The Arxiv Is Not a Journal

There's been a lot written recently about academic publishing, in the kerfuffle over the "Research Works Act"-- John's roundup should keep you in reading material for a good while. This has led some people to decide to boycott Elsevier, including Aram Harrow of the Quantum Vatican. I'm generally in favor of this, but Aram says one thing that bugs me a bit:

Just like the walled gardens of Compuserve and AOL would never grow into the Internet, no commercial publisher will ever be able to match the scope and ease of access of arxiv.org. Nor can they match the price. In 2010, there were about 70,000 new papers added to arxiv.org and there were 30 million articles downloaded, while their annual budget was $420,000. This comes to $6 per article uploaded (or 1.4 cents per download). Publishers talk about how much their business costs and how even “open access” isn’t free, but thanks to arxiv.org, we know how low the costs can go.

This is very nice, but has one major problem: The arxiv is not a journal.

The arxiv does do one of the things that we associate with academic journals, which is to distribute papers to a broad audience. But that's only one thing out of the list of jobs performed by a typical journal. The arxiv does no peer review-- as I understand it, new submissions get a quick glance to make sure they're not barking mad, but that's it. There's no copyediting or layout work done on the submissions, as anybody who's tried to slog their way through some of the confusing crap that's up there ought to know. And those services cost money.

"But referees work for free!" hard-core partisans break in. Which is true-- academic journal referees are not paid for refereeing papers. So, technically, it's true that referees work for free, to the extent that they work at all.

I mean, what's the usual refereeing process? You get sent a paper, glance at it quickly, and set it aside. Two or three (or eight or ten) weeks later, you get an email from the journal editor reminding you that you were sent a paper to referee a while back, and then you dust it off and read it over and either write a report or write a letter saying that you can't review it.

I'm exaggerating a tiny bit, but not by much. I've been very lucky with the peer review process in my career, but I have never gotten a paper through without at least one email sent to the journal asking what the hell was going on with my paper. The APS journals provide (or did, back in the day) a helpful service allowing you to track your paper through the process, which revealed that mine were not the only emails being sent to nag people.

That stuff doesn't happen for free. Somebody has to coordinate the sending of papers to referees, so they're not all going to one person, or randomly going to blood enemies of the first author. Somebody has to keep track of how long each paper has been out with a referee, and send the appropriate nagging emails. Those people won't work for free-- somebody is going to want to draw a salary for doing those jobs.

So, citing the low cost of the arxiv is basically the academic equivalent of citing the low cost of paper and ink for trade publishers. It's true-- I'd be shocked if Elsevier was spending more than $6/paper on web hosting-- but not really that significant a part of the overall cost.

(To be fair to Aram, he seems aware of this, and has some suggestions toward the end of that post about what it would mean to make a journal out of the arxiv. I've seen the same figures thrown around elsewhere, though, without any acknowledgement of the problem with the comparison, and it bugs me.)

Now, does this mean that Elsevier isn't gouging people? Almost certainly not. All it means is that the arxiv isn't really a valid comparison. If you want to get all outraged about their profit margins, there are legitimate comparisons-- any of the PLoS journals, for example. And, in fact, if you look up the publication fees for PLoS, you see that they're not all that far off the $3,000 that Elsevier charges for their open-access options. So when you compare two actual journals to each other, rather than comparing a journal to a hosting service, the difference isn't quite as dramatic.

(You could, of course, argue that even PLoS is grossly overcharging people for their services, for whatever reason. But while the real cost may not be measured in thousands of dollars, it's not going to be $6/paper. If you want to try to do a Fermi-ish estimate of the real cost per paper, I'd guess that coordinating peer review of all those articles probably requires a minimum of an hour per paper of the time of somebody who will expect to be paid for their time. Which probably puts you up in the neighborhood of $50/paper, maybe even $100 as a more realistic lower bound to the total cost of running a real journal.)

More like this

Via Twitter, Daniel Lemire has a mini-manifesto advocating "social media" alternatives for academic publishing, citing "disastrous consequences" of the "filter-then-publish" model in use by traditional journals. The problem is, as with most such things, I'm not convinced that social media style…
It is infuriating how stodgy biomedical sciences are in terms of information sharing. It's not clear how much of this is bred of inherent conservatism, the pressures of a very competitive field or just plain technobackwardness. But while mathematics and physics have had preprint servers for years,…
There is one small event from the conference (Publishing in the New Millennium: A Forum on Publishing in the Biosciences) that I would like to share with you. I asked Emilie Marcus, head editor of Cell, a few questions. But before we go there I'd like to delve into one aspect of the whole open…
John Bohannon of Science magazine has developed a fake science paper generator. He wrote a little, simple program, pushes a button, and gets hundreds of phony papers, each unique with different authors and different molecules and different cancers, in a format that's painfully familiar to anyone…

About 5 years ago I saw a comparison between the cost per page for The Astrophysical Journal (a society owned, nonprofit journal that aims to break even) and an average Elsevier journal. There was a factor of 10 difference. To be fair, The Astrophysical Journal gets slightly more than half its revenue from page charges, so that actual difference is only a factor ~4. That's still a big difference.

Your U$S 100 estimate is spot-on. The journal papersinphysics charges exactly that for a 4-page paper, plus U$S 15 per extra page. It's a real journal with peer review, and a CC-by license for all the papers. It's very new, so it is not very well known yet, but this shows it can really be done for a very little amount of money.

Disclaimer: I've got no affiliation with the journal, but I've published in it.

Yeah, I've always been skeptical of the comparison to Arxiv, too-- without the peer review, the value of the publication drops from journal, through conference, and does a dead-cat bounce off of self-published pamphlet.

(And the bounce is mainly because I guess you might get a citation from Arxiv.)

By John Novak (not verified) on 30 Jan 2012 #permalink

I've seen that argument bandied about a lot, in the context of the evils of publishers or the failure of the referee system. Always struck me as a symptom of a larger problem -- to a certain extent we are stuck with the current system because no better alternatives exist. To a certain extent, no better alternatives exists because the alternatives most often discussed cannot do an adequate job in replacing the current system. The academic system is conservative and resistant to change, and all the utopian ideas and radical changes proposed are unlikely to gain much traction, especially if you can poke holes in them after thinking for 20 seconds. The only success stories I am aware of, for example JHEP in my field, concentrated on achievable incremental changes (moving to web-based system, taking on more of the administrative duties, negotiating better deal with a professional society based publisher). This did not achieve the orders of magnitude savings one can dream of, but at the very least the money goes somewhere more appealing.

There do need to be people you can contact, so the costs for a peer-reviewed journal won't be as low as for arXiv, but sending nagging emails is a job for a computer if there ever was one. In fact, I'd be surprised if publishers weren't already doing it that way.

Two recent items from the New York Times throw some light on issues in science publishing.

http://www.nytimes.com/2012/01/11/opinion/research-bought-then-paid-for…

http://www.nytimes.com/2012/01/23/opinion/should-research-be-more-freel…

The second, by Gene Sprouse and Joseph Serene, who are the editor in chief of Physical Review and the treasurer of the APS, respectively, make the point that, yes, running a journal costs money. Papers come in, they have to be tracked and assigned to editors. Editors have to find referees, read the reports when they come in, and then decide what further actions to take. There is technical editing to be done. The APS journals have pretty impressive IT systems for handling papers, all of it designed, developed and maintained in house. These tasks require people, and those people have to be paid. It is the case, however, that the subscription costs of society journals tend to be much lower than the subscription costs of journals from commercial publishers, so the high prices charged by some publishers have more to do with profits than anything else.

There is another related issue, which is how to pay for journals. Should they be open access with an author-pays model, or subscription based with a user-pays model? There are some people, especially in the bio-medical area who want everything to be open access, author-pays. There are other people, such as some theoretical physicists and mathematicians (there was an anti-author-pays editorial in the October issue of the Notices of the American Mathematical Society), who, because of the low level of support for theoretical research in the US, and the fact that you can do theory with little or no grant support, think the present system is fine. Publish your paper in a subscription based, user-pays journal and put it on the arXiv, so anyone who wants to can read it for free.

I should point out that though I am an associate editor of Physical Review A, all of the opinions expressed here are my own and not necessarily shared by the APS.

By Mark Hillery (not verified) on 30 Jan 2012 #permalink

I'd just like to point out that for normal unfunded or between funding people, there's a huge difference between a couple hundred dollars and a couple thousand dollars, in terms of ability to pay.

Yes, that's one aspect. There is another aspect that I find missing in the debate about the 'Elsevier boycott':

Elsevier and many other publishers are for-profit organizations. Complaining that they want to make money is like planting potatoes and then complaining that they need water. If you want to keep your water, you better not plant the potatoes in the first place. That's a dumb analogy, I know, but my point is that the publishing system that we presently have is for-profit. That's the facts. I have said many times that if it's a public service, it should be funded like a public service, but, needless to say, nobody is listening to me. As you say, publishers do provide services, and somebody has to pay for it, and presently that's for-profit.

Now the point that most people seem to miss here is the question who will suffer if the publishers' profits fall. Well, if you don't water the potatoes, in the end you'll be the one suffering. Same with the publishers. If their profit margin shrinks, they'll not declare bankrupt from one day to the next. They will try to cut back on things they can no longer afford. What will they do? They will merge or dump non-profitable journals, or book series, and export copy editing to India (if it's not already there). That means that it will in future be difficult, if not impossible, to get any journals stated in fields that are not considered profitable (and textbook quality will drop). Look, publishers use their profit-buffer to finance the non-profitable part of science publishing. Unless you have an alternative way to cover that, you better think twice what you're doing to science when you boycott a system that's been serving us quite well so far.

To make that really clear: I am not saying it's a good system. I have written many times on its shortcomings. What I am saying is it can easily get worse.

Essentially, this comes down to the value of peer review. Can't one, as a scientist/researcher, evaluate a paper on arxiv him/herself? Why does an expert in a field need a journal gatekeeper to determine whether or not a submitted paper is worth reading/citing?

Even if there is some value in that process isn't there also cost -- e.g., the unwillingness to publish statistically insignificant results or results that fail to replicate -- that make the net value of journal's peer review more ambiguous?

Essentially, this comes down to the value of peer review. Can't one, as a scientist/researcher, evaluate a paper on arxiv him/herself? Why does an expert in a field need a journal gatekeeper to determine whether or not a submitted paper is worth reading/citing?

An expert in the field doesn't need peer review to evaluate a paper in their area of expertise. As long as you're comfortable with only experts being able to read the scientific literature, there's no need for peer review.

If you would like to broaden the scientific conversation to include students (who, by definition, are not yet experts), or even scientists who are experts in one field but want to explore a different field, there is great value in peer review as a filter.

Two other reasons why peer review is essential:

The record produced by the community should be usable for people displaced both in space and in time. I've used results from 40 year old papers without having to check them for myself, I am glad someone was there to check them and give them the seal of approval. Equally, I see my responsibility now to maintain a record of which small part of the literature is correct and useful.

Secondly, publications are used for evaluation of scientific merit, both for promotion and for allocation of grants. This is typically not done, and cannot be done, by experts of that person's community alone (it is easy to come up with reasons why that would not be a good idea). As with any other kind of evaluation, some kind of bean counting is unavoidable, and making sure your effort are judged favorably by your peers is not a bad piece of information to use.

Also, ideally the process of ensuring a paper is fit for publication involves a dialogue with the referee, which improves the quality of the finished product. Some communities are better than others in coming close to that ideal.

@Chad: One other difference between Arxiv and a journal: the journal can offset costs through advertising.

Certainly PLOS advertises, but compared to a publishing house like Elsevier, Wiley, or NPG that can offer an advertising platform coordinated across web and print versions of multiple interrelated high impact journals, I doubt PLOS takes in anywhere near as much advertising revenue per paper. I could see an open access medical journal potentially offsetting all of its costs through advertising, but most fields won't be able to charge anywhere near as much per paper as medicine.

@Bee

That means that it will in future be difficult, if not impossible, to get any journals stated in fields that are not considered profitable (and textbook quality will drop). Look, publishers use their profit-buffer to finance the non-profitable part of science publishing. Unless you have an alternative way to cover that, you better think twice what you're doing to science when you boycott a system that's been serving us quite well so far.

I think if Elsevier keeps any of its publications operating at a loss for long it's not out of some sense of duty; it's because keeping those journals going enhances their overall profitability. If those journals didn't help Reed Elsevier PLC , they would get merged or scrapped. I'll refer you to Reed Elsevier PLC's corporate strategy page. If it's cheaper (associate editors aren't paid much) to outsource copy editing they'll do it even if profit margins are already healthy .

As to an alternative way to cover unprofitable fields, that has already been well established for many years: unprofitable publishers. I.e., non-profit. See comments #1 and #2 above yours for examples. In terms of your analogy: the farmers chose to plant drought resistant varieties of potatoes and keep their water too.

@cb: I'm not certain how much value publishing ambiguous results would add. It would be great if there was more of an incentive for both scientists and journals to publish papers that confirm the null hypothesis or reliably fail to accomplish the intended goal (again, you can do that more in medicine, but less so in other fields). If the research was meticulously carried out and written up, those can end up saving other scientists a lot of time and effort. Research that ends up with ambiguous and unreliable results still provides a cautionary tale, but it's the least useful form of communication.

At any rate, putting both negative result and @!$#!?? result type publications through peer review might help the authors (and others) find out where to go from there.

Peer review does not only happen at the refereeing stage. Rebuttals are written of questionable claims. Citations occur or don't. Area reviews are written to summarize and synthesize the numerous articles published in a field. Professors recommend articles to students, certain areas of research are selected to appear in broader circulation textbooks. There are even awards to recognize exceptional achievement.

The Internet is mostly cruft. But we reward those who help us manage the garbage. Search engines, aggregator blogs, review sites, consistently high quality content producers: all seem to end up with good readerships. In the end, I don't find it difficult to find high quality content on the Internet.

My point is it doesn't need to be a choice between publishers-as-gate-keepers and sifting-through-garbage. Was academia to move to an everybody-in-arXiv model, alternative mechanisms for recognizing paper quality would appear. I'm not saying we should just jump headfirst into full, unrefereed publishing. But very rarely is a single bottleneck the best quality filter. The appropriate filtering mechanism is going to be different for a subject expert, a funding committee, a student and a private-sector researcher, and that filtering could all occur after the process of putting a paper on a server happens.

Several scattered comments about the whole thing:

1) My issue with Elsevier is not with the notion of making profits, but with the notion of making profits off other peoples' uncompensated work. My understanding is that Elsevier journals are still operating on the unpaid professorial/senior grad student review mode. Plus also, the lobbying attempt to remove the public access requirements for work funded by NSF, NIH, etc.

2) Nothing does stop anyone from reviewing Arxiv papers, but unless you're 19 years old, you should realize that "anarchy" is not a good way to institute "quality control." And that's what peer review and editing are, in large part: quality control.

(In fact, Arxiv is exactly what you get when your quality control strategy is benevolent anarchy.)

3) Speaking of quality control and peer review, here's a rather topical
failure
in a new journal that's actually supposed to have peer review.

If you aren't an expert in the field, you can probably tell that that's pure effing drivel, but unless you're a Nobel Laureate, there are bound to be border cases. Hell, some of us are interdisciplinary, and it's nice to have peer review to fall back on as an indicator of quality.

And even if you are an expert in the field capable of masterfully distinguishing all good papers from all bad papers, do you want to spend all your time on the alert for nonsense like that?

4) If I actually have a general point, here (and I'm not sure that I do) it's that perhaps neither of the main systems (high-profit organizations like Elsevier, vs high-anarchy systems like Arxiv) are working out, so perhaps it's time to do something different.

(Yeah, easy for me to say.)

By John Novak (not verified) on 31 Jan 2012 #permalink

The arxiv does no peer review-- as I understand it, new submissions get a quick glance to make sure they're not barking mad, but that's it. There's no copyediting or layout work done on the submissions

Publishers don't do peer review either. Journal editors organize it, and most of them work for free just like the reviewers do. The few that are paid... I know an editor of an Elsevier journal who is paid 250 US$ per year, which amounts to 5 $ per manuscript that he handles.

Open-access journals do have peer review. Rumors about PLoS ONE lacking it or having less of it are false; it merely doesn't care about whether a manuscript is a sensational breakthrough (the most important criterion for the journals with the highest impact factors).

Publishers do usually provide layouting, but increasingly less copyediting. Many journals don't bother with it anymore; if the authors don't catch a mistake, it's printed. Nature Itself published on a "tuberocity" in 2001 and hasn't gone back since. As far as layouting goes, there are already journals where authors â much like in arXiv â are expected to submit their manuscripts in the final LaTeX format.

To summarize:

â Authors write the manuscript, make the figures, usually copyedit it and sometimes layout it. Authors are paid by tax money.
â Authors send manuscript to journal, at no cost to the publishers.
â Authors sign over the copyright, their mortgages, and their firstborns to the publishers. Have you ever read a copyright transfer notice? They range from scary to deeply laughable. (Some even pretend they can limit the number of times you can send the pdf to your colleagues, for fuck's sake.)
â Editors read the manuscript to check if it's barking mad and send it to reviewers, at no or negligible cost to the publishers.
â Reviewers write reviews, at no cost to the publishers. BTW, while I've had your experience with lazy/overworked reviewers, I've only had it with 1 or 2 reviewers; most reviewers in my field do decent work in a reasonable time... at no cost to the publishers.
â Editors collect reviews and send them to authors, at no or negligible cost to the publishers.
â Authors improve the manuscript and resubmit it, at no cost to the publishers.
â Editors read the revised manuscript to check if it conforms to the demands of the reviewers and decides whether to accept, at no or negligible cost to the publishers.
â Authors pay nine hundred bucks per color illustration to the publishers. Your institution can't pay that? Your graph with the 14 lines will have to be printed in 14 shades of gray, too bad.
â Editors pass on publishable manuscript to publishers, at no or negligible cost to the publishers.
â Publishers send finished manuscript to copyediting (rarely), layouting (usually) and printing (still almost always), provide web hosting, paywalling and shipping.

The complaint isn't that for-profit publishers make profits. It's the fact that the top 4 publishers make profits of a height that is really hard to believe and continue to rise throughout the crisis of the rest of the economy.

BTW, academics are not expected by their employers/grant agencies to do any reviewing or editorial work. It's a service academics expect of each other, but technically they do it in their spare time, unpaid (or, see above, for 5 $ per paper in very rare cases).

While I am at it.

4) If I actually have a general point, here (and I'm not sure that I do) it's that perhaps neither of the main systems (high-profit organizations like Elsevier, vs high-anarchy systems like Arxiv) are working out, so perhaps it's time to do something different.

1) Author-pays journals with fee waivers for poor authors, like the PLoS journals.
2) Taxpayer-funded journals like Acta Palaeontologica Polonica (Polish Academy of Sciences), Contributions to Zoology (museums in Amsterdam and Leiden) and the like.

By David MarjanoviÄ (not verified) on 01 Feb 2012 #permalink

You know, repeating something over and over doesn't actually make it true. In particular, the repeated bold assertion that the time of professional editors has "no or negligible cost" is a lovely rhetorical device with no connection to reality.

Editors are professionals, often highly-educated professionals-- most of the editors at the APS journals have Ph.D.'s-- and their time is not free. Reading submissions-- even in a cursory way-- takes time. Reading referee reports takes time. Reading resubmissions to see if they've addressed the concerns of the referees takes time. And educated professionals expect to be compensated for their time.

The service that journal editors provide is not free, and no amount of repeated assertion will make it free.

Correction: here's a case of a researcher who is expected by his university to spend 10 % of his time providing "service". Arguably this means he gets tax money for reviewing manuscripts â though whether 10 % for everything together is realistic can be discussed.

Right underneath that, in contrast, is a case of a researcher whose museum expects him to work on nothing but his project full-time. The reviews he does are explicitly written in his spare time, for free.

Two comments farther down: confirmation that math and logic journals require authors to typeset their own manuscripts in TeX before submission.

By David MarjanoviÄ (not verified) on 01 Feb 2012 #permalink

Editors are professionals, often highly-educated professionals-- most of the editors at the APS journals have Ph.D.'s-- and their time is not free.

I wrote "work for free" and "no or negligible cost to the publishers".

In my field at least, most editors are unpaid; the guy with the 250 $/a is the rare exception.

What do you mean by "free"? Do you believe I wrote editors don't do anything, or that they aren't professional, working scientists like you & me? I didn't. I said they aren't getting paid, with exceptions that only support the rule.

By David MarjanoviÄ (not verified) on 01 Feb 2012 #permalink

David, just because the editor himself isn't being paid (and I'll stipulate that many people who are listed on mastheads as editors are not being paid) doesn't mean that nobody is being paid. The journal with which I am most familiar has a paid staff of editorial assistants who work behind the scenes, and if something goes wrong with the process (technical difficulties with supporting material, for instance) it is they, not the editors, that authors and reviewers work with. Back in the days of snailmail submission these editorial assistants were often collocated with the titular editor (who normally had that job for three years before rotating off), and the assistants typically dealt with correspondence and the like. (I know a guy in San Antonio who started doing science by being an editorial assistant for an editor located there; he became interested in the science he was reading about. He does have a Ph.D., but in some humanities field.) Nowadays, with submission happening over the web, these assistants are typically at publisher HQ, but they haven't gone away. Journals wouldn't remain functional for very long if the editorial assistants went away.

By Eric Lund (not verified) on 01 Feb 2012 #permalink

A PI of a lab where I had an intership got a paper for review on which he was a co-author. Yeah, a lot of work goes into the selection of reviewers.