Slashdot Mirror
Paywalled Science Journals Under Fire Again
The Real Dr John sends this report from The Guardian:
Emeritus professor Stephen Leeder was sacked by the Medical Journal of Australia (MJA) in April after challenging a decision to outsource some of the journal's functions to the world's biggest scientific publisher, Elsevier. This month he will address a symposium at the State Library of NSW where academics will discuss how to fight what they describe as the commodification of knowledge. Alex Holcombe, an associate professor of psychology who will also be presenting at the symposium, said the business model of some of the major academic publishers was more profitable than owning a gold mine. Some of the 1,600 titles published by Elsevier charged institutions more than $19,000 for an annual subscription to just one journal. The Springer group, which publishes more than 2,000 titles, charges more than $21,000 for access to some of its titles. "The mining giant Rio Tinto has a profit margin of about 23%," Holcombe said. "Elsevier consistently comes in at around 37%. Open access publishing is catching on, but it requires researchers to pay up to $3000 to get a single open access article published. What other options are there for making scientific publications available to everyone?
You make it sound as if some journals employ experts on the topic of the journal. I don't believe that is true. The peer reviewers are volunteers who also publish in the field.
I had a meeting with the director of research for a university once, regarding career issues... publication and citation record on Web of Science or Scopus concerning a certain type of publication (refereed, internationally recognized journals) were the only two things considered. The only. There is good reason for this. There is also good reasons for the complaints about "paywalled" journals. But when the existance of my job depends on the traditional publishing model, you bet I'm motivated to stick with it. Loss of my job -- should I choose to stay in the same field of work -- involves finding another job in (at a minimum) a different city, if not (more likely) a different country.
When you apply for funding, you don't actually get paid any more than you do before. Your salary is set by the instutition, and is fixed. This is because research is unpredictable. So when in the media (or even Slashdot) there are comments about how scientists are in it for the money... wtf?!? Extra funding money, if not for needed equipment, at best goes towards supporting students that get paid barely above minimum wage.
You want to create a new publishing model, you go for it. I'll cheer from the sidelines until it's been successful for a decade or two.
Obviously you need a scientist to go over your work but I think they might lower the costs if they can make the papers easier to read or potentially release them as a series.
The peer reviewing is done for free. As in the people who review do it because it's part of one's duties as a member of the community. Making papers easier to read would certainly be nice (better journals tend to have better written papers). Not sure what you mean by "release as a series".
I wouldn't want yet more stuff to review to be honest. I already am atmy limit and have to turn things down occasionally due to lack of time.
The other problem with "easier to read" is few people set out to write a badly written paper. A few do---those who want to obscure something---but most don't. The thing is scientists are on the whole scientists. Writing, and GOOD writing is a whole other career and one that scientists ocasionally master in addition to their own, but often just get by.
Writing a good paper actually has a lot in common with writing fiction. Even though it's about facts, there still has to be a narrative flow guiding the reader towards the author's ideas. An additional problem is that unlike fiction where you alter the world to fit the narritive, with science you have to bend the narritive to fit the immutable world, which can make things a bit confusing when you have s triangled of dependencies and you have to start somewhere.
I'm not a writer, and I was like many only a so-so scientific writer. I've seen a few truly excellent
There's also the problem of different styles in that some people think in different ways. And by the time it gets to a really cross disciplinary paper, not one author will understand the work in its entirety, never mind reviewers.
Oh and there's ALSO the problem that people don't seem to knoe what a literature review is for. For many years didn't, thinking like many that it's a bunch of citations you need because REASONS. They I read one by a skilled writer and I was enlightened.
A lot of your suggestions are about better tech. That would help a little, but not all that much. Citations are essentially old-fashioned hyperlinks. Modern systems can hyperlink the citation to the place in the reference list and from that to the actual paper. It would be nice if more places did that and it would be nicer if they supported richer hyperlinking as you suggest. But ultimately that's not the main problem. Reviewers tend to be familiar with the field and so often I don't have to go and read much if anything extra when reading a paper. It's the big blob in the middle describing the technique that needs improvement.
If papers could be made easier to read, that would be great, but I don't see an obvious way to do it.
Data can be provided, and some journals require it, but many don't. The problems with releasing working algorithms is that you have to be able to distribute working portable code to other people with a moderately sane user experience. This is something I actually can do and have done. It's a *lot* of work single handed (and I don't support Macs, and only partially support Windows). but I'm something of a hacker and I've got considerable industry experience. I also care about it and do it even though I've left academia for the most part. Most people simply don't have the training to do such things. I've encountered released code that I and several other good people have been unable to make run even after expending considerable effort.
One thing I learned as well is that it's not always useful. I wrote a paper with a good number of citations (650 so far and climbing at the rate of a few hundred a year which in the academic world is a very good) and I made a big effort to release the code, keep it up to date and fix bugs. The only thing people use is the code bundled with it for a much earlier, simpler system: to my knowledge not a single person has used the code I released specifically for the paper.
Other bits of code/whole programs I've
SJW n. One who posts facts.
The cost comment was in reference to the complaint that OPEN journals cost upwards of 3000 dollars to get a single paper published.
That number is highly variable. There are plenty of open-access journals that cost only a few hundred dollars per article for publication, not several thousand. See here, for example. As that article notes, quite a few big open-access publishers admit that their internal costs for publishing are around a few hundred dollars per article.
As to the notion that the scientists are not being paid to audit the papers, then why do the paid journals only have a profit margin of 38 percent?
If I'm getting the papers for free, people are auditing the papers for free, my cost structure is a website, and people are paying me 20k for a subscription to access the journals... then why is my profit margin so low?
I think you are significantly underestimating the amount of administrative work that goes on in sustaining a publishing operation, even an online one. See the first big chart in the link above, which breaks down costs percentage-wise in publication, and see the amount needed for "administering peer review; editing; proofreading; typesetting; graphics; quality assurance... covers; indexes and editorial; rights management; sales and payments; printing and delivery; online user management; marketing and communications; helpdesk; online hosting... " etc.
There's a lot of random overhead required.
The profit margin if what you're saying is true should be closer to 97~99 percent basically meaning the journal has a small staff that matches X scientists with Y papers... and then whatever the web hosting costs which in any of these businesses is basically nothing.
Um... yeah... again, see above.
That said, it's clear that something fishy is going on with commercial publishers. As this article notes, the for-profit publishers seem to charge 2-3 times as much as non-profits, so it seems like they should be making more than 38% profit. I don't know what the explanation is there, other than that I imagine for-profit companies pay upper-level administrators more.
Anyhow, wherever that excess money is going, your weird conspiracy theory that there's some sort of "kickback" scheme to scientists or reviewers or universities just isn't happening.