reposted from my New Media Blog: This afternoon, in our New Media Seminar, we had the bizarre experience of following the Twitter stream of Marshall McLuhan, whose writings we were discussing. McLuhan passed away some time ago so apparently he was tweeting from the grave. I struggled a bit to comprehend McLuhan’s message that the “medium was the message.” How could the format, the vehicle, be more important than the content?
Regarding contemporary science, content is everything. More specifically, data are everything. Someone might articulate the most elegant and compelling theories regarding the most significant unknowns in the natural world, but until evidence in the form of reproducible data are generated to support those theories, little attention is given to these thinkers. Rare is the scientist-philosopher, and that is a shame.
How do scientists communicate their content, their data? Through the 20th century, “serious” scientific content was communicated in print, specifically, in peer-reviewed journals. The format of a scientific journal article in my field is fairly rigid: Abstract (150-300 words, depending on the journal), Introduction (literature review), Materials and Methods, Results, Discussion, References. Strict page or word limitations exist for most journals. The “meat” of the scientific article is the Results section, conveyed through a series of figures and tables (usually 5-10 per article). Each figure possesses a descriptive legend of a few sentences. The Results section also contains a narrative “walk” through the data in prose, with references to the figures and table.
The Discussion section that follows should provide deeper interpretation of the data and make connections to other published works in the field. For the most part, speculations that extend beyond the scope of the data are not permitted. Philosophizing is prohibited entirely.
There is little room for literary creativity in scientific articles. I once worked in the same department as a brilliant developmental biologist who submitted a research article with the Abstract written as a limerick. All of the requisite information was present, the abstract fell within the 200-word limit, and it began with “there once was an embryo from…” The journal refused to publish his article.
Scientific articles are written, submitted to the journal of choice and then reviewed by two or three scientists selected by the editors of the journal. The reviewers are kept anonymous. Depending on the opinions of the reviewers (the jury) and the final decision of the editor (the judge) the article will be accepted as is (rare), accepted with modifications, or rejected entirely.
How has scientific communication changed in the era of electronic media? Most journals are published both in electronic and print format. Some are published entirely on-line. Nowadays, the submission and review process is conducted electronically, which has accelerated the time from submission to decision from a matter of months to a matter of weeks. Beyond these efficiencies, the medium has not changed the message very significantly.
The switch to electronic journal format has not really altered the kinds of data submitted, with the exception of videos, which can provide compelling records of temporal experiments. Undoubtedly, each article typically contains more data, since page charges are no longer a restriction. However, much of these data are relegated to “Supplementary Materials” accessed only as a link from the main page. Perhaps the greatest change is in the references, which are now hot links to the article cited, accessible provided the linked article is published in an open access journal or the reader works for an institution with a paid subscription to that journal. This practice of linking journal-to-journal is a bit reminiscent of Bush’s Memex with the added dimensions of the World Wide Web.
With these exceptions, the electronic journals have adopted the conventions of the print medium: Abstract-Introduction-Methods-Results-Discussion-Reference. Reviewers remain anonymous, even more so, because in the old days, it was sometimes possible to determine the reviewer from the handwriting, and a few bold souls would proudly sign their reviews.
And most significantly, once an article is published, it becomes a fait accompli. The work becomes a defined and static entity, a unit known as a “pub”. This is where the greatest opportunities are missed, I believe. Once a scientific article is published, even in electronic format, there is no simple mechanism for amendment to that body of work. If the scientists who wrote the article discover that they made a mistake, they must submit a formal Erratum, which remains a real mark of shame in the scientific community. “You lazy bum – why weren’t you more careful in the first place.” It would be so simple and seems so logical in the current medium for scientists to be able to amend their own work, to post additional data, whether confirming or contradictory to the original conclusions, in order to continue a line of inquiry.
And the process of linking one data set to another remains hindered by the packaging of bodies of work into the conventions of an article. When one published article is linked to the referenced article therein, the link is to the other article in its entirely, not to the specific relevant figure or statement from that work.
In writing many of my own articles, I have wanted to compare my own results to a particular data set from another article. There is no convenient way to link to that information, even though it is readily available. Part of the problem is copyright, but equally challenging is the mindset that the body of work as originally published must be kept intact. Even with proper referencing and permission, it is simply not considered acceptable form to deconstruct a scientific publication in order to make new meaning.
Some journals have added moderated comment threads to the on-line version of the articles. I haven’t seen much dialog generated from these, but I suspect there are lively discussions on some of the most active and disputed areas of inquiry. Hopefully, this will be a step in the direction of a more free-flowing discussion and exchange of scientific ideas.
However, I doubt that many scientists will post unpublished data in such a forum for fear of being scooped. As long as (academic) scientists are judged by the number or publications and the impact factors of the journals in which they publish, scientific dialog will largely be restricted to the current norms.
However, if the medium is the message, then change seems inevitable. Much of that change is probably underway, but I lack the vantage point to see the forest for the trees that surround me. I look forward to the time in my life where I can appreciate these changes in scientific communication. The nature of the current medium seems to almost guarantee that the practice of science will be more collaborative and more inclusive than the environment in which I was trained. My students will be working in that new realm, and I pledge to do all that I can to help them prepare for what could well be another revolution in science.