Abstract In 1998, Michael Bowen summarized an ICSU Press workshop on electronic publishing [1]. It is reprinted below (in this issue of Chemistry International) and I have been invited to give a 2023 perspective on its conclusions. In 1998, Wendy Warr was chair of the IUPAC Committee on Printed and Electronic Publications (CPEP was the predecessor of the current CPCDS) and Mike Bowen was Secretary. It was an advisory function of the committee to keep-up with how the landscape of electronic publishing was evolving. Before I address those conclusions more specifically, it is worth mentioning some electronic publishing advances of the 1990s. Carnegie Mellon University advertised an opening for an “electronic librarian” as early as 1991. Some electronic products predate the World Wide Web (e.g., arXiv preprints which were first emailed using TeX in 1991). The Digital Object Identifier (DOI) was launched in October 1997. SpringerLink, Elsevier’s ScienceDirect, and PubMed were all available in 1998. The American Chemical Society (ACS) and Chemical Abstracts Service launched ChemPort in December 1997, in collaboration with seven other publishers. ACS had reinvented its journals for the Web with Articles as Soon as Possible (ASAP), search tools to find specific articles easily, and links to databases, and to cited articles through ChemPort. The scene was clearly set for electronic publishing, and so let’s move to the eight 1998 predictions from the ICSU Press workshop.