James Williams


At Carleton College, I studied math and physics and learned how scientists think. And I longed for my California home. At U.C. Berkeley, I lived two blocks from the Oakland Black Panther headquarters and learned about blacks' problems with cops. And I got a Ph.D. in mathematics.

At Berkeley and later at BGSU and the MITRE Corporation, I wrote research papers. Back then, the rules for writing reference lists were informal and reasonably simple.

After retiring, I became curious why my doctors thought medical marijuana could help with my prostate cancer. I learned that CBD, THC, and a dozen other cannabis compounds are relevant to serious illness. I went after the details.

PubMed is now posting thousands of cannabis articles per year. My reference lists have grown to more than a thousand entries. Meanwhile, the requirements for writing reference lists are growing like Topsy. The National Library of Medicine guide includes 25 chapters, not counting five appendices on the art of abbreviation.

In writing my citations and references, I tried out an approach that involved writing raw citations and references, which I then transformed to numeric citations and references by hand. I thought, why not automate the transformation?

And while I'm at it, why not use a reference style suited to my medical marijuana audience? But which style—NLM, APA, MLA, Chicago, other? As far as I can tell, none of these styles are supported by needs-based research. We who venerate science are using reference styles that are not backed by science. Moreover, all of these styles reflect ingrained traditions that predate the Internet. The web is becoming our medium of choice, yet we still use styles created for an ink-and-paper world. What styles would result from doing needs-based research?

Today, I have done the research, shown that author needs differ from reader needs, designed separate styles for authors and readers, and written the transformation from the author style to the reader style.