the sources is from 1992 or directly records opinion at the time, and E4 notes the internet was widely used in the USA from 1992 onwards, so the support is indirect.