Web metadata standards: observations and prescriptions

The World Wide Web has spawned numerous standards initiatives that aim to facilitate more powerful and interoperable functionality based on text exchange, but beyond mere Web page transfers. Software can take a Web page's data as input to further value-added processing, such as filtering items of interest, comparison shopping, finding potential business partners, and executing transactions. But software can do this only if the page's meaning is formalized and explicit. We review various Web content metadata standards and offer observations on their development efforts. We're motivated both by the sheer number of standards and a concern that in our haste to advance these standards and their promised functionality, we might overlook key lessons learned in various disciplines, including software engineering, software reuse, and library science. We call particular attention to the apparent confluence of standards development and artificial intelligence, which raises additional possibilities and concerns.