Slashdot Mirror
Cutting Out the Middle Men in Scientific Publishing
Black Parrot writes: "Just got a message that was sent to several mailing lists used by machine learning researchers, announcing the mass resignation of the Editorial Board of one prominent ML journal (i.e., the scholars who make a peer reviewed journal work). The reason? 'Times have changed. ... We see little benefit accruing to our community from a mechanism that ensures revenue for a third party by restricting the communication channel between authors and readers.' It's the music industry vs. artists and consumers, writ small. You can see the full text of the message at the UAI archive. This sort of thing has been bubbling for a couple of years. The letter mentions other cases, and I know that several thousand biological researchers have threatened to go on strike against any journal that does not make their articles downloadable for free after a fixed delay from the date of publication. The trend toward more toll booths is not the only force at work in the Internet Age!"
> Good, but the editor at first was the one separating good studies from stupid ones...
:-)
Well, no, not really. The whole idea in a peer-reviewed scientific journal is that the people performing the editing are experts in the field (ie peers of the person/people submitting the article for publication), not employees of the company which publishes the journal.
There's no suggestion of doing away with peer review.
> Will Scientists be able to apply this instead of the usual bickering ?
No, they'll still bicker...
-Andy
http://www.gimbo.org.uk/
- Published papers are available virtually forever. Go down to the Bodleian library and book out articles published by Michael Faraday or Robert Hook. Whatever they do here has got have the same sort of permanence.
- Remember all those tapes that NASA has that they don't have the drives for anymore, or don't know the format? Paper doesn't have that sort of problem. Again, they need to ensure that whatever they produce can be moved on to whatever the current technology is
Other than that I think it is an excellent idea. I hope it scares the shit out of whatever the journal publishing equivalent of the RIAA is.Why limit the peer review process to the opinions of a select group of people?
/. -- very often comments are moderated up as "insightful" or "informative" even though the comments are actually *completely wrong*.
Because there's a lot of idiots in the world.
The purpose of peer review isn't so much to decide if results are interesting enough to be worth publishing; the purpose of peer review is mainly to decide if results are *correct* enough to be worth publishing. For an average paper in a major scientific journal, there might only be a few dozen people in the world who are qualified to make that judgement. Fortunately, the journal editors know who does what research, and thus the journal editors can send papers to the right reviewers.
If you open up peer review to the masses, you'll have people "moderating up" papers because "they look interesting", even if the papers are complete BS -- because most people can't tell the difference between a scientific paper and BS. We see the same thing on
At an absolute minimum, peer review should be restricted to people who have published papers in related fields; ideally, it should be restricted to people who have published several major papers in related fields... which is exactly the status quo of peer-reviewed journals.
Tarsnap: Online backups for the truly paranoid
The new open journal, JMLR that replaced the old one is a great solution to the problem. However, I'm dismayed to find that they publish using proprietary formats. Namely PDF and Postscript. Wouldn't it, thus, cost money to save to those formats? I think they should use open standards *only*. Why not use LaTex or just plain old HTML 4? This would better allow scientists from developing countries to publish their work rather than wasting precious money buying licenses of Adobe Acrobat. It seems they are fighting closed proprietary standards in the first place and should not be supporting them.
Furthermore, I could actually read the articles much easier than saving to a file and launching a bloated Adobe application or, gasp, use that annoying PDF plug-in that usually crashes my browser.
What I don't understand about all of this is why are there so few people who can see 'the big picture', including our law makers, businessmen and leaders.
.com boom and bust as demonstrating that the internet isn't as important as was generally thought, because they assumed that what would happen is just a variation on what went before. They don't understand that the internet is causing fundamental changes to the way the world works.
Many people stupidly saw the
The world now has a new a new set of rules which over the coming decades are going to completely change the business landscape. Record companies - which currently make money by duplicating, distributing and promoting physical manifestations of music - are going to die, because duplicating, distributing and promoting music can be done at virtually zero cost by anyone now.
Software companies that create and charge high prices for infrastructure and very widely used software are going to die out because their software is so widely used that it will make more sense for their bigger customers (companies and governments) to contribute programming time to open projects than to pay for products from a single company.
General travel agents are going to die out because they are middle men that will serve no useful purpose in the future. Ditto with publishers of scientific journals.
Unfortunately some of our law makers seem to think this natural progression is unjust and are creating laws to restrict or outlaw the technology. There is historic president for this type of response - the Luddites, who smashed machinery during the industrial revolution because it threatened their professions. Hundreds of business types and professions died as a result of the industrial revolution. The same will happen over the coming decades of the Internet revolution.
Unfortunately our leaders and lawmakers, under the influence of the threatened professions, are acting like Luddites in a very literal sense.
ArXiv is great, but the problem is the non-refereed aspect of it. We need properly refereed free (or cheap :) journals... and there is really no reason why not. People don't get paid to referee papers, or submit papers... there is just this man in the middle eating all this money for no net gain to the community.
A journal that had all its articles freely available electronically could easily charge for paper copies as long as it was reputable. All we need is for some big names to get their act together. An example where they *have* got their act together is from my University:
Geometry & Topology
This is a peer reviewed journal on (guess what) geometry and topology. All the articles published in the journal are freely available electronically, or you can order the paper version at a low price.
I don't see any reason for any extra distribution fee.
-- Help Digitise the Public Domain at DP.
There should be some kind of registration process so some 12 year old kiddy can't submit a journel on UFO study and get all his friends to rate it up. The registration won't stop that, but most kiddies won't bother going through a registration to screw with a website.
But a bunch of creationist adults will devote 5 hours a day every day to doing just that, but with papers sufficiently well written that they seem scientific to a casual reader. And all the sudden, you have a bunch of "peer reviewed" highly rated anti evolution creationist BS* sitting on a respected journals lap. NO WAY!
And don't even tell me that the negitive ratings from all the good scientists will ballance it out. Even if they suspect its BS, a decent scientist will not moderate something down until she has taken the time to look it over and check the methodology to make sure she isn't rejecting it out of hand because she has tons of experience that the end result has to be wrong. As for "peer reviewers are biased against some conclusions", sure you get biased after the 20th time the same conclusion turns out to be supported by sloppy work, wishful thinking and all out lies, but you still check the methodology to see which one it was this time.
But the point is that honest work takes longer than lies. Debunking lies takes more time and effort than presenting them. Moderating down conclusions that contridict your holy book takes less time than the propperly designed research it took to come to them. And the people who care the most about spreading lies are often devoting most of their lives to it, while the people most motivated and qualified to correct those lies are doing other possitive research and don't have the luxury of playing wack-a-mole with the latest psuedoscientific voodoo all day.
Nutshell: TRUTH IS HARDER. In an open marketplace of ideas managed by libertarian principles and voted on democratically, the truth will get its ass kicked. I'm sorry that we don't live in that perfect world where "the solution to bad information is good information, not supression" or "the truth will out" or any of those other nice thoughts with no basis in reality. Really sorry.
*and just to not pick only on the biggest target, lets not forget perpetual motion, psychic healing, ESP, alien visitation, racial infer/super-iority, gender work from both sides of the fence, conspiracy theories, power lines cause cancer, soil theory, homeopathy, dowsing, ok just put "Flim Flam" table of contents here....
Kahuna Burger
...will work for Chick tracts...
As someone who publishes in traditional peer-reviewed journals, let me add a few comments to this. What journal peer review provides that can never be replaced by an open forum like Slashdot is credibility. The journal editor is responsible for selecting reviewers who are knowledgeable in their field. These editors, at least in the physics journals I've worked with, are responsible professionals committed to publishing a high-quality journal and not just ego-driven hacks. Because the editors are considered trust-worther by the journal readers, articles subjected to a peer-review process can be take seriously (they can still have mistakes, of course, but usually not obvious ones).
Similarly, because article reviewers are selected based on their professional credentials, reviewer comments in a peer-reviewed journal are worth my serious consideration. I may disagree, with the reviewer, but that's part of the process and I'm given the opportunity to respond to the reviewer's criticisms as a normal part of the review process. Finally, because the review is blind (either single-blind or double-blind depending on the journal), the reviewer can safely criticize a more established colleague without fear of retribution. On a peer-reviewd journal, with an editor as the final arbitrator, this procedure works. On the internet, though, either reviews are public which opens the viewer up to retribution, or the reviews are completely anonymous in which case the reviewer lacks all credibility.
A comment made by a random AC on slashdot, in contrast, is not worth my [professional] time. There are just too many posts by too few knowledgable people. Even with the established pre-print servers (which are not peer reviewed but of considerably higher signal-to-noise ratio than most other public forums), it isn't worth the time and effort to read the articles unless I already know one of the authors, either personally or by reputation.
In sort, peer-review is not some arbitrarily imposed requirement from on high; when handled properly it is a valued part of the scientific process and I will not take the time to read any journal that does not maintain peer review. There are only so many hours in a day and I would prefer to use them to get real work done instead of wading through a stack of dubious articles on the off chance that one of the authors will have something worth-while to say.
-JS
P.S. For the record, I read slashdot on my own [personal] time. This is entertainment, not work.
Vanity of vanities, all is vanity...