One scientist's take on three of the big journals

Randy Schekman will no longer be publishing in Science, Nature, or Cell.

http://www.theguardian.com/commentisfree/2013/dec/09/how-journals-nature-science...

He thinks that flashy scientific reporting harms science.

Robert

Looking at this issue from 'way on the outside, I can say it does really befuddle and discourage the average Joe to see science papers retracted after a huge splash.

I was particularly interested in the mention of "replication studies" where published results are repeated to acquire the same results as the original. I had recently wondered how those were funded & actually thought it would be a great job for people skilled and coming up through the educational system. I remember when reading some of Craig Venter's books he mentioned there were problems with other researchers not being able to repeat his work - until he had someone{s} from his team visit & troubleshoot. This sounded to the layman {me} like there is some protection of technique perhaps for ego or patentable monetary reasons. For example I have seen, in print, people complaining that China uses patents for engineering instructions.

It sure crushes the hopes of a lot of people and I am sure wastes a lot of resources when publicity grabbing stuff like what happened at Duke University in my state where a professor massaged some data to come up with genetic markers to use for identifying effective cancer treatments. I am surprised that the science journals mentioned would factor in a "wow" value into what they publish. I was under the impression they looked for cutting edge or Nobel quality stuff.

I like a "wow" factor and like to browse magazines like Discover etc. but had always looked to the "old" Scientific American, Science etc. to bore me silly with detail and if I understood over 80% of any article I was/am disappointed thinking that it has been "written down". ;-)

Schekman is welcome to dedicate the rest of his career to replicating studies--to be published in his very own journal, obviously. (I can't believe he had the gall to plug it!) Absolutely no problem, no conflict of interest, with THAT setup.

To me this smacks of sour grapes. The truth is that even Nobelists' labs struggle at producing a paper for Nature, Science or Cell to publish every month--I'd like to know the history of recent submissions and rejections for his lab. Incidentally, it doesn't make sense to categorise all three together. Nature and Science publish research in every branch of science, from astronomy to zoology; Cell is strictly, as the name sort of gives away, cell and molecular biology. It follows that Nature and Science are therefore going to mean even tougher competition for space, as not even Nobelists tend to be astronomers AND zoologists at the same time.

Big journals indeed seek out big news, that's what made them and keeps them big. "Fashionable" "sexy" science isn't necessarily (or ever) an empty-headed bimbo, wined and dined for her boobs. In science, things come into "fashion" for a reason, following or accompanying some breakthrough, novel methods, useful new approach, a whole new field. If it's bubbling, something interesting is going on, whether actually big, or small. It could be as simple as some imaging mode becoming a hundred times cheaper, a microscope more powerful, a new labelling molecule ten seconds more stable. Sometimes the hopes are dashed and the gold rush ends--the method/field/notion has been exhausted, or problems with it surfaced--but the more people work at them, the sooner mistakes and misconceptions, if any, are discovered and the sooner everyone moves on. Or, if it's truly a goldmine, people settle in and build towns and roads and sometimes, whole new countries. The history of science in terms of actual research can be described in a succession of such "fashions".

I want big journals to keep publishing big sexy pieces and not fucking drone work because that's how I keep abreast of what's hot & happenin' in the larger scientific world--the universe that contains astrophysics and organic chemistry and materials science and geology and paleontology, and not just my three-tatami closet. They are Andy Warhol's Interview, not the place to get morning news, the lowdown on the Joneses next door, or blow-by-blow weather predictions in our county. And only n00bs and big bosses who grace the lab with their presence once every century could imagine that ordinary scientific working stiffs somehow... I don't know, don't know this? Every month I review--look at/skim/read the contents/abstract of/ scan results/methods/print out graphics/ read partly/read in full/read again/look up references--of several hundred articles in 40-60 journals. And that's because I'm a sorry slacker with way too many extra-curricular interests.

Yes, publishing in a big journal can kick-start a career--and why on earth not? But it can't guarantee it. No one gets to rest on the measly laurels of a single big paper in science. The idea is that if you had it for this big paper, you might have it for another one. And then another one. And another. The race never ends. Some people manage, some don't (even Nobelists).

Yes, papers published in big journals are not ALWAYS flawless--there are corrections and retractions. It would be comfy to live in the world where this was never the case, but, errare humanum est. There is actually something ludicrously naive--or blisteringly dishonest--in insisting on such a record in science. All science is presented in terms of probabilities, potentialities, "to the best of our current knowledge". Reputations aren't secured by never being wrong, but by knowing how far you might be from being right.

I'm disappointed that Schekman fails to mention how the percentage of retractions in the "big" journals compares to that in others. Well--I expect he'd claim it's somehow worse for Nature to have a paper retracted, than for Biochimica Biophysica Acta. Almost as if he thought impact factors factored, eh?

Yes, the impact factor is indeed used to assess a paper's or (cumulatively) a lab's importance. But, darned if I know anyone who doesn't understand that this is only an indirect indicator of quality--it is not in itself "quality", and, most important, without context--such as anyone working in the field provides as a matter of course--you can't determine much about a paper's value from it. There are different kinds of value, local and universal.

For years, the highest citation in my sub-sub-field was held by a methods paper published in some modest mag, not even a full article. Because tons of people used it and cited it in every piece based on that precedent-founding method! If you asked the authors themselves whether they thought such a paper deserved publication in Nature, they'd laugh you out of your shorts.

I have now committed my lab to avoiding luxury journals, and I encourage others to do likewise.

Well, we must all do our modest bit for the improvement of humanity's lot. One little step for Dr. Schekman, but oh what a leap for science!

Ha.

Good points Lola. I realized after I posted I knew nothing of Dr. Shekman, thanks for the background.

Well, being a biomedical scientist myself (although way WAY lower on the recognition scale!!), I agree with Lola that the article is a bit biased. While I have long resented the absolute fawning over the Cell-Science-Nature publications, it is important to note that the "free-access" journals need to be viewed cautiously. I don't know about Randy Schekman's journal (I am sure it is legitimate), but I get approached to be on editorial boards of random free-access journals constantly - journals for which I am not remotely qualified. There is a place for both types of journals. I think Dr. Schekman's article should have included a discussion of this country's (the US) anti-progressive move towards less funding for basic biomedical science, fewer academic jobs etc. This pushes scientists to scrap for every possible dollar, publishing before they are ready. We have an extreme imbalance right now between the number of scientists, number of scientific jobs and amount of money. Until that is corrected, good science will suffer.

Of course that does not excuse true ethical violations, such as mentioned by DugsBooks. Some people will cheat no matter what. It is a sad thing.

PS. Thanks for bring this up, Mr. Durick.

#4

Hm, not sure what you mean by "background"--I know nothing about him as a person. I'm familiar with his work (not in profound, first-hand sense--it entered textbooks before I entered uni), and I have some notion of rich, high-powered labs. Schekman is old, his career was made decades ago, and he'll never have to worry about getting recognition. He's Marie Antoinette; most of us are not. His giving up cake means nothing to people living on bread.

I'd have more to say about apparent assumptions in the linked article, but I must say I dislike doing so when I'm not even sure I'm hearing the message correctly. Perhaps he simplified a lot--it's Guardian, it's mass journalism, it's "talking to the people"--but I'm genuinely puzzled by how he views what he calls "luxury journals" (never heard that before, clearly I don't mingle with the right crowd) and what he thinks abandoning them might achieve. There are bits of truth in his statements, but the general picture projected by the article is off.

It's difficult to explain this properly without going at length about the system of rewards in science, the significance of publication, even describing scientific culture. How the hierarchy is built, maintained and replenished.

First, there's a problem with the idea that Nature, Science and Cell are ruining science. I'll just leave it that, if I start digging I'll end up in Shanghai by tea-time.

Second, let's say that we agree that the infernal trio is ruining science. We follow Schekman's lead and never send them another paper--not any of us thousands or tens of thousands of labs. They die overnight. But, we still need to publish, we still WANT to publish, we want recognition, we NEED recognition, so, well... we publish elsewhere. But not all work is equal in quality and importance, so some weeding and classification are needed, lest your scientific readership drowns in the humongous ocean of information--an ocean that is already unswimmable to individuals.

So you start regulating, classifying, categorizing: this is grade A, this is grade B, this is grade C... And that's only one manner of discernment. One of the puzzling things about Schekman's manifesto is that he appears to assume people only send papers to Nature, Science and Cell, when those journals publish a minuscule portion of all published research. (And as I mentioned, Nature and Science are further diluted by including ALL of science, not a single large field.)

In reality, myriad journals are necessary, simply because science itself has grown exponentially. I know NOBODY that thinks only Schekman's "luxury" journals publish excellent science. Quite a few of my colleagues never pick up Nature or Science. Months may pass before a paper even relating to our work shows up in one of them. That's how specialised we've become--how specialised everyone has become.

So, you need to break up this gigantogargantuan mass of papers so that people can find what's useful to them. And not just by specialisation. Not all studies have the same scope and duration, or carry equal significance. It must be possible to publish a stunner on the six-headed transgenic mouse who likes Mozart as well as the tweak to the miniprep protocol that results in cleaner DNA. Both are "quality", both are interesting and valuable--in different ways, according to different needs. Mozartomouse may be the next big thing in science, but the little tweak to DNA purification may save my experiment.

So after all this... what do we end up with, if not with something... similar to what we have? A system of specialised journals choosing papers based on merit, as assessed by peer review. And how would one escape hierarchy when scientific work itself is based on discovering what matters and what doesn't, on comparative evaluation of data, on criticism?

No idea whether Schekman's journal is capable of replacing Cell, but if it does so, won't it become Cell?

#5

I think Dr. Schekman's article should have included a discussion of this country's (the US) anti-progressive move towards less funding for basic biomedical science, fewer academic jobs etc. This pushes scientists to scrap for every possible dollar, publishing before they are ready. We have an extreme imbalance right now between the number of scientists, number of scientific jobs and amount of money.

Not something a Howard Hughes investigator has to worry about!

Frankly? I wish I had HIS problems.

Not something a Howard Hughes investigator has to worry about!
Frankly? I wish I had HIS problems.

OK, you got me there.

k4k

Not to mention the mass of predatory, open access journals. A blog called Scholarly Open Access keeps track of not-recommended publishers and not-recommended journals. http://scholarlyoa.com/

Whoa. 120 computer-generated papers published by academic journals! Including some Springer journals:
http://www.cnet.com.au/publishers-remove-gibberish-computer-generated-research-p...

If you go to the site where the software's creators talk about this software, they have a couple of example papers, one of which was accepted at a conference.

It quickly becomes clear that there is no real topic, just a kludge of various buzzwords, like "network," "private-public key encryption," etc. And the references list contains some not very subtle jokes, e.g., a paper in computer science in which one E. Schroedinger is listed as a co-author. Twice. Along with his noted co-author S. Hawking.

The software’s creators also managed to get a set of three papers accepted as a complete session - with the computer-generated title "The 6th Annual North American Symposium on Methodologies, Theory, and Information" - at a conference.

#9.....Hilarious. I wonder if that would work for padding term papers in today's schools? -"What?? You don't like my documentation??" ;-)

OMG. There are sites that sell papers now. Imagine what it will be like when they can just generate them. Oy.