SoylentNews is people
SoylentNews
In the ongoing open access debate, which oldmedia publishers have been able to drag out for decades, oldmedia publishers have repeatedly made the assertion that articles in their very expensive journals are greatly improved during the publication process. Glyn Moody, writing at Techdirt, discusses the lack of value added by expensive, subscription-only journals over the original, freely-available pre-prints of the very same papers, thus negating the claims from the oldmedia publishers.
Such caveats aside, this is an important result that has not received the attention it deserves. It provides hard evidence of something that many have long felt: that academic publishers add almost nothing during the process of disseminating research in their high-profile products. The implications are that libraries should not be paying for expensive subscriptions to academic journals, but simply providing access to the equivalent preprints, which offer almost identical texts free of charge, and that researchers should concentrate on preprints, and forget about journals. Of course, that means that academic institutions must do the same when it comes to evaluating the publications of scholars applying for posts.
Scientific method requires that hypotheses be testable, and that means publishing anything necessary for a third party to reproduce an experiment. So some might even say that if your research ends up behind a paywall, then what you are doing is not even science in the formal sense of the concept.
Previously on SN :
New York Times Opinion Piece on Open Access Publishing (2016)
India's Ministry of Science & Technology Join Open-Access Push
(2015)
Open Access Papers Read and Cited More (2014)
(Score: 2) by Lester on Friday March 16 2018, @09:49AM (14 children)
Publishers filter contents, and that's an important task. They don't accept crap, well, sometimes they do, but 99% of accepted papers are serious stuff. When you read one of this media you are certain in a 99% that it is a serious paper. Without publishers it would be like googling, you would have to pick the gems from a lot of crap. Of course it has a drawback, a lot of good papers are never seen.
Do they improve articles? I don't think so, and this research shows it.
Could filtering be done without publishers? Repositories of filtered articles are needed, and someone has to filter them, read them, get some and ditch another, and then add it to the repository. Call them publishers, editors, peer review committee, or whatever you want. Probably it needn't to be a private company, a group of academics could do it. But then there is the danger of endogamy. But in digital age, even a private publisher needn't to be that expensive, look O'Reilly, it is not an scientific publisher but has to filter a lot of stuff.
The problem is that it is an oligopoly, it looks that after a lot of year a few publishers have built up a lot of prestige and nobody dares to try to defy them starting a new digital "magazine"
(Score: 3, Informative) by PiMuNu on Friday March 16 2018, @10:09AM
Elsevier support some crap journals (publishing rings). I couldn't find a good reference; I recall reading about it a few years back, and have seen some evidence to support the case (same publication in two journals).
(Score: 2) by PiMuNu on Friday March 16 2018, @10:11AM (1 child)
Two replies to the same post, sorry, but I believe arxiv now require sign-off from a "wise person" before papers can be submitted.
(Score: 2) by Lester on Friday March 16 2018, @11:32AM
Is every paper in arxiv published in one of the big ones?
(Score: 2) by MostCynical on Friday March 16 2018, @10:12AM (2 children)
Quality is usually (but not always) helped by peer review.
Whiłe there are plentyof journals that are open and peer-reviewed (yes, some have membership requirements), the problem is that *most* universities have "deals" with the profit-making journals, so the researchers don't really notice the cost. When all the universities say "no", things might change - but not in my life time.
http://www.google.com.au/search?q=open+peer+reviewed+journals [google.com.au]
https://reviews.libraryjournal.com/2012/04/reference/plan-b-life-after-the-big-deal/ [libraryjournal.com]
"I guess once you start doubting, there's no end to it." -Batou, Ghost in the Shell: Stand Alone Complex
(Score: 4, Interesting) by canopic jug on Friday March 16 2018, @10:24AM
Whiłe there are plentyof journals that are open and peer-reviewed (yes, some have membership requirements), the problem is that *most* universities have "deals" with the profit-making journals, so the researchers don't really notice the cost. When all the universities say "no", things might change - but not in my life time.
It may happen sooner than you think. Germany has given up on Elsevier [the-scientist.com] as have a few other countries. Developing countries can't afford even one basic bundle. (Bundling a la cable tv, with the good journals spread over several different subscription packages.) It won't take much in some fields to break their hold, but the hold is based more on lobbying than inertia, though inertia is also a big reason. And it will go field by field and not all fields at once. Researchers in each field publish in certain journals because of the Impact Factor [omicsonline.org] which is weighed in their tenure assessments and other career-essential evaluations.
Money is not free speech. Elections should not be auctions.
(Score: 3, Informative) by Wootery on Friday March 16 2018, @10:43AM
I'm more optimistic. Hopefully we'll see a continuing rise of the open-access mandate. [wikipedia.org] Things are going in the right direction.
(Score: 5, Informative) by AthanasiusKircher on Friday March 16 2018, @12:03PM (3 children)
Uh, just to be clear, a "group of academics" almost always DOES do it -- and almost always free of charge. Almost all reputable journals have an editorial board staffed by academics, an editor (who is usually an academic), and solicits 3rd-party peer reviewers, who are generally not paid.
The only paid people tend to be people who do the copyediting, typesetting/design (where necessary), distribution, etc. All of the actual content review is done by academics.
Which is why you've had many "declarations of independence" [simmons.edu] by entire journal editorial boards, who then go off and found a journal often with lower publication costs. They're the ones actually doing the most important work of filtering good articles from bad and suggesting content improvements... they only tend to be attached to traditional journals because those journal titles still carry some weight of reputation in the field, which helps promote citations and a wider readership, provide information during tenure review, etc.
Which brings us to the biggest hole in the proposal as mentioned in TFS:
That's a HUGE issue. The reason why big-name traditional journals have a stranglehold on academics is because academics publish in them to get noticed. People in their fields know the "big name" journals. They know the ones with a reputation for excellence. They carry more weight when you see them on a CV for a job application, or when someone is applying for tenure or promotion.
Having an article just posted randomly as a "pre-print" on an online repository is just an unknown to a review committee trying to judge the work of someone -- is the research worthwhile at all? Unless you're a subject-area specialist, you probably have no idea. (And is it still a "pre-print" if it's never intended for PRINT???) Publishing it in "Bob's Backyard Low-Fee Crap Journal" also won't carry weight, unless you have a solid editorial board with recognizable names for the journal.
And again, that's why some editorial boards have seceded en masse to form new journals -- but that only can work if you have some really big names, and even then it might take quite a few years until the new named journal gets a reputation.
(Score: 3, Interesting) by Lester on Friday March 16 2018, @12:45PM (2 children)
Yes, I know. I suppose that the protocol is: The journal receives a paper, reads it, decides after that first skim if it is worth and then sends to academics for a deeper verification.
I said: "But then there is the danger of endogamy". In the end it is read also by academics, so what's the difference. The third party in the middle makes the difference, a subtle but important psychological difference. If scholar sent the papers directly to other colleague it could degenerate in a Quid pro quo. This third party is a barrier against such endogamy.
Yes, I agree. Nevertheless , it is nice to read that OMICS International [omicsonline.org] is getting momentum. Hopefully we won't have to wait so much years.
(Score: 3, Informative) by shrewdsheep on Friday March 16 2018, @01:36PM
https://en.wikipedia.org/wiki/OMICS_Publishing_Group [wikipedia.org] is considered predatory. It's a complicated game. OTOneH there are traditional journals extorting money from academia. OTOH there are the open access journals with an incentive to accept (for money of course), polluting science, and also extorting academia.
Open access, done right, is difficult.
(Score: 5, Interesting) by AthanasiusKircher on Friday March 16 2018, @01:50PM
Actually, NO. As I said in my post, the editor and editorial board are academics too. Typically, at a journal, the submission is processed by an editorial assistant, who is often -- though not always -- an academic too (frequently a grad student or someone near the start of career). The editorial assistant generally ONLY looks to see that the submission conforms to BASIC formatting guidelines, such as a page/word limit, requests that examples/images be provided in a separate file or separate part of the document, etc.
Then the editorial assistant will pass the article onto an editor (an academic). If the editor is familiar with the author of basic subject content and it looks promising, the editor will send it out to other academics for peer review. If the editor isn't familiar with the topic directly, the editor may send it to another member of the editorial board (also an academic) who has more specific subject area competence. The editorial board member may give it a skim before it goes out for peer review.
And again, it's typical that none of these academics are paid anything for what they do -- except editorial assistants. Sometimes academics who serve as main editors get a small honorarium or something, but the reason most do it is for the prestige in the field. This is not true for all journals: less reputable ones sometimes pay more academics, because academics are sometimes less willing to do work for less prestigious journals.
But the point is -- the non-academic staff at the journal generally do NOTHING in terms of quality-control for the research. They're only concerned with formatting, layout and design, copyediting, etc. in terms of quality. (And frequently these days a lot of that is outsourced to India or something even at top-tier journals. And that often leads to communication issues in the publication process.)
Sorry, but I'm not sure you have an informed perspective on academic publications and journal politics. Except at the top-tier journals in a given field, it's common for academic editors to send out articles for peer review by "favorable reviewers" if they think an article is worthwhile themselves. Although peer review is supposedly anonymous -- both for the author and the reviewer -- it's frequently easy to guess who someone is in a given subfield given the topic of the article or by the type of comments made by the reviewer. I personally know people who have held grudges against other people in their field whom they ASSUME were reviewers on a rejected article based on circumstantial evidence. Some emerging subdisciplines have a culture where they tend to review other papers in the subdiscipline a little more favorably, because everyone in the field wants more attention paid to their stuff, etc. Heck, even REDACTED citations in an article can sometimes be a clue as to who the author is.
I don't mean to portray this as though the whole system is "quid pro quo" -- because it certainly isn't. Most academics have a certain level of integrity and standards. But there's at least the theoretical potential for politics to play a role in the process at just about every stage, even the supposedly "anonymous" ones.
Point is -- whatever you imagine third-party journals are doing as adding a "gatekeeper" role by being a disinterested party... it probably isn't happening. Academics basically run the whole process. (Also, by the way, publishers who actually employ in-house editors to do things like make publication decisions on content or to set up book contracts with academics, etc. -- guess who those in-house editors generally are? Mostly drawn from academics in the field. Everyone knows who they wrote their dissertation with, what it was on, etc. After all, who else are publishers going to employ if they actually want someone to JUDGE CONTENT except a subject matter expert??)
(Score: 4, Interesting) by bzipitidoo on Friday March 16 2018, @02:25PM (2 children)
My experience is that peer reviewers and journals are also unduly swayed by trivial considerations, that they, like everyone else, use mental shortcuts to evaluate a submission. For instance, does the layout, format, and manner of presentation fit with tradition? Of course that has nothing to do with whether the finding is valid, but reviewers will be swayed by that one, in that any departure from such norms could indicate that the authors don't know what they're doing, perhaps are noobs, and so should be scrutinized more harshly.
Sometimes reviewers don't get it. They misunderstand a paper, and reject it when it should have been accepted. It is of course impossible to know how significant a work will be before it is published.
Another huge problem is their slowness and jealousy. They take months to review a paper, and mention, over and over, that it is naughty and unethical to submit a paper to more than one journal simultaneously. So a paper gets held hostage a very long time. If it is then rejected for bad reasons, the authors have lost a lot of precious time during which others could make the same discoveries.
One of the worst possibilities is the dishonest rejection so that a reviewer can take what is a good result, rework it some, and submit it elsewhere as their own work, thus stealing the paper. The pressure to Publish or Perish tempts professors into such things. One hears all the time about professors taking credit for the work of their grad students. So why steal only from grad students?
(Score: 1, Interesting) by Anonymous Coward on Saturday March 17 2018, @02:08AM
> So why steal only from grad students?
Especially when there are post doctoral researchers are available. One post doc had done a great dissertation and stayed on doing post doctoral research for his advisor. He was doing phenomonal work for his former advisor and still getting many single-author research articles published in the prestigous journals for his field yet over and over again could not find a job. There was always massive interest whenever he first applied, which was always followed by silence and a brush-off. After some years of that he had some friends pretend to be potential employers. When they got to calling his former advisor for references, the former advisor badmouthed him severely in order to be able to keep him on doing work at slave wages. Once that was corrected, he found a good position at a top university within a few weeks and moved on.
(Score: 2) by FakeBeldin on Saturday March 17 2018, @12:28PM
That's an interesting (and valid) point.
There are some ways to mitigate this - e.g. by uploading a "preprint" to a preprint server such as arxiv.
In the fields of research in CS with which I'm somewhat acquainted, discoveries happen to quick for journals anyway. So the main publishing venues are conferences (workshops, symposia, whatever). These typically have a far faster turnaround time (a few months). Then, once published, an expanded version is sometimes submitted to a journal.
Of course, that's my experience for academics in well-off countries. In countries with far less science funding, they obviously cannot afford to travel to conferences too often. There, journal publications are far more important - even if the journals typically carry far, far less academic prestige than the conferences.
(Score: 0) by Anonymous Coward on Friday March 16 2018, @09:30PM
No, they don't.
All of the content decisions are made by academics - typically by publicans funded ones (ie. by your taxes).
Publishers provide the website, the review/editor website, and the dead tree version. All the rest is being done by folds in academia.
The one thing of value publishers "provide" is some assurance of quality. Mostly needed for new publication venues (new conferences, new journals). For anything with an established reputation, they sit back and rake in the money.
Which, by the way, in many countries comes ultimately from taxes.
So that's why everyone should care. Publishers produce (typically) a black and white magazine, that appears less then monthly, don't write the content, and charge more than thousands times what you'd pay for a monthly magazine in a store. Which is paid with (your) taxes.
(Score: 4, Insightful) by rigrig on Friday March 16 2018, @10:06AM (1 child)
No it doesn't, it involves formulating a hypothesis, thinking of a way to test it, predicting what will happen, performing the experiment, and updating your hypothesis based on a comparison of your result with your prediction.
No need to get anyone else involved.
This would imply that what you are doing is not science unless you make sure everybody has unlimited access to your results:
This is like saying that racing someone by running 100 meters is not sports in the formal sense of the concept if you post who came first behind a paywall.
No one remembers the singer.
(Score: 0) by Anonymous Coward on Friday March 16 2018, @11:57AM
Not that what you posted is badwrong, but in general if you have to resort to analogy to argue something maybe you need better arguments.
(Score: 4, Interesting) by Anonymous Coward on Friday March 16 2018, @11:45AM (2 children)
There are several things journals traditionally provide:
I guess it would be possible to separate some of those functions. We already have repositories like arXiv which provide some of those items (permanent storage, availability, some degree of authenticity, a minimal filter).
One might think of a service that provides refereeing of arXiv submissions, and provides more authenticity by digitally signing those it accepted (so you see that this, and not another, version of the article is the one reviewed by this organization). Unlike journals, several organizations could review the same article, giving articles who have the stamp of several organizations a higher degree of credibility. The referee service signatures could be taken into account by general and specialized search engines to give articles with many signatures from respected referee agencies higher visibility. Those organizations would likely also have hand-curated indices of articles they refereed.
There also might be paid services which provide pre-publication refereeing: The author can pay them for a refereeing process, so that at the time they publish the article, they already have the referee service signature applied, just as with today's journal articles when the authors choose not to put a preprint on arXiv before publication. The digital signature ensures that the authors cannot put up a different version and claim that were the one refereed.
(Score: 2) by Thexalon on Friday March 16 2018, @12:27PM (1 child)
Most importantly, what these journals provide is bragging rights for the people published in them. "I published 3 articles in Nature, so you should give me a tenure-track position or at the very least a raise." Since a substantial percentage of academic papers are never read after publication, that matters more than any of the contents of said papers. If your system doesn't provide the same kind of bragging rights, it won't be adopted. Period.
The only thing that stops a bad guy with a compiler is a good guy with a compiler.
(Score: 1, Insightful) by Anonymous Coward on Friday March 16 2018, @01:18PM
I don't see why being approved by a high-reputation refereeing service shouldn't give the same bragging rights as being approved by a high-reputation journal. And the ability of having several independent approvals of the same article just increases the bragging opportunities ("my article was approved both by Nature and Science!")
(Score: 2) by All Your Lawn Are Belong To Us on Friday March 16 2018, @12:50PM (2 children)
They provide an established name and branding. This may seem like marketing BS, but there is a big difference between, say, Journal of the American Medical Association or New England Journal of Medicine and Lawn's Huuuge Open Journal of Medical Stuffz!. (Yes, I have ones I can pick on. No, I won't.) Generally anything that makes it through peer review at JAMA or NEJM is something that is reliable. Are there exceptions? Yes - they can get snookered too. (That isn't an argument for Open Access - it is the opposite, in fact). But JAMA and NEJM have names to protect - and if they want the prestige they still have to protect it. I hate to say it, but their profit motives do ensure that they take care with their content and process.
What protects Open Access from False Academy [nih.gov]? That article hints at the broader truth - the issue isn't the profit motive. It is one of trust and reporting accurately. Charging for a journal doesn't eliminate that problem, true. Neither does making it free - in fact, it could worsen the reliability issue as the "publish or perish" motivation is strong and be an unethical motivator. Science can suffer either way.
And yes, there is a big difference on a resume of, "I published an article in NEJM," and, "I published an article to Lawn!" There should be - there is a hierarchy of importance of information and discovery. I would love to see an open journal take that first place. But this will occur organically, when an open journal gains more credibility and quality than them.
Last thing.... if the publishers didn't exist, would there be preprints to give open access to? (Not talking OA journals - talking preprints of pay journals - pay journals disappear then do the same amount of articles occur? That's an open question that AFAIK can't be answered.)
If the old world journals survive, there are reasons that they do. And, in my opinion, they should. If the system eventually eliminates the need for them, they will disappear. But not this year.
This sig for rent.
(Score: 1, Interesting) by Anonymous Coward on Friday March 16 2018, @01:25PM
Open Access does not mean the journal doesn't have a reputation to lose. If a journal accepts just anything, then publishing there will be no better than self-publishing. Therefore people will not be willing to pay for it (and author payment is how OA journals make their money).
Indeed, OA journals have even more to lose, as unlike traditional journals, they cannot hope to make money from continue to sell (access to) earlier issues. They can only thrive as long as people publish there, and people will only publish there as long as they see an advantage to do so, especially given that it costs them money.
Moreover, the higher the reputation of an OA journal, the more money they can ask from authors, as the authors will be more eager to publish there.
(Score: 0) by Anonymous Coward on Friday March 16 2018, @04:31PM
American Medical Association is getting a lot of money for doing almost nothing, just sends papers for review (for free) to some of the same people that will read it latter. Why couldn't peers that review papers join in a association and do what they do for American Medical Association with much less money?
(Score: 3, Insightful) by AthanasiusKircher on Friday March 16 2018, @12:50PM (3 children)
I agree with a lot of other comments here that journals mostly add only a content-filtering mechanism. But the research cited in TFA is fundamentally flawed, since it compares pre-print versions with final published versions of articles accepted for publication.
Which means the accepted articles were good enough to be accepted, so they met certain expected professional standards, because authors were motivated to improve their research/writing to get it published.
You can't go from the observation that high-quality pre-prints are similar to high-quality published journal versions to the idea that if you just removed content and quality filtering, it would still all just "be good."
To put it another way -- authors aspire to be published in a reputable journal. Take away that motivation and those standards, and you now have authors who don't take as much care preparing their articles, perhaps not even as much care in their research itself.
TFA is like a business executive arguing that we don't need Quality Assurance in manufacturing, because we don't see any significant improvement in product quality by having it. Except, that analogy is a little flawed, given the power journals have over academic careers. Imagine if to get another job, you needed a recommendation from the people in your Quality Assurance department. Imagine if promotions at a future company were contingent on stellar reviews from former Quality Assurance personnel. I think you'd be darn sure when you sent your products to QA, they were already strongly vetted and were solid work.
That's effectively the situation scientists are often in, because publication reputation is so critical to their careers. Remove that, and a lot motivation for high-quality output disappears. An executive who argued in such a situation, "Oh, we don't see any 'valued added' by our QA personnel, so let's get rid of 'em" is not only out-of-touch, but perhaps delusional.
NOTE: I'm NOT arguing in favor of the current system, with its ridiculous costs. But it's going to take a lot more to come up with a substitute system than simply saying, "We don't need journals because they add no value." The value often isn't in the revision or copyediting or whatever -- it's in the aspirations of those who seek to publish in high-quality journals in the first place.
(Score: 2) by AthanasiusKircher on Friday March 16 2018, @12:54PM (2 children)
Oh and sorry for the self-reply -- but to make something explicit that was implicit in my argument: the study mentioned in TFA obviously doesn't account for articles sent to the journal but REJECTED. In the QA analogy, it's like an executive looking only at the products that made it through QA and saying, "that department doesn't add anything."
(Score: 0) by Anonymous Coward on Friday March 16 2018, @01:20PM (1 child)
There was that editor of a medical journal who said they selected papers by throwing them down a flight of stairs and publishing whichever reached the bottom. Then for one issue they only published papers that failed peer review and nobody noticed.
https://en.m.wikipedia.org/wiki/Richard_Smith_(editor) [wikipedia.org]
(Score: 2) by AthanasiusKircher on Friday March 16 2018, @02:16PM
To be clear, I didn't say that peer review process is good. There in fact seems to be conflicting evidence of its value, and I know many studies question it. My point was that the research mentioned in TFA is a poor way to judge the potential value of the journal process. (That is -- just because we know peer review is often broken doesn't necessarily mean that we should accept the conclusion of even a poor study with bad methodology because it agrees with what we expect. I'm sure there's a meta-moral of sorts in this.)
(Score: 3, Insightful) by cocaine overdose on Friday March 16 2018, @12:59PM