APIs and Agents: a NISO training series, spring 2027
What the series is
In spring 2027 I'm teaching "APIs and Agents: MCP and WebMCP, the Next Generation for Scholarly Infrastructure," an eight-session online training series for NISO.
From NISO's description of the series:
The arrival of AI agents has not made APIs obsolete—it has added a second class of consumer to the interfaces our organizations already maintain. This series treats agent access as an extension of API strategy rather than a replacement for it, working from the standards participants know toward the ones now forming, including MCP and WebMCP, COUNTER's optional Agent access method, rights signaling, and trust markers.
The series meets on eight Tuesdays, March 30 to May 18, 2027, for 90 minutes each, online over Zoom. Each session is recorded for registrants. Times and the full schedule are on NISO's page.
Who it is for
NISO describes the series as "ideal for technology professionals, systems librarians, publishers, product managers, and organizational leaders responsible for APIs, digital infrastructure, interoperability, and emerging AI applications."
Why I'm teaching it
For two decades I have taught and built with web APIs, beginning at the UC Berkeley School of Information, where I taught courses on APIs, mashups, and working with open data using Python.
I co-wrote two reports on how the US government can improve its efforts to make data and services available through APIs.
I served on the Zotero Project's initial advisory board and as its Integration Advisor, and managed the Zotero Commons collaboration between George Mason University and the Internet Archive.
I'm author of Pro Web 2.0 Mashups: Remixing Data and Web Services (Apress) and was a contributing writer for ProgrammableWeb.
I've been an invited speaker at the Library of Congress, O'Reilly Emerging Technology Conference, American Library Association, Code4lib, Educause, and other venues.
Much of my current work is in scholarly infrastructure. Most recently I was a technical contributor to iSamples, an NSF-funded system for describing and finding physical research samples, where I led the technical close-out: archiving the final data with permanent identifiers and handing the system over. I'm also modernizing unglue.it, the Free Ebook Foundation's ebook platform, using AI-assisted development.
The question running through all of this is how people and programs get at the data and services an organization offers. AI agents put the same question to the same interfaces, with a new kind of user on the other end.
More of my background is on the About page.