The Impending Death of Scientific Journals
by
If Copernicus had a website – assuming “www.earthrevolvesaroundthesun.com” was still available – it probably would have been much easier to spread the word about his novel theory. Instead, he depended on a publisher to reprint and distribute the work. For centuries, scientists have relied on publishers and scientific journals to make public their findings and until about fifteen years ago, it was hard to imagine that system might ever change. But we’re in the 21st century now – an era that promises interplanetary travel, holographic tv, personal jetpacks, and an end to Dancing with the Stars. In such a fast-paced, technologically-savvy society can traditional publishing mechanisms really survive or are we about to witness death of scientific journals?
We’re taught the Central Dogma is DNA –> RNA –> Protein, but most scientists recognize that the true dogma is closer to DNA –> RNA –> Protein –> Publish. A researcher’s publication record and resulting impact factor directly impacts hiring, compensation and funding decisions and thus, is an integral component of research as most of us know it. However, scientists have acknowledged that impact factors are flawed and may be manipulated by journals to artificially prop up their importance. In fact, in recent years alternative measures of author performance such as the H-index have emerged in an attempt to more accurately describe researcher productivity, although these methods are not immune to criticism.
Having read this far many researchers will experience an uncontrollable, instinctual response to the idea that journals could ever disappear – “not possible!” they say, “we need journals”. Did we need cassette tapes to play and share music? Did we need to hold up an entire line of people by paying for groceries with a personal check? Maybe we’re wrong. Maybe journals will live forever, but there’s a certain irony in scientists – among the strongest supporters of the theory of evolution – refusing to at least consider that their own profession will evolve as well.
This is not to discount or trivialize the role journals play in the scientific process. On the contrary, journals have been the primary mechanism of reviewing, validating and distributing results for centuries. [Full disclosure: my own father is the founding editor of a scientific journal, so it’s not that we’re cheering the death of these publications.] The idea is that using modern technology, these functions may be achieved through alternate – and arguably more efficient – means.
So what would life without journals look like? Consider how modern technologies might pick up the slack.
.
Distribution
Believe it or not – for those of you under 30 – there used to be a time when looking up articles on a specific topic meant going to the library (gasp!), pulling out a 30-pound chemical abstract index and guessing at relevant keywords for your topic. After finding the keyword, you’d be directed to any number of references that may or may not be of interest, but you’d only know after looking up every single one. This usually meant risking getting squashed among the rolling towers of journal archives in the library’s basement. Similarly, staying current in a field meant wandering over to the library and scanning the relevant print journals – often a couple of general and a couple of specialized journals – for a couple of hours.
Good luck with that approach today. The ever-expanding number of journals has made it nearly impossible to stay on top of the literature with over 50% of scientists admitting they read “zero” journals, opting instead to follow specific research topics. In fact, many subscribe to services that provide regular keyword searches of the literature and notify them if anything has been published (perhaps a future role for journals as advanced content aggregators?). The point is, we’ve already moved to searching for research by topic, author or keyword not necessarily by browsing intact journals.
If you’ve got a website, you’re a publisher. And these days, you’ll be hard-pressed to find a lab that doesn’t fit this bill. In fact, since access to most journals is subscription-based, a freely-accessible lab webpage actually presents an opportunity to increase visibility among a wider audience. Like with journal publications, work published on your own website could be delayed until any necessary legal issues (patents, etc.) were resolved, or until you felt you had a significant head start on the follow-up experiments if you’re worried about competition. Website publications could be time-stamped like enotebooks so it would be clear who was the first to a new discovery. So what’s the difference between staying current in the literature by setting up a keyword search to cherry pick from journals vs. scientists’ own websites?
For a moment, close your eyes and imagine a world without file type errors (do they accept .tif or .png files?) or formatting (and reformatting) issues. No wasted time searching for 6 words to remove to make the cutoff, no image resolution issues when converting between three different file formats. Ah, feels good doesn’t it?
.
Review
Scientific research and publication is built upon peer review and journals have been the primary mechanism of maintaining this standard. While our peer review system isn’t perfect, it’s been the only logistical solution available to us since it’s clearly been impossible to imagine sending the work out to everyone in a field for a review prior to publication. But in it’s purest form, wouldn’t peer review include feedback from all of your colleagues in a given field – not just two or three? Twenty years ago, the prospect of gathering widespread sentiment on an article was daunting. Now it happens on the web every day. Some will argue that the reviewers play an important role in making sure the story is complete. However, there will always be more work that someone could suggest, so appeasing two reviewers doesn’t mean the work is finished. Just ask any graduate student who participates in a journal club how easy it is to rip apart a published, peer-reviewed piece of work.
But in the absence of peer review how would we know the value of the work, particularly if it weren’t in our area of expertise? Countless web-based measures of user sentiment and interaction have been developed in the past decade. Actions such as rating articles, leaving comments, tracking downloads, counting bookmarks and quantifying social sharing are all valid, real-time metrics of user response to content.
.
Cost
OK, now close your eyes again and imagine what it would be like to publish an article without spending thousands of dollars in publication costs. What else could that money be used for – equipment, reagents, computers? Of course, uploading the results to your webpage would take time, but compared to the amount of time that goes into preparing and publishing a paper in a traditional journal, it would be far less. Most people have stopped ordering reprints of their papers, so there wouldn’t be significant printing costs associated either. Interested parties could download or request a pdf directly from your site.
The idea here is not to imagine that one day all journals become open access – it’s to consider what research would look like with no scientific journals, when scientists’ lab webpages serve as the publication mechanism. The Internet has changed almost every facet of our lives so when it comes to our own profession we shouldn’t be too quick to assume evolution won’t apply to us too.
.
[poll id=”88″]
.
.
Have any ideas ideas how scientific publishing may look in the future?
.
.
Katie_PhD
wrote on October 11, 2011 at 2:31 pm
I voted for under 20 years, but I'm probably being optimistic. My biggest gripe with the current system is peer review. As you said, there is always more to do, but often a reviewer will suggest an experiment outside the scope of the paper, or that is prohibitively expensive for a small lab, and regardless of a well worded rebuttal the editor will go with the reviewer's opinion. It can be incredibly frustrating, and hold up publication by not months but YEARS. In a world where we can publish what we ate for breakfast, or post a picture of the mess our kitteh made, in a second (on facebook or twitter), it's pretty shameful that we can't share scientific information more efficiently.
Dangerous Bill
wrote on October 15, 2011 at 5:52 pm
You can always debate peer review with the editor. In general, the editor tends to side with the author, since anonymous peer review, whatever its advantages, invites sniping and obstruction by rivals. Even so, it's the only realistic method of quality control.
Publication followed by commentary means that the paper stays in the literature, available for citation, regardless of the number of negative comments. I suspect that those online sites that implement a conventional and time-tested peer review mechanism will have the greatest acceptance among readers.
EyeLab
wrote on October 11, 2011 at 4:30 pm
If we lose the journals, what happens to peer review?
Katie_PhD
wrote on October 13, 2011 at 9:54 am
My guess/hope is that people would use post-publication peer review (as is already in place on sites such as Faculty of 1000) as a guide. In this way, papers viewed favorably by their field would have far more good reviews than bad, and vice versa. Peer review would not go away, it would actually become a much more open and thorough process.
@FLOSciences
wrote on October 11, 2011 at 8:53 pm
Yeah the process is flawed, but I don't think the solution is to scrap peer review. Instead, we need reform. The idea that I could open publish what I ate for dinner and that carries the same degree of scrutiny as a Nobel Prize winning experiment just doesn't sit well with me.
If you could open publish anything then you would see so many trivial or incremental "papers". As it is, we're overwhelmed with all the information. And to be fair I see so many papers today and think "I could have published for THAT!?"
Again, I think we'll see, or that we need, reform, not a scrapping of traditional journals/peer review.
Anthony Salvagno
wrote on October 12, 2011 at 10:40 am
"The idea that I could open publish what I ate for dinner and that carries the same degree of scrutiny as a Nobel Prize winning experiment just doesn't sit well with me." Sure this is true, but publishing what you ate for dinner will get you scrutinized a whole lot more than the Nobel Prize winning experiment, which shows you the system works. Either that or it won't even get noticed.
The idea that a paper is trivial is silly. If you realized you could have gotten published for a simple idea, then why didn't you do it? Something like that doesn't have to be impactful, it just has to be useful and accessible.
Right now I'm repeating an experiment from the 1960. I don't think it is a trivial experiment because there is a lot more available to me now then there was back then and the topic hasn't been discussed in over 40 years! Obviously I don't think it is going to be the next "Neutrinos" experiment, but I think it has purpose.
As for the scrapping of traditional journals, I don't think anyone said to just throw it all away. We do need to transition though. At first this will be a reform, but as everything goes electronic I think people will realize that publishing on your lab website will be very similar to publishing on the website of a journal, just more cost effective and things may start to migrate that way. But like the poll suggested, that may not be for another 10-20 years. It all lies in the hands of the new generation of scientists who grew up with Facebook, Twitter, and Google and have no recollection of going to the library.
Scott
wrote on October 12, 2011 at 8:55 am
It seems to me that in fast-moving fields, a researcher would need to know about the most important research BEFORE his or her competitors–or at least, at the same time. Rather than waiting for the crowd to vote (because if you wait for that, you're already behind). The best way for that to happen, still, is to trust the human judgment of scientific journal editors, in my opinion.
Plus there still is the issue that funding decisions are made by people responsible for multiple disciplines, who therefore can't keep up with all the disciplines, and must rely on the judgments of impartial arbiters (i.e., scientific journal editors) in making funding decisions.
Given this, I would suggest, journal publishers must recognize this as their fundamental mission…
NB full disclosure: I work for Springer, but the opinions above are strictly my own and I am not speaking for Springer.
Anthony Salvagno
wrote on October 12, 2011 at 10:30 am
I'm all for this (obviously being an open notebook scientist). We already use search engines to find papers, which makes storing them in journals obsolete. Blogging formats (posts with comments) allows peer review to still be relevant. The added benefit of that is that reviewers won't be anonymous so they wouldn't be able to just say anything to prevent your paper from being published. Eventually we will have a facebook for science which will expedite the process of bringing labs together which could deliver meaningful peer review. Finally, I believe that impact factor is a thing of the past. Information moves so quickly now that any future publication can be impactful, it just has to be found.
I eventually think that publication will work more like facebook or friendfeed. You will publish something on your lab website which gets linked in the facebook of science and all the labs that follow will be notified and read and scrutinize and celebrate. That thought makes me happy!
Nick Harris
wrote on October 12, 2011 at 11:45 pm
My main qualm with your argument lies with the replacement of peer review. I wholeheartedly agree that our system is flawed. The opinion of two reviewers means nothing in the grand scheme of things, and the art of peer review has been lost in our modern world. However, the answer does not lie within Facebook "likes" and other options based on similarly guided principles. There needs to be some sort of regulated process by which the opinion of those whose voice is valid does not get drowned out by the masses.
A side note, too: Just because only 50% of scientists do not read entire journals anymore, does not mean that it is not worth it. Just because our world, of quite the short attention span, does something a majority of the time does not make it the right thing to do.
S.Pelech-Kinexus
wrote on October 14, 2011 at 3:36 pm
At Kinexus Bioinformatics Corporation, we plan to launch by the end of this year Kinetica Online, an open-access, meta-website dedicated to supporting cell signaling research. Kinetica Online will feature original research articles from our scientists and collaborators, blogs, experimental databases, knowledgebases, educational materials including animations and videos, and links to hundreds of other resource websites and blog sites. As an interactive website, viewers will be able to add their own feedback to the articles and blogs on Kinetica Online. I fully agree with Alan Marnett that this type of website could become increasingly popular for dissemination and storage of scientific ideas and ideas.
For laboratories in academic settings, it should become feasible for the universities to host open-access websites that allows for the dissemination of results from their faculties. Many university-based presses already exist for publishing books and journals. With PubMed and Google searches, it is actually no longer necessary that website-based journals are thematic for a particular field or discipline. Post publication peer-review from the scientific community will ensure that mostly high quality publication will still be produced, as it could be very embarrassing for authors to have their findings openly criticized at the end of their manuscript if the work is poorly performed or presented.
Morten
wrote on October 30, 2011 at 6:17 am
Well journals could certainly live a lot longer if they started publishing in epub or a similar e-reader format.
Mark Hoemmen
wrote on November 8, 2011 at 3:50 pm
I think your article confuses incidentals of journal publication (like file format incompatibilities and expensive publication costs) with essentials. The essential feature of journal publication is peer review, not by random internet readers, but by a group of trusted reviewers who defines the journal’s “canon.” Whether a “journal article” is a piece of vellum with calligraphy on it, a blog post, a multimedia presentation, or some medium we haven’t invented yet, publication is nothing more or less than gaining the approval of those reviewers.
Of course, the incidentals are of major concern: journals are too expensive, the publication process takes too long, their peer review processes aren’t as fair as they could be, etc.
Dana Roth
wrote on November 10, 2011 at 3:49 pm
I think it is a serious mistake to generalize about journals without making a distinction between high quality society published journals and commercially published journals. There is a difference and I doubt that high quality society journals will be easily replaced.
Stewart Lyman
wrote on December 7, 2011 at 1:57 pm
I have proposed a new concept for the publishing of science papers called iPubSci. The basic idea is to blend together PubMed with iTunes, so that papers can be purchased on an individual basis. The approach solves three current problems that plague accessing the scientific literature: iPubSci will be easy to use, legal, and affordable. A link to the paper describing the concept in detail can be found on the iPubSci homepage at http://www.ipubsci.com.
Jefferson Brown
wrote on July 12, 2012 at 12:40 pm
I think the biggest problem with the journal model that we use is the outrageous intellectual property issue – why on Earth do I have to pay someone else a thousand dollars to give them essentially exclusive rights to intellectual property that I have slaved for countless hours for and used huge amounts of grant money that I agonized to get? Yes, the peer review system they have in place is nice…but it seems like there should be an alternative way to get the same thing done. Perhaps something like a subdivision of NSF or NIH that you pay to have your paper arbitrated (similar to a journal), then they issue some sort of electronic certification to you if it passes, which can be posted on your website. That way, you can get both the establishment-type review and the wiki-type review on your personal site if you so choose. And you keep your publication rights. Thoughts?
alan@benchfly
wrote on July 20, 2012 at 12:21 pm
This is a very cool idea- peer review for hire with a certification badge or something posted alongside the data. You could also imagine the lab itself (not just each individual paper) having a rating based on the quality of their publications over the last X years. Moving the impact factor to the lab side of things, not the journal side.