The Dayside : Where's my flying journal?

By: Physics Today
13 June 2014

Earlier this year, the International Association of Scientific, Technical, and Medical Publishers (STM for short) convened in Washington, DC, for its annual spring meeting. My friend Terry Hulbert helped to put together the program. When he returned to the UK, where he's based, he wrote an engaging and concise summary of the proceedings.

One of the speakers at the STM meeting was Clifford Lynch, a computer scientist who heads the Coalition for Networked Information. Terry reported that

Lynch of CNI made the provocative, although some would say accurate, comment that publishers have not yet achieved much in the digital age. Content has been digitised and reference and citation pathways have been created, but, despite this, the process and output of science would still be recognisable to a scientist from 100 years ago.

If the American Institute of Physics, which publishes Physics Today, and other academic publishers have failed to deliver the publishing equivalent of the flying car, it's not through lack of effort or imagination. Online-only journals, open-access journals, single-article metrics, semantic enrichment, and interactive PDFs are among the innovations of the last two decades. Other new features and products are surely on their way.

Imaginary flying cars have appeared on the cover of Popular Mechanics several times in the magazine's 112-year history. CREDIT: Popular Mechanics

Imaginary flying cars have appeared on the cover of Popular Mechanics several times in the magazine's 112-year history. CREDIT: Popular Mechanics.

But it's conceivable that the process and output of science, as Terry put it, looks essentially the same to us as it did to Lord Rayleigh because the peer-reviewed scholarly paper cannot be surpassed as the most effective way for scientists to communicate the results of their investigations to each other.

But is that the case?

Before considering how the peer-reviewed scholarly paper might be reinvented, let's look at what the current version does. A paper in, say, Chaos begins with an introduction that reviews the problem at hand and explains its context and importance. The main section describes the investigation by means of words, mathematics, and figures. The conclusion summarizes the results of the investigation and their implications.

I don't see how a scientific paper could take any other form. Granted, videos, comments, hyperlinks, and other digital affordances enhance a paper, but they do not transform it. Where I think the scientific paper can be improved is in how it's written and read.

Most scientists read papers in two ways. For papers in their immediate field that report what might be called new facts, scientists simply skim the papers and note the new information. There's no pressing need to reinvent such papers.

But some papers contain significant advances that break new ground. Diligent scientists can't afford to skim those landmark papers. They have to digest and understand them. Two barriers potentially impede that process.

First, despite the work of journal editors and referees, some papers are poorly written—by which I mean the facts and concepts of the paper are neither well organized nor well described.

The Write Brothers' Dramatica software takes users through the steps of developing a screenplay's plot and characters.

The Write Brothers' Dramatica software takes users through the steps of developing a screenplay's plot and characters.

The second barrier arises when readers need more information than the author supplies. For instance, readers might want a bit of help with a mathematical derivation or they might want more details about an experimental step. Some journals provide that information in online supplemental material, but most do not.

To obviate the two barriers to understanding, I envision a new software tool that helps authors write better papers. The tool would guide the author through filling out the title, coauthors, references, and other standard features of a paper. But it would also analyze prose as it's written and prompt the author both to improve it and to supply additional derivations and details; in the final paper, they would appear as hotlinked popups or sidebars.

Far-fetched and ambitious? Possibly. But artificial intelligence is advancing rapidly. Software that can grade students' essays is already in use. And if you want to write a Hollywood screenplay, a company called Write Brothers markets software that can guide you through the process of developing characters and a plot.


Submit comment
Comment moderation successfully completed
This is a required field
Please enter a valid email address

Oops! This section, does not exist...

Use the links on this page to find existing content.

632fc4064c2120686ad139879e7bbc09 weblog.blogpostsparsezxybnytfddd
Scitation: The Dayside: Where's my flying journal?