Why Is Science Behind A Paywall?
May 18, 2013 8:01 AM   Subscribe

A large portion of scientific research is publicly funded. So why do only the richest consumers have access to it?
posted by reenum (62 comments total) 28 users marked this as a favorite
 
Yes. One of the things I miss most about not being engaged in university study is the access to online periodicals.
posted by BenPens at 8:24 AM on May 18, 2013


Open access is on of the major initiatives of the Creative Commons foundation. If you feel strongly about this you should send them some cash.
posted by humanfont at 8:29 AM on May 18, 2013 [3 favorites]


Because the most influential academics have made no effort to increase access to research, because they have well-funded libraries and grants, and it doesn't affect them.
posted by jb at 8:32 AM on May 18, 2013 [2 favorites]


In justifying the margins earned, the publishers point to... the complex typesetting, printing and distribution activities

Wow! I guess they really are using a business model from the 1600's!
posted by vorpal bunny at 8:33 AM on May 18, 2013 [4 favorites]


When tenure decisions include the accessibility of the research, e.g. as a weighting factor on publications, then I suspect you'll see change.
posted by idb at 8:45 AM on May 18, 2013 [4 favorites]


vorpal bunny:
In justifying the margins earned, the publishers point to... the complex typesetting, printing and distribution activities

Wow! I guess they really are using a business model from the 1600's!
I'm not sure about other fields, but Physical Review Letters just has everyone use a TeX flavor. There's virtually no additional cost that I can see with methods like this.
posted by Brian Puccio at 8:47 AM on May 18, 2013 [1 favorite]


On the other hand, you've got University libraries pushing back, as noted in the article.
posted by Existential Dread at 8:51 AM on May 18, 2013 [1 favorite]


Don't most public libraries in the US have access to these periodicals? I know mine does.
posted by Doleful Creature at 8:53 AM on May 18, 2013


Another example of that kind of thing that I've always resented: big telescopes are almost universally built using public money. But when astronomers do observations, they have exclusive rights to the results. That's particularly the case with the Hubble, but also applies to things like the VLA, and the Gemini.

If we paid for it, why don't we own it?
posted by Chocolate Pickle at 8:57 AM on May 18, 2013 [2 favorites]


Hey, if I had an extra $1,350 to publish in PLOS One, I'd be glad to.
posted by ChuraChura at 8:58 AM on May 18, 2013 [2 favorites]


A lot of researchers aren't fully competent TeX users, is the thing. Some people can produce papers with clear, legible equations without any outside assistance. A lot of people can't, or won't. (And a decent subset of the people who can't still think that they can....) So if a journal wants to be printing consistently readable papers, there is an additional cost associated with that.

It's like copyediting. Authors think they don't need one because their writing is perfect. But you send their shit to a copyeditor anyway, because you want to give the reader something as clear as possible — and because you want to save the author the embarrassment of ending up with obvious errors in print.

Now, that doesn't mean journals have to be for-pay. That's a complete non-sequitur. You can have an open-access journal, funded by grants or donations, which hires copyeditors and TeXwranglers and so on. A journal set up like that is doing a real service to readers and researchers — the readers get clean legible papers to read, the researchers get a wider audience without having to be experts on mathematical typesetting, everyone wins. But it takes a decent pile of grant or donation money to make it happen.
posted by Now there are two. There are two _______. at 8:59 AM on May 18, 2013 [2 favorites]


(Yeah, and then there's the author-pays setup in PLoS One, which I kind of feel like shouldn't even be called "open access." But I think the good guys lost that particular terminological battle, so we're stuck with it. Oh well.)
posted by Now there are two. There are two _______. at 9:05 AM on May 18, 2013 [1 favorite]


Because the most influential academics have made no effort to increase access to research, because they have well-funded libraries and grants, and it doesn't affect them.

Assuming you can get physically to the library of a state university (or even, in many cases, community colleges), you can usually get access to all of the literature they purchase. So, more people have access to this stuff than they seem to think.
posted by GenjiandProust at 9:21 AM on May 18, 2013 [1 favorite]


This is an issue I've been following mostly from the humanities side, along with MOOCs and the emergence of "alt-ac" Ph.D. programs. On the one hand, knowledge is supposed to be free (as in beer and as in speech). On the other, furthering knowledge takes money (and a whole lot of it in the hard sciences) and it takes people who have to eat, live, send their own kids to college, etc.. Who should bear the brunt of the costs of research: those learning from it or those creating it? The money, I think it's clear by now, ultimately goes to neither group.

The open-access journals sound great on the consumer side, and it's true that the current model frankly blows for getting research out there to be used or criticized in a reasonable timeframe, but the pay-to-publish model just shifts the costs from one group of people who don't think they should have to pay to another. If this model takes off, cost of publishing will have to be written into research grant applications, which seems like it will just intensify the current imbalance of whose research gets adequate funding to get out there (stuff with defense/pharmaceutical applications = yes, other stuff = no). With academic journals in particular, there's the added issue that (let's be honest) the majority of the people who will ever crack open a serious publication are academics themselves, so the pay-to-publish model is really just shifting costs from university libraries to individual research teams.
posted by oinopaponton at 9:27 AM on May 18, 2013 [1 favorite]


Assuming you can get physically to the library of a state university (or even, in many cases, community colleges), you can usually get access to all of the literature they purchase. So, more people have access to this stuff than they seem to think.

Yeah, if they buy physical copies of a journal and keep it in the open stacks, you can go in and browse.

But at least at the schools I've been associated with, you have to be a student/staff/faculty to request old volumes out of storage, or to get stuff via interlibrary loan, or to access online databases like Jstor. And the stuff that's actually sitting on a shelf in the open stacks is a fraction of the stuff that can be accessed through those other routes. So it's still pretty limiting to outsiders.
posted by Now there are two. There are two _______. at 9:33 AM on May 18, 2013 [2 favorites]


Yeah, if they buy physical copies of a journal and keep it in the open stacks, you can go in and browse.

And that's a big "if" these days, for recent copies of journals. There are also several journals that publish papers online before they reach print because of the printing backlog -- in which case you might get access eventually, but can be years later.

The pay-to-publish model is terrible though.
posted by Kutsuwamushi at 9:38 AM on May 18, 2013 [1 favorite]


That's a pretty weird way to frame this issue. I mean, I'm not accustomed to thinking of well gated research journals as consumer goods. And all the people I've known who actually do consume well gated research journals were never particularly rich, gaining access through professional or academic affiliations.

But certainly in this day and age, the well gated research journal seems a woefully obsolete model.
posted by 2N2222 at 9:40 AM on May 18, 2013


A lot of researchers aren't fully competent TeX users, is the thing.

I don't disagree with most of your comment at all, but by now, researchers should be expected to be competent TeX users, i.e. PhD programs should require candidates to pass TeX exams, because it's just not that hard and it does save everyone a bit of work. There's a whole free (speech and beer) infrastructure for making beautifully typeset shit, so, short of basically nonexistent researchers who don't have computers, there's not much excuse for anyone, in any field, not to make their papers pretty and legible before they're circulated.

(ArXiv versions of papers are very often prettier than journal versions, although copyediting will be invaluable forever.) Journals want typesetting and appearance to be uniform, not necessarily better than what authors produce. This is reasonable, but authors do a lot of this work, anyway, by imposing journal-dictated TeXmanship on their own articles after acceptance. This is itself fine, but I have trouble accepting typesetting as a justification for journal pricing. (Which I think is part of your point, too.)

Typesetting is different from copyediting: copyeditors perform an extremely important function whose importance is orthogonal to the particular publishing model being used (and I would like to buy all of them a beer). Typesetting is different; one could image peer-reviewed sections of the ArXiv with, eventually, the same function as journals. In such a context, copyediting is no less necessary (i.e. there would probably have to be some centralized means of doing this), but there's no need for a central authority saying "You have to left-justify your displayed equations even though it looks ridiculous", since there's no good reason for uniform typesetting (as long as everyone is TeX-literate) in such a situation.

(Maybe this comment is math-centric; I don't know anyone in my field, regardless of age, who's not TeX-competent. Even loads of undergrads take notes in classes with a TeX editor.)
posted by kengraham at 9:43 AM on May 18, 2013 [3 favorites]


Yeah, if they buy physical copies of a journal and keep it in the open stacks, you can go in and browse.

But at least at the schools I've been associated with, you have to be a student/staff/faculty to request old volumes out of storage, or to get stuff via interlibrary loan, or to access online databases like Jstor.


Were you associated with public institutions? Pretty much every public institution (and the one private) I've worked or gone to school at (four in three different states), had, as part of their mission, giving some access to their electronic resources. Obviously, students, faculty, and staff, as primary users, get precedence, but there is access.

Given that Interlibrary Loan has significant costs associated with it, unaffiliated people should not expect that. On the other hand, your local public library can usually do ILL, so there is that.
posted by GenjiandProust at 9:51 AM on May 18, 2013


kengraham: "I don't disagree with most of your comment at all, but by now, researchers should be expected to be competent TeX users ... (Maybe this comment is math-centric; I don't know anyone in my field, regardless of age, who's not TeX-competent. Even loads of undergrads take notes in classes with a TeX editor."

This comment (and the article!) are both STEM-centric. There are a ton of fields outside the STEM real in which research and publishing go on, and if you asked us to start learning TeX you'd have a revolution on your hands.
posted by barnacles at 9:57 AM on May 18, 2013 [3 favorites]


To defend this a little, most of the "pay to publish" journals offer a waiver for authors wo can't pay. And even in non OA journals, you usually have to pay for e.g. each color figure. Finally, at least for people funded by the NIH, publishing costs are an incredibly tiny part of the budget, though for labs with primarily NSF or other funding sources and for academics outside of STEM I can imagine it could be spendier.
posted by en forme de poire at 10:06 AM on May 18, 2013


I've published a couple of papers in PLos ONE recently, and I write publication costs into my grant proposals. I'd rather pay a little more to publish in PLoS ONE and have the resulting paper free available to anyone who wants it (and the paper metrics can be humbling when you see that only about 15% of the people who find the paper download the PDF) than pay half as much for it to be embargoed for a year.
posted by wintermind at 10:06 AM on May 18, 2013


Open access is on of the major initiatives of the Creative Commons foundation. If you feel strongly about this you should send them some cash.

Thanks (I work at CC). We're also trying to push the discussion beyond mere access toward licensing that allows reuse. On that front, the White House directive does require that agencies build:
an approach for optimizing search, archival, and dissemination features that encourages innovation in accessibility and interoperability, while ensuring long-term stewardship of the results of federally funded research;
To me, that means (or should mean, anyway) open licensing.

This comment (and the article!) are both STEM-centric.

Yup. And too much of the movement is too, in my opinion. The argument for open access to publicly funded research is just as strong in the humanities as it is in science.
posted by roll truck roll at 10:11 AM on May 18, 2013 [1 favorite]


I don't disagree with most of your comment at all, but by now, researchers should be expected to be competent TeX users

I agree, for physical scientists and mathematicians. Biology and medicine… work differently.

Most submissions to most biology journals occur as PDFs of double-spaced Word (or equivalent, but Word) documents, with citations managed by Endnote, a pure creation of Satan, and with images composited in, I shit you not, Powerpoint. Attempting to get a legacy paper I've inherited into shape has been a nightmare, for bibliography and format reasons.

Non-computational biologists are just not expected to know sophisticated computing tools. Were I to give a TeX document to my august and skilled scientific advisor, she would ask me to get it to her in Word format so she could edit it. There are a lot of potential advantages to embracing good tools, but they are not as strong as in the physical sciences, where frequent need for equation formatting and the availability and use of (at the time) powerful computing resources forced everyone to learn to deal in the 80s and 90s.

I will be lucky if I can contribute to the transition by introducing Illustrator, which at least will output our line art in an appropriate format and allow appropriate figure alignment and spacing. So I do appreciate the work of the typesetters at the journals, who have their work cut out for them. Granted, it's still not worth the huge costs, but it is an important component of the journals' service. The $1350 or whatever PLoS charges is likely a fair representation of costs with a little program continuity thrown in.

Alongside this, biologists have no Arxiv. In fact, in journals following the Ingelfinger rule as it is normally done in medicine, Arxiv publication would invalidate a publication for consideration. There are revenue and ethical reasons for avoiding preprints in medicine (all studies should undergo peer review before release of interpreted results), though less so for biology in general.
posted by monocyte at 10:16 AM on May 18, 2013 [7 favorites]


en forme de poire:To defend this a little, most of the "pay to publish" journals offer a waiver for authors wo can't pay. And even in non OA journals, you usually have to pay for e.g. each color figure. Finally, at least for people funded by the NIH, publishing costs are an incredibly tiny part of the budget, though for labs with primarily NSF or other funding sources and for academics outside of STEM I can imagine it could be spendier.

Yeah, sorry, I'm sure that my humanities affiliation is skewing my own perception of the financial side a bit. Just to throw some numbers in here, the NEH had an budget of $167 million in 2010 compared to the NIH's $30.9 billion (thanks Wikipedia), and that's for all the humanities. An extra $1000-$2000 really matters in the humanities, especially in fields where it's unusual to have more than one author on an article.
posted by oinopaponton at 10:22 AM on May 18, 2013


There are a ton of fields outside the STEM real in which research and publishing go on, and if you asked us to start learning TeX you'd have a revolution on your hands.

But by the same logic, converting a manuscript from word or plaintext or whatever to whatever their typesetting software is should be simple enough for a few scripts and someone's brief attention to fixing the few inevitable fuckups.

I've published a couple of papers in PLos ONE recently, and I write publication costs into my grant proposals.

This is one of those things that's just not relevant to disciplines where the overwhelming majority of research is not supported by any grant.

Given that PLoS bases publication charges on where the primary funding for the paper comes from, I wonder how they deal with research that is simply not funded by anyone?
posted by ROU_Xenophobe at 10:24 AM on May 18, 2013 [2 favorites]


The biology department where my girlfriend got her PhD had also basically never heard of TeX, which blew my mind. Everyone used Microsoft Word. I kept pointing out that the time spent to learn TeX would be repaid the first time she had to change layout or citation style for a different journal, but MSWord was too entrenched in the department for a lowly grad student to buck the trend.

(On preview: what monocyte said.)

Hasn't the Obama administration been pushing for open access to government-funded research this year? e.g. [pdf]?
posted by hattifattener at 10:28 AM on May 18, 2013


ROU_Xenophobe, I think that you can request that PLoS journals waive the publication fee for various reasons, including financial hardship. This is far from perfect, though, and I recognize that my ability to recover publication costs in grant proposals may be STEM-centric, but that's what I know.
posted by wintermind at 10:40 AM on May 18, 2013


I don't quite get the hassle. I haven't used TeX in almost 30 years. I submitted a paper fairly recently (which was rejected :-D) to a computer music conference, and they gave us a template document in a bunch of formats where you simply had to replace their text with your own. I did it and got it right the first time - I don't even remember what program I used, it could have been Word for all I remember.

I assure you, computer music people don't have oodles of money. If they can do it, these huge journals can definitely do it.
posted by lupus_yonderboy at 11:07 AM on May 18, 2013 [1 favorite]


Fron the Deutsche Bank report cited in the article:
We believe the [Elsevier] adds relatively little value to the publishing process. We are not attempting to dismiss what 7,000 people at [Elsevier] do for a living. We are simply observing that if the process really were as complex, costly and value-added as the publishers protest that it is, 40% margins wouldn’t be available.

The CEO of Elsevier makes $2 million per year, by selling research paid for by taxpayers. Elsevier also sells research that is available for free.
posted by elgilito at 11:22 AM on May 18, 2013 [7 favorites]


Count me as another library staffer with almost un-fettered public access to print and online periodicals. We don't offer ILL for anyone but current students and faculty because it is expensive and there are concerns about who is liable for damaged materials, but we also send out a lot of articles to public libraries. Many state schools also do offer library cards and access to local residents.

Not that it makes up for those access roadblocks and staggering costs, but just as a heads up that it's worth a try to ask around. And if you publish articles, please do consider putting them in an institutional depository or reserving your rights to post a copy or per-print online.
posted by jetlagaddict at 11:49 AM on May 18, 2013 [1 favorite]



Alongside this, biologists have no Arxiv. In fact, in journals following the Ingelfinger rule as it is normally done in medicine, Arxiv publication would invalidate a publication for consideration. There are revenue and ethical reasons for avoiding preprints in medicine (all studies should undergo peer review before release of interpreted results), though less so for biology in general.


In physical science and math, you still have to submit your papers to peer-reviewed journals, and something equivalent to the Ingelfinger rule is usually enshrined in copyright forms -- they won't publish something still under consideration by another journal -- although with exceptions made for circulating preprints (they are picky about which versions you circulate, usually; leave some minor typos in the ArXiv version).

It's true that a big difference is that it's generally much easier to independently verify non-experimental results before citing something than it is to reproduce research, so something like math lends itself unusually well to the existence of the ArXiv. However, I'm curious what the "revenue and ethical reasons" for avoiding preprints are. Are the revenue reasons relevant in publicly funded research? Are there good reasons why there shouldn't be easily accessible preprints resulting from publicly funded medical and biological research?

Also, to what extent do preprint servers exist in the humanities? It seems like an ArXiv-type thing would be ideal in this context.

This comment (and the article!) are both STEM-centric. There are a ton of fields outside the STEM real in which research and publishing go on, and if you asked us to start learning TeX you'd have a revolution on your hands.

Folks in the humanities seem threatened by societal pressures that elevate profitability above all else, which is why people in the humanities should think about eschewing things like Word (or many of its equivalents), at least for symbolic reasons. TeX is not a specialized scientific typesetting tool (or a "sophisticated computing tool", as was suggested above; my partner is running free TeX software on an ancient $100 used netbook in the other room right now). It's for typesetting anything beautifully, with a high degree of control. It's also way less difficult to acquire TeX facility than it is to, say, do research in any field. Assembling your CV in TeX, from scratch, will take a few hours and teach you much of what you need to know.

Actually, disciplinary norms that require things like Word documents probably contribute to public money being spent on software licenses unnecessarily, which is a similar problem to that of public university libraries having to buy ludicrous bundles of journals from Elsevier or whatever. It's inappropriate corporate incursion into a public institution, and the resulting expense doesn't help anyone make a case for research funding to a public that is increasingly batshitinsane about research (especially humanities research).
posted by kengraham at 12:13 PM on May 18, 2013 [2 favorites]


I know it's way up the thread, but this is wrong and I can't just leave it here:

Chocolate Pickle: Another example of that kind of thing that I've always resented: big telescopes are almost universally built using public money. But when astronomers do observations, they have exclusive rights to the results. That's particularly the case with the Hubble, but also applies to things like the VLA, and the Gemini.

If we paid for it, why don't we own it?


You didn't pay for most of it, but you do own it anyway.

1) The NSF/NOAO/NRAO/NASA/etc. contribute some, but most big ground telescopes are also built with funds kicked in by universities, private foundations, museums, and public agencies in other countries, in a public-private partnership. A breakdown for optical-IR telescopes with public involvement is here; note that it does not include some of the largest telescopes like Subaru, the VLT telescopes, and GTC, for lack of significant US public involvement.

2) Most data on these telescopes is made public after some embargo period. Gemini, in your example, gives 18 months; VLA and Hubble give a year. I don't see this as unreasonable---if you planned and carried out the observations, why wouldn't you get a crack at the data before your competition?

3) Moreover, that data is publicly available to anyone, generally from archives hosted by the individual observatories. The VLA and Gemini archives can be queried from here and here, respectively; the data for many space astronomy missions, including Hubble, is available from MAST.
posted by Upton O'Good at 12:19 PM on May 18, 2013 [2 favorites]


With all due respect, I do not understand how anyone could write:

There are a ton of fields outside the STEM real in which research and publishing go on, and if you asked us to start learning TeX you'd have a revolution on your hands

without embarrassment. Folks in STEM fields didn't spring fully formed from the forehead of Gauss with genetic memory of LaTeX-- we learned it. And you can too.

Saying that you can't learn LaTeX is tantamount to saying that you can't follow step-by-step instructions. The first 90% of LaTeX, which is as much as most people ever use, can be learned in an hour, tops. Environments, sections, figures, BibTeX; that's really all you need to know to typeset 99% of papers. I don't understand how anyone could say with a straight face that they're capable of doing research, i.e. uncovering heretofore unknown truths about the world, but incapable of understanding what \section{Introduction} means.

C'mon, people.

PS: Also, bonus points for accusations of being "STEM-centric". That's like Metafilter bingo.
posted by lambdaphage at 12:55 PM on May 18, 2013 [2 favorites]


Just pointing out: Access to a paper is pretty useless if you don't have the money to do anything with it. Duplicating what is in a chem paper is expensive. If you can't afford the paper then you REALLY can't afford to do anything useful with it.

I'm not saying the fees shouldn't be more reasonable, and there shouldn't be more respected publishers to drive the price down, or perhaps price regulation down to something sane, but suppose the price was dropped down to something sane; I don't see why it has to be free, since if you can't afford the journal, there is no damn way you can afford the reagents and lab space to USE anything in it.

Alternatively, why not set up a barter system? Instead of referring papers for free, each time a prof at the uni referees a paper for that journal, the cost of the subscription goes down by a certain amount. This would also give us a GREAT way to fund verification of results, which is something that has always been problematic. Duplicate an experiment independently, get a huge cut off the journals fee.
posted by Canageek at 1:10 PM on May 18, 2013 [1 favorite]


Honestly, though, if your papers aren't full of equations, diagrams or complicated multilingual annotations, there's no reason you should feel the need to learn (La)TeX. Hell — even if your papers are full of any of that, if you're willing to put in the work to make that shit readable in Word, more power to you.

The real issue here isn't TeX vs. Word. It's "legible and comprehensible" vs. "illegible and incomprehensible."

But so here's the thing. Putting out a legible and comprehensible paper is harder in disciplines where there's a lot of special notation (and that's not just STEM, it also includes linguistics, some corners of textual criticism, music theory, etc. etc. etc.). In all those fields, there's extra work involved in making sure that all the notations are correct. And in all those fields, there are some authors who are willing and able to do that extra work themselves, and some who aren't. Which means that in all those fields, typesetting wonks (whatever software they use) are doing genuinely useful work that wouldn't get done consistently in a world where everything just went up on ArXiv and that was it.

(Physics might be an exception. I don't read physics papers, so I don't know. Math certainly is not an exception. I have read math preprints that are painfully illegible, prepared by mathematicians who could have made it legible but didn't bother to. Math journal editors do a real service to the field by either sprucing that shit up themselves or bouncing it back to the author with a list of demands.)
posted by Now there are two. There are two _______. at 1:12 PM on May 18, 2013 [1 favorite]


It's ridiculous to assert that people doing university-level research can't learn TeX fairly easily, I agree. But what's mostly going on, I think, is different: it's that people don't want to learn another publishing system. And even though I think it costs them a lot of hassle in the long run, I completely understand this position. For lots of researchers, writing up results for publication, and getting the fiddly journal-specific formatting right, is one of the least fun parts of their occupation. It's not what they want to be doing; what they want to be doing is more research! So of course they don't want to put in the time to learn some new system, even if they're told that a little investment now will make everything easier next time— they've heard that claim before (every time they had to upgrade to a new version of MSWord, for example, and how often did that claim actually pan out?)
posted by hattifattener at 1:13 PM on May 18, 2013 [1 favorite]


PS: Also, bonus points for accusations of being "STEM-centric". That's like Metafilter bingo

You're writing that like it's a stupid, meaningless distinction. But the difference between including publication fees in your grant proposal and paying them out of your pocket is, I assure you, less than subtle.

Folks in STEM fields didn't spring fully formed from the forehead of Gauss with genetic memory of LaTeX

There's little reason for someone writing in history or literature, or other fields where your articles are likely to be just text, to learn TeX. A publisher could just feed the Word/plaintext to a script that would convert it to tex, troff, adobe-whatever, or whatever it is the publisher/typesetter uses.

(I think STEM-types are also overestimating the extent to which publishers and typesetters use LaTeX for production)
posted by ROU_Xenophobe at 1:41 PM on May 18, 2013 [3 favorites]


For lots of researchers, writing up results for publication, and getting the fiddly journal-specific formatting right, is one of the least fun parts of their occupation. It's not what they want to be doing; what they want to be doing is more research!

So true!! You were hired to do RESEARCH-this concept that everybody is expected to be good at everything has to stop! Comparably-and arguably a derail-why is the IRB committee sending the consent back for the sixth time to the exact same researcher for a rewrite-they think that this time the Russian native speaker is going to figure out what a 6th grade USA reading level is? That's like asking an education major to do my statistical analysis! When trainees ask for advice in their job search, I never mention work-life balance, salary, blah, blah. Instead, I recommend they look for the place that allows them to use their skill set and offers support for the ones they don't have. For instance, does it offer good IT support, professional assistance with posters/papers beyond mere printshop service, finance people to assist with budgeting, etc. The right people, the right tools = the right job-when did this get forgotten!
posted by beaning at 2:28 PM on May 18, 2013 [4 favorites]


You're writing that like it's a stupid, meaningless distinction. But the difference between including publication fees in your grant proposal and paying them out of your pocket is, I assure you, less than subtle.

I sincerely don't understand this point. If you can't ask for pub fees in the grant proposal, doesn't the cost of manuscript preparation become more pressing? That is, shouldn't folks in the humanities care even more than scientists about using free, professional-quality typesetting? Sentiments about resisting corporate domination, the system and the man seem widespread within the humanities, yet when actually shown a concrete and practical way to dump some of the academic bosses of their backs, the same scholars become, in the phrase of one famous humanist, "monsters of incuriosity".

The central point here is that most journals have to retain someone for typesetting; if those journals were to simply publish a TeX template instead, that problem would become completely trivial. Any point in the publication process in which human intelligence performs a machine-amenable function is a waste of research money, and that observation is invariant with respect to academic field.

There's little reason for someone writing in history or literature, or other fields where your articles are likely to be just text, to learn TeX. A publisher could just feed the Word/plaintext to a script that would convert it to tex, troff, adobe-whatever, or whatever it is the publisher/typesetter uses.

I thought the entire point of this discussion was to identify the middlemen whose existence justified charging the public for research that they paid for. The problem with such an hypothetical script is that .docx is still a quasi-proprietary format (to say nothing of .doc) and the viability of such a script depends on Microsoft not doing to XML what they are singularly famous for doing to other formats. I'd love to see it happen, but at present converting .docx to .tex convincingly in any less time than it would have taken to typeset the original document seems to be a crapshoot.

Though the advantages of TeX are unignorable if you have to typeset a single equation, they are still substantial even if your paper is pure structured text. I actually started using TeX when I was an undergraduate philosophy major, and have never looked back.

it's that people don't want to learn another publishing system

It's not as though I'm asking academics to learn something else in addition to what they already use; I'm asking them to renounce their old ways entirely. As an academic, you'll have to learn some system or another for composing papers, and it might as well be a free and technically superior system rather than a proprietary, inferior one. Kieran Healy (a sociologist!) has some remarks worth reading about this.

...getting the fiddly journal-specific formatting right...
The entire idea of TeX is to divorce form from content. Take the same document, structured in .docx and .tex, and convert them from one journal's style requirements to another's. Tell me how long it takes in each format.
posted by lambdaphage at 2:44 PM on May 18, 2013 [1 favorite]


TeX is okay for programmers or scientists in fields that involve programming (physics, math). But it is a difficult, fussy and flaky programming language for non-programmers. Sorry, it just is. If you want to complain about why this is true, that's fine. But it is true, nonetheless. If you lament this fact, then apply your programming skills to an open-sourced word processor that presents a fully-featured (and reliable) Microsoft Word-like front end, which uses TeX well in the background. That is how you will win converts. Up to you.
posted by Blazecock Pileon at 2:58 PM on May 18, 2013 [1 favorite]


Yes, I know, lambdaphage. That's precisely my point: it's going to save hassle in the long (or even medium-short) run to switch to TeX. What I'm trying to say is not that, in a rational analysis, it makes sense to stick with MSWord. What I'm trying to explain is why people don't even want to make that analysis. Because they don't see the writing and especially the preparation as a principal part of their occupation as a scientist.

So true!! You were hired to do RESEARCH-this concept that everybody is expected to be good at everything has to stop!

I think this goes too far in the other direction, though. A very central part of science is communicating your discoveries to other scientists and this means being able to write and present clearly. And that means not just being able to string words together clearly and grammatically, not just being able to put together a comprehensible figure or present an engaging talk, but also being able to use the necessary tools for doing those things.

then apply your programming skills to an open-sourced word processor that presents a fully-featured (and reliable) Microsoft Word-like front end

Some do exist (eg LyX or whatever its current incarnation is). But presenting a "Microsoft Word-like front end" immediately loses a lot of the benefit of a TeX-like system. As an analogy … have you ever worked with documents from someone who learned all their skills in the typewriter era? Have you noticed how they do things that make everything harder, like using spaces to indent paragraphs, hard line returns for every single line break, and more spaces for tabular alignment, instead of just telling Word what to do and letting it do that? The difference between Word and TeX is like that, but more so.
posted by hattifattener at 3:04 PM on May 18, 2013


Also I kind of think that the TeX and typesetting argument, while dear to my heart and not totally irrelevant to the FPP topic, is a derail here. Sorry for perpetuating it.
posted by hattifattener at 3:07 PM on May 18, 2013


LyX as a word processor is pretty awesome, once set up (though the process is still fiddly compared to Word or Libre).

Another major issue, though, is that many journals in, e.g., biology require manuscripts to be submitted as Word files. Even a .pdf of a LaTeX document isn't acceptable for the final version. You can convert to rtf or html or whatever, and I've done that, but those tools are still pretty imperfect, so you end up wasting time cleaning it up in Word on the other end. (And omfg don't get me started on the eldritch abomination that is Endnote.)
posted by en forme de poire at 3:13 PM on May 18, 2013


I have also had inexplicable experiences in biology journals where people send back proofs and all of a sudden the equations are nonsense - like, natural log has been changed to the word "In" and parentheses and brackets have been dropped all over the place. Uh, wtf was the point of me sending you a digital copy?
posted by en forme de poire at 3:17 PM on May 18, 2013


To get back to the FPP I think one big problem is that while there are a few OA journals with that compare in prestige to, e.g., Nat/Sci/Cell, or Nat Genet, or PNAS, there are just way fewer of them. So you can't play this game of getting rejected after mediocre reviews at Science and then turning around, making a few tweaks, and trying your luck at Nature and/or Cell. If you don't get into PLoS Biology (or I guess eLife these days), your next option is going to be a significant ways down the impact ladder. Not every PI plays this game, for sure, but if you are trying to maximize the conventional measures of impact it is one strategy.

This is unfortunate because there are major rewards for moving to online-only OA journals. A big one that the authors mention, and one that's going to keep getting more important IMHO, is that the data can live together with the paper in an online format. You may even be able to require scientists to submit raw data, either to the journal or to a repository (like GEO). Someone I know actually submitted their entire lab notebook as supplementary information, essentially making literally everything they did part of the permanent scientific record, which is both awesome and super brave. It would be amazing if that became more common (though honestly I would be a little terrified - and I would have to make my notebooks a lot clearer from the beginning if I ever wanted to do this!).
posted by en forme de poire at 3:35 PM on May 18, 2013 [2 favorites]


I sincerely don't understand this point. If you can't ask for pub fees in the grant proposal, doesn't the cost of manuscript preparation become more pressing?

No, because the cost of manuscript preparation to the researcher is $0 + a little time.

The problem with such an hypothetical script is that .docx is still a quasi-proprietary format

That, for pieces that are just text, can be converted into a stream of ascii text by copying it into a text editor.

Take the same document, structured in .docx and .tex, and convert them from one journal's style requirements to another's. Tell me how long it takes in each format.

Look, I write all my stuff in LaTeX unless I have coauthor problems, but this particular benefit of TeX is oversold. How long will it take to convert from one journal's format to another? Depends. Do they both use or accept the same citation and bibliographic packages, or do I have to go through and fuck with all my \citeasnoun and \citeyear and so on? (I've actually given up on this and just put everything in manually followed by a \nocite to the appropriate bib entry). Do either of them use a custom .cls file with weirdo environments that want things to be different than vanilla LaTeX?

You can convert to rtf or html or whatever, and I've done that, but those tools are still pretty imperfect, so you end up wasting time cleaning it up in Word on the other end.

I get that too, but at that point it's already accepted. So I just take whatever latex2rtf gave me and send them that. Here, have some ugly shit. You deal with it.
posted by ROU_Xenophobe at 4:25 PM on May 18, 2013 [2 favorites]


Look, I write all my stuff in LaTeX unless I have coauthor problems, but this particular benefit of TeX is oversold.

Yep. L-O-fucking-L at LaTeX being portable or at all an instance of well-separated concerns. Is it the most powerful member of its class? Absolutely. Is it elegant or modular or predictable? No, it's an ugly and anachronistic pile of shit with insane and inconsistent syntax and it's stuffed with environment-dependent gotchas. Anybody that relates to it with an emotion other than resignation is clearly suffering from Stockholm Syndrome, because any programmer that cares about expressiveness would have to forget themselves to think that LaTeX is well-designed. I get why it's used, but I don't get why it's loved.
posted by invitapriore at 8:31 PM on May 18, 2013 [4 favorites]


Yeah, the main things to like about LaTeX are math mode, pretty enough default settings, and being plain text (no risk of Word suddenly garbling your document because you added too many figures). Everything else is definitely "least worst" rather than "awesome." Journal articles follow such a stereotyped form, it seems like another plain text format could easily replace LaTeX for most uses - like if something like Markdown had clean-looking figure, table, citation/bib, and math mode support, plus maybe an easy way of leaving comments/notes, that would about cover it. I can dream, I guess.
posted by en forme de poire at 9:14 PM on May 18, 2013 [3 favorites]


http://thepiratebay.sx/search/peer%20reviewed/0/99/0

Oh hey isn't that interesting there's nothing there

What happens when the largest group of fuck-shit-up crowd-sourcers, people who are disproportionately college students with access to these paywalled articles, want to protest the prosecution of Schwarz in a potentially illegal manner? oh hey DDOS that's useful.

Nobody cares. The only people who want access to these articles are academics (who know how to access, who have access). Well, and a handful of people that are probably thought of as tin-foil-hatters.
posted by nathan v at 10:41 PM on May 18, 2013


Well, not all academics have access. Yeah, if you're HYP, you can just cough up whatever Elsevier asks for, but if you're a community college with barely a photocopying budget, not so much.

As far as end-running around paywalls, I think you're just looking in the wrong venue - if you look you can find entire boards where people ask for .pdfs of particular papers to be uploaded, they're just not torrenting them on the Pirate Bay.
posted by en forme de poire at 11:00 PM on May 18, 2013


(hey look, there are some journal torrents! Not very extensive tho.)
posted by en forme de poire at 11:08 PM on May 18, 2013


Yeah, I'm sorry-- I meant that academics have access, if not to good libraries, then to knowledge that leads them to access. Community college professors have done the rigamarole. They realize that if you need an article, there are ways to get it.

That's not quite true of the amateurs. They don't have access. If one of your parents wants to double check the studies justifying some prescription, they probably have no way to do that. (They should have a way to do that.)

What bothers me is that the solution for this is so simple. It doesn't require boycotting Elselvier. It doesn't require a presidential order. All it takes is the accumulation of some "direct action."
posted by nathan v at 1:20 AM on May 19, 2013


It has personally taken me days to get a single article already in LaTeX to the specs on a new journal because of package / style / bib requirements. Margins without the geometry package! Table spacing! I have flashbacks about getting the formatting done on my dissertation. Even my tex-savy colleagues do not like editing tex + diff for collaboration; we invariably end up scribbling on the pdfs.
posted by a robot made out of meat at 4:40 AM on May 19, 2013


Upon reflection, my status as someone who comes home after a few drinks on a Friday evening to rant about LaTeX in the sciences on the internet fills me with a funny sort of pride. Anyway, no more from me on that tangent.
posted by invitapriore at 9:10 AM on May 19, 2013 [2 favorites]


Just pointing out: Access to a paper is pretty useless if you don't have the money to do anything with it. Duplicating what is in a chem paper is expensive. If you can't afford the paper then you REALLY can't afford to do anything useful with it.

I think it's kind of silly to suggest that the only reason you'd pick up a paper is to find the method/synthesis/whatever that you need. People can read all kinds of papers for all kinds of reasons -- sometimes just for fun, and that's okay too. They paid for it. They should be able to.

And when we're talking about medical research, people want to be able to read papers because those papers are directly relevant to their own health. I've prepped for important doctors' appointments with pubmed sessions more than once in my life. More people should have that option.
posted by gerstle at 1:00 PM on May 19, 2013 [3 favorites]


nathan v Check out #icanhazpdf or #icanhaspdf on twitter. A lot of people will ask for a paper then someone will discreetly slip them a copy under the table. Not a big mass share, but people helping people out.
posted by Canageek at 7:51 PM on May 19, 2013 [2 favorites]


gerstle: That seems like a really minor use. The main point of papers is for other scientists to use them to do research. I think it would be cool if they'd offer a reasonable single-person subscription option ($30 a year like a normal magazine?) but my main concern is for people at standard, 20,000 person universities and up being able to use papers to further science. Everything else is gravy. The commercial entities? Feh. They make money, they can afford it.
posted by Canageek at 7:54 PM on May 19, 2013


If universities and industry search committees stopped relying so heavily on the impact factor of publications in applicants' records, then researchers would stop feeling the pressure to publish in "high-impact" journals. But it goes beyond this. Research institutions want researchers with good publications because these researchers are better able to acquire funding in competitive review processes. Research institutions care about the money a researcher can acquire, 40-60% of which goes to institutional "overhead". So really, it is the granting process which needs to be reformed. Once it focuses more on quality of work rather than quality of publication, everything else will start to sort itself out.
posted by FUD at 9:59 PM on May 19, 2013


"Really minor use" in frequency, yes. Really minor use in importance, no.

A single-person subscription rate wouldn't really solve this problem. When you need to know something you read ten papers on the subject. They're all in different journals. You'd still be spending $300 every time you had an important medical decision to make.

I admit that probably a small minority of patients would ever do the research themselves. But the idea that it's out there, that they paid for it, and that they can't see it is pretty unconscionable.

And again, I would also stick up for the right of the average citizen to read whatever they want just for fun. I certainly read things that are far afield of what I'm actually trained to understand, and enjoy it. I'm sure other people would like to be able to do that as well.
posted by gerstle at 5:06 AM on May 20, 2013 [4 favorites]


FUD: At the same time, I want to publish my research somewhere it has a chance of being read. Demphasising impact factors is good, but I think I'm more likely to have someone pay attention to my research in JACS, Inorg. Chem. or Organomettalics then The Journal of Obscure Science.

gerstle: I disagree. If I want to read a novel then I am fine paying $10 to pick up the paperback. If I want to read National Geographic then I get a subscription. The problem is that the costs of a single article are crazy ($30 for 48 hours? That is three novels!) and institutional access adds up super fast. I think that they need to get the fees to a reasonable level, and stop double dipping (Pay to publish OR pay to read. Pick one dammit).
posted by Canageek at 9:22 AM on May 22, 2013




« Older The Myth of Nazi Efficiency   |   Rubber Ducky, you're my very best friend, it's... Newer »


This thread has been archived and is closed to new comments