Asolid brick of paper landed on my desk with an audible thunk. It was one article, just one, that my adviser wanted me to read — over sixty pages of densely packed mathematics written in German. Oh, and not just German, but the sort of arcanely worded German that prompted Google Translate to throw up its hands and say, “You’re on your own on this one, bub.”

But after the initial language shock, what struck me most about this article was how little space it devoted to its audience. Despite its length, the abstract and introduction — the explanation of why these results mattered — was scrunched into half of the first page. A review of the article (also in German, but much less arcane) remarked that the utter lack of context for the results meant the paper would generate little interest; and indeed, in the near century since it was published, only one other writer, besides myself, has cited it.

I have noticed journals even into the ‘70s and ‘80s following this same terse format: no context for results, just the results themselves. No space is wasted. As soon as one article ends, there is a bold black line, a blank space best measured in fractions of a millimeter, and then the next article begins.

But today, I hardly see this format any more. Sometime in the last few decades, the expected audience for a research article shifted from the mere dozen who might have looked at a particular issue before to a much broader selection of researchers.

Why the change?

I suspect shifting publication standards are the culprit. It is easy to forget that research journals are a business, just as much as newspaper or book publishers, and they want to publish high-quality, attention-grabbing articles to increase their sales. At some point, the value of a well-written introduction outweighed the value of cramming in one more article into a given volume.

Since the dissemination of one’s work is an important goal of any researcher, the publication standards of journals can have a massive impact not only on what researchers write down, but what research they actually do in the first place.

I have spoken with scientists who lament that journals focus so much on publishing bright, shiny new results. Discovery is only one part of scientific progress, after all. We also need to retest results and confirm that they work. This work, while necessary, is not terribly prestigious, and so these results are often overlooked by major journals. Even papers that show flaws in previous work, a vital duty in science, often get skipped.

In mathematics, too, we often only write down and publish the one method that worked. All the methods that failed, all the ideas that petered out without results — they never find their way into print. That is just fine if all the reader needs to know is the truth of some statement. It is awful if the reader wants to expand, refine or improve upon the results of a previous paper: Researchers can spend most of their time rediscovering the same dead ends that the previous writer did.

The company Elsevier provided one of the more worrisome reminders that publishers are a business. The methods they used to boost their profits earned such widespread condemnation that nearly 13,000 researchers boycotted them, refusing to review for, edit for or publish in any of their journals.

But, once again, dear reader, the Internet is changing everything.

Already there are several research journals published completely online. Some even have high-profile editors and a bit of prestige to their name. Most printed journals I see today also have all their articles available for download online (for a subscription fee, of course).

The big difference between print journals and online journals is the associated cost of running them. There is a massive difference between a 20-page and a 60-page paper in print; not so much difference between a 20 kb and a 60 kb PDF file online. The biggest remaining cost is the time and effort to take articles through the peer review process.

Without the same pressure to publish the best stuff possible in the few pages allotted, online journals can take on more and more varied articles. Researchers, in turn, can take slightly greater risks as well: The maxim, “Publish or perish,” may just become, “Publish, or publish somewhere else.”

Joseph is a graduate student in mathematics. He can be reached at [email protected].