Text in a data-centric world

i first met jim gray when he was the moderator of the database subject area of arXiv, part of the expansion into computer science that arXiv initiated in 1998. Soon afterward, he was instrumental in facilitating the full-text harvest of arXiv by large-scale search engines, beginning with Google and followed by Microsoft and Yahoo!—previous robotic crawls of arXiv being overly restricted in the 1990s due to their flooding of the servers with requests. Jim understood the increasing role of text as a form of data, and the need for text to be ingestible and treatable like any other computable object. In 2005, he was involved in both arXiv and PubMed Central and expressed to me his mystification that while the two repositories served similar roles, they seemed to operate in parallel universes, not connecting in any substantive way. His vision was of a world of scholarly resources—text, databases , and any other associated materials—that were seamlessly navigable and interoperable. Many of the key open questions regarding the technological transformation of scholarly infrastructure were raised well over a decade ago, including the long-term financial model for implementing quality control, the architecture of the article of the future, and how all of the pieces will merge into an interoperable whole. While answers have remained elusive, there is reason to expect significant near-term progress on at least the latter two