PDA

View Full Version : Peer review and the ugly universe



Argos
2004-Jan-19, 02:21 PM
João Magueijo is the author of a theory that questions the basic premise behind Einstenian Relativity: that the speed of light is constant (observations, still very controversial, suggest that light might have been faster in the past). In his book “Faster Than The Speed Of Light” (I´m translating from the Portuguese) he approaches such questions and tells how difficult it was to publish his ideas, fighting against the process of "peer review”, which he believes is failing in selecting quality papers for science bulletins. “Nobody reads science magazines. Scientists don’t read science magazines”, he says.

Magueijo also has strong opinions against the aestheticism of physics, personified by the Superstrings Theory (“a hollow thing, without experimental content, without experimental predictions”). This elaborate construction has provoked amazement for the virtuous use of mathematics. However, Magueijo states that we need to accept the fact that nature might be intrinsically “ugly”, perhaps to the point of not allowing the unification of Relativity and Quantum Mechanics (“to make maths is something beautiful. But making physics is not always that beautiful”).

I tend to agree with his opinions on peer-reviewing. In many cases (maybe most of them), scientists put their articles as files on the Internet prior to publishing them on a magazine. The internet files are a means of systematically disseminate new ideas. Through the files people get to know what the other ones are writing, and the files don´t have peer review. It works fine. There is a natural selection. Peer review does not seem to be an objective way of selecting good articles, and in the end we don´t need this. The reader knows from the start what is and what is not a good article. The days of science magazines seem to be numbered.

Cougar
2004-Jan-19, 05:25 PM
João Magueijo... tells how difficult it was to publish his ideas...
To publish in a scientific journal, one needs more than an "idea." One typically needs evidence showing there is merit to the idea. Maybe this was his problem....

..."peer review”, which he believes is failing in selecting quality papers for science bulletins.
I see no evidence of such "failing."

“Nobody reads science magazines. Scientists don’t read science magazines”, he says.
I read the journal Nature, and I'm not even a professional scientist.


...nature might be intrinsically “ugly”, perhaps to the point of not allowing the unification of Relativity and Quantum Mechanics...
I'll go along with that. As Goldsmith put it, "To be sure, beauty or the lack of beauty does not determine the truth, whatever John Keats may have thought."


The internet files are a means of systematically disseminate new ideas.
In case you haven't noticed, the internet is full of misinformation. How do you tell what is good and what is hogwash? Does the author have a personal agenda camouflaged in pseudoscience? This wastes a lot of time. I can be pretty sure that the information in Nature is not hogwash, and this is because it is a peer-reviewed journal.

Through the files people get to know what the other ones are writing, and the files don´t have peer review. It works fine.
Define "work".

Peer review does not seem to be an objective way of selecting good articles...
What do you mean? That's exactly what peer review does.

The reader knows from the start what is and what is not a good article.
This is exactly what you DON'T know about articles that are not peer reviewed.

Argos
2004-Jan-20, 02:06 PM
To publish in a scientific journal, one needs more than an "idea." One typically needs evidence showing there is merit to the idea.

What is the evidence behind superstrings? For most journals, its “mathematical beauty”, the idea, seems to suffice.


I read the journal Nature, and I'm not even a professional scientist.
That is a Magueijo´s assertion, not mine, though I suspect he may be right. Perhaps he means that (printed) science journal readers over general population yields a so infinitesimal figure that it is equivalent to “nobody”.


I'll go along with that. As Goldsmith put it, "To be sure, beauty or the lack of beauty does not determine the truth, whatever John Keats may have thought."
Maybe this is exactly what Magueijo is trying to say.

In case you haven't noticed, the internet is full of misinformation.

Agreed, but I believe there is a critical mass on the Internet capable of making reasonable judgements. Do you know anyone who really takes the Nigerian Bank Account story seriously?


How do you tell what is good and what is hogwash? Does the author have a personal agenda camouflaged in pseudoscience? This wastes a lot of time. I can be pretty sure that the information in Nature is not hogwash, and this is because it is a peer-reviewed journal.


I see no evidence of such "failing."

Peer-review is not working properly because it is undergoing a total breakdown. It simply can´t keep up with the mass production of science material. What happens is that people (e.g. popular science reporters) take journals like “Nature”, from the start, as a reference of quality. This is hardly true. Articles are not separated by quality when they go through the peer review process. Peer review is a sociological response. It’s a rite. It´s not guaranteed that referees read the papers. The result is that people are continuously fooled when they use science journals as reference of quality (or maybe they are only fooling themselves in search of an authority argument to justify commonly accepted opinions).

Basically, mediocrity is what´s being accepted into science journals. Anything that is different than the common understanding, in the sense of being completely new, has a problem from the start.

There is a stronger response by the scientific community when you cast new ideas on the internet: “o, this is a new idea; lets try it”, and they go and write new papers to criticize that idea, generating new ideas. Theories flourish like organisms, and a kind of evolution process automatically eliminate the defective ones, that is, the unsustainable ones. So, this kind of debate is a completely new thing and will ultimately work better than the old peer review. Magueijo´s “shooting platform” on the Internet, the site http://www.arxiv.org is an example of this new scientific community.

wedgebert
2004-Jan-20, 02:29 PM
What is the evidence behind superstrings? For most journals, its “mathematical beauty”, the idea, seems to suffice.


They have a mathematical model that "works" for many cases, and as the theory evolves, it works for more and more cases. The evidence is in the fact that the theory has the potential to work.



That is a Magueijo´s assertion, not mine, though I suspect he may be right. Perhaps he means that (printed) science journal readers over general population yields a so infinitesimal figure that it is equivalent to “nobody”.

That is probably correct. The general public doesn't read scientific journals. You know why? Because they're BORING. They're not like Popular Science, Scientific American or National Geographic. The papers are written for other scientists in that field as a way of sharing ideas. They're not writtenf or mass consumption and most people would have no idea what was being talked about.




Agreed, but I believe there is a critical mass on the Internet capable of making reasonable judgements. Do you know anyone who really takes the Nigerian Bank Account story seriously?

I don't know anyone who does, nor do I know anyone who takes any kind of Spam seriously. However the originators of both still make enough money to warrant sending more e-mails.


Peer-review is not working properly because it is undergoing a total breakdown. It simply can´t keep up with the mass production of science material.

I think the problem is more that there is a mass production of pseudoscience. Most scientists don't sit down and pump out scientific paper after paper. They have to actually have an idea, find supporting evidence and at least run some basic experiments first.



What happens is that people (e.g. popular science reporters) take journals like “Nature”, from the start, as a reference of quality. This is hardly true. Articles are not separated by quality when they go through the peer review process. Peer review is a sociological response. It’s a rite. It´s not guaranteed that referees read the papers. The result is that people are continuously fooled when they use science journals as reference of quality (or maybe they are only fooling themselves in search of an authority argument to justify commonly accepted opinions).

Basically, mediocrity is what´s being accepted into science journals. Anything that is different than the common understanding, in the sense of being completely new, has a problem from the start.

I don't see that as true. Scientists are not as close-minded as psueduoscientists would have you believe. Most of them are after two things, the truth and a steady job. If a mainstream theory is cast into doubt, that means that there is an increased need to either validate or invalidate the new theory. That means more work.

The problem is that the radial against the mainstream ideas tend to be flat out wrong. You get these theories that would require the invalidation of many current theories and decades or more of observations and research. The evidence for the new theories tends to be shaky and circumstantional at best. Yet when the theory is not considered seriously, the originator cries foul.



There is a stronger response by the scientific community when you cast new ideas on the internet: “o, this is a new idea; lets try it”, and they go and write new papers to criticize that idea, generating new ideas. Theories flourish like organisms, and a kind of evolution process automatically eliminate the defective ones, that is, the unsustainable ones. So, this kind of debate is a completely new thing and will ultimately work better than the old peer review. Magueijo´s “shooting platform” on the Internet, the site http://www.arxiv.org is an example of this new scientific community.

I can't say that's how I see scientists reacting to a article published online (outside of a major scientific website). In fact, I'd bet most scientists wouldn't even realize it exists.

Normandy6644
2004-Jan-20, 04:10 PM
I think this goes along with the idea that new and revolutionary ideas against the current mainstream scientific principles are being "suppressed" by a coalition of conspiratorial scientists. I don't buy it, since it negates the whole purpose of science. Science does not work to reach some stability point where everything is explained nice and neat, if that were the case no one would be doing anything in science! it's more fun when things don't work out quite right, because it means that the scientists get to do what they do, which is explain how things work.

Cougar
2004-Jan-20, 04:48 PM
Peer-review is not working properly because it is undergoing a total breakdown.
No evidence for this.

It simply can´t keep up with the mass production of science material.
No evidence for this.

What happens is that people (e.g. popular science reporters) take journals like “Nature”, from the start, as a reference of quality. This is hardly true.
Pop sci reporters are not peer reviewers. And if Nature isn't a quality reference, what the hell is?

It´s not guaranteed that referees read the papers.
Well, now you're falling victim to the conspiracy theorists. You may find an anecdote here and there where a paper didn't get read, but I'd say 99.9% of them are indeed read. Prove me wrong.


Anything that is different than the common understanding, in the sense of being completely new, has a problem from the start.A common misrepresentation of the conspiracy theorists. What has a problem from the start is anything that is completely new and is not well supported.


Theories flourish like organisms, and a kind of evolution process automatically eliminate the defective ones, that is, the unsustainable ones.
Sort of like the idea of memes? In the common culture, defective ideas do get eliminated, but on the internet, defective ideas don't get eliminated. Maybe if webpages could be voted OFF THE INTERNET, but I don't see this happening....

Argos
2004-Jan-20, 05:23 PM
Cougar, this is the problem with fighting against arguments of authority: you are likely to be taken for a conspiracy theorist. I did not say Nature has no quality. I only said that it (like all the rest) can´t be taken as a quality reference "a priori".

To be frank, we shouldn´t be using Nature as an example, because it is still at the top of credibility. But trust me on this: I know a tenth quality science journal in a south american nation that pays special attention to the family name of the researcher before considering any submission. Does it tell you something? To me it looks like we were under the "ancient regime" sometimes. And this is not conspiracy stuff (it happens that i´m very well accquainted to certain academic practices in certain regions of this planet). Internet publishing can get around this kind of barrier, and let fresh concepts see the sunlight.

Argos
2004-Jan-20, 05:28 PM
They have a mathematical model that "works" for many cases, and as the theory evolves, it works for more and more cases. The evidence is in the fact that the theory has the potential to work.


I wonder how many promising concepts (with models working for many cases) editors kept in the dark in the past, shocked by the New.



They're not written or mass consumption and most people would have no idea what was being talked about.


I read Nature and I find it very pleasant. I don´t think that selling only for specialized people could keep “Nature” for long time. The fact is that journals like Nature shape the opinion that will reverberate across the media. They can become formidable barriers against the diffusion of new concepts (it happened in the past).



Most scientists don't sit down and pump out scientific paper after paper. They have to actually have an idea, find supporting evidence and at least run some basic experiments first.


So, how do you do to publish theoretical science?



The problem is that the radial against the mainstream ideas tend to be flat out wrong. You get these theories that would require the invalidation of many current theories and decades or more of observations and research. The evidence for the new theories tends to be shaky and circumstantional at best.


Ok, but creative review can cope with falsities as well. Internet publishing (and the dynamic criticism it generates) has the potential to boost the thinking power for the solution of problems. In fact, the old-fashioned peer-review system (a legacy of aristocracy) pales before the potential of Internet publishing and creative dynamic review.



can't say that's how I see scientists reacting to a article published online (outside of a major scientific website). In fact, I'd bet most scientists wouldn't even realize it exists.


Yes, but this will change inexorably, whether you and I like it or not.

wedgebert
2004-Jan-20, 06:59 PM
To be frank, we shouldn´t be using Nature as an example, because it is still at the top of credibility. But trust me on this: I know a tenth quality science journal in a south american nation that pays special attention to the family name of the researcher before considering any submission. Does it tell you something? To me it looks like we were under the "ancient regime" sometimes. And this is not conspiracy stuff (it happens that i´m very well accquainted to certain academic practices in certain regions of this planet). Internet publishing can get around this kind of barrier, and let fresh concepts see the sunlight.
Well, if a journal is known to only publish the works of people based on their family name, odds are that scientists who know this aren't going to take it seriously.


I wonder how many promising concepts (with models working for many cases) editors kept in the dark in the past, shocked by the New.

I would say very rarely. Scientists, as a whole, are seeking answers. To ignore something because it's "new and scary" is totally contrary to what science is about.


So, how do you do to publish theoretical science?
By showing that is has an applciation. You can't just say "I have a theory". You have to say "I have a theory, this is why I think my theory warrants more study, and here is some evidence that my theory at least partially works".


Ok, but creative review can cope with falsities as well. Internet publishing (and the dynamic criticism it generates) has the potential to boost the thinking power for the solution of problems. In fact, the old-fashioned peer-review system (a legacy of aristocracy) pales before the potential of Internet publishing and creative dynamic review.

That's just not true. The only people who can peer-review a given theory, are other scientists in that field. It does nothing to stick a theory out on the internet where anybody can "dynamically review" it because in 99.9% of the cases, those people have no idea what they're talking about.

Peer-review is done for a reason, and it's not because of aristrocratic backgrounds. It's because the only people qualifed to comment on a theory are those who are experts in the field.

Putting your theory on the internet (on a site that isn't classically peer-reviewed) is going to gain you two things. A: Lots of praise by people who think like you do, and B: Critisim by people who don't think like you.

It will not get you an honest, relatively unbiased appraisal of your theory.

The only online publishing that scientists will take seriously is going to be the online version of what we have now. Experts are the only people with any right to review theories, and most of the time, they do a good job.

Demigrog
2004-Jan-20, 09:28 PM
Peer-review is not working properly because it is undergoing a total breakdown. It simply can´t keep up with the mass production of science material. What happens is that people (e.g. popular science reporters) take journals like “Nature”, from the start, as a reference of quality. This is hardly true. Articles are not separated by quality when they go through the peer review process. Peer review is a sociological response. It’s a rite. It´s not guaranteed that referees read the papers. The result is that people are continuously fooled when they use science journals as reference of quality (or maybe they are only fooling themselves in search of an authority argument to justify commonly accepted opinions).


The increase in science research has merely resulted in more specialized journals. Nature is a great place to get a feel for the cutting edge of a field with less field-specific jargon than, say, Journal of non-Newtonian fluid mechanics. And, I can assure you, journal editors do read the articles they publish, speaking as someone with several “reject” letters. I might point out that most rejects are due to sloppy methods, poor writing, or simple (embarrassing :oops: ) mistakes in calculations that get noticed by the editors.

In my opinion, the reason many oddball theories don’t even get a first look is that the writing is so bad and hard to follow that any real new ideas are lost in the jumble.



Basically, mediocrity is what´s being accepted into science journals. Anything that is different than the common understanding, in the sense of being completely new, has a problem from the start.

There are enough journals out there that pretty much any oddball paper has a good shot at publication—if it is well written, follows rigorous scientific method, has repeatable results, and has conclusions that follow from the results. No editor is going to lower the quality of their journal by publishing junk just because it is against the mainstream. I’ll bet, however, that a small journal would be deliriously happy to publish something truly revolutionary, if it was supportable, as it would be a boon for circulation.



There is a stronger response by the scientific community when you cast new ideas on the internet: “o, this is a new idea; lets try it”, and they go and write new papers to criticize that idea, generating new ideas. Theories flourish like organisms, and a kind of evolution process automatically eliminate the defective ones, that is, the unsustainable ones.

Publication on the internet has three key problems as I see it: lack of credibility, lack of traceability, and lack of audience.

The lack of credibility is obvious; a quick look over crank.net will show you a lot of sites that purport to be science—but are pure bunk. How can you tell if a site is credible? For me, the best way to tell is lots of references to supporting evidence in peer reviewed journals…

Tracability is a problem because of the intangible nature of web pages. How do you prove that an idea was on your webpage in 1997? How do you prove you came up with an idea when the messageboard goes down or gets hacked? Paper journals will always be around for that reason alone. Also, copyright infringement and plagiarism are so rampant on the web that you’ll never be able to prove you found the original source of something and that the data has not been altered. Plus, not many researchers I know would put their papers online prior to publication in a journal, as many journals want exclusive rights to the article. While some scientists (I would say the ones that cannot get published any other way) may be “enlightened” and support the open sharing of papers freely, the reality is that somebody has to pay for research; that somebody usually wants credit where it is due, as well as the prestige of publication in a peer-reviewed journal.

Lack of audience may be counterintuitive with regards to the web, but how does the scientific community know what pages to look at? With the thousands of pages devoted to single topics, it can be a nightmare just to find something relevant to your needs. The solution, obviously, is pages that index and organize the webpages by discipline and filter out the noise. If you do this, however, you simply wind up with peer-reviewed electronic journals very similar to their printed counterparts. (This is, in fact, what is happening today in the world of specialized journals).

Sparks
2004-Jan-20, 09:42 PM
Publication on the internet has three key problems as I see it: lack of credibility, lack of traceability, and lack of audience.
Actually, you missed one - longevity. I can get a paper published in the IEEE Transactions in the early 70's and still find it today. Whereas if I go looking for a paper from the mid-90s on someone's homepage (and I've had to do this), they may have moved colleges, changed career or even died in the meantime and the paper may no longer even exist, let alone be findable!

AK
2004-Jan-21, 12:08 AM
Peer review is very necessary and is ultimately a beneficial process. Demigrog is correct; there are journals with various levels of specificity and circulation. Journals are arranged by the number of times they are cited, divided by the number of articles. This ratio is used to determine which have the largest audience and/or the most important science. The journals that are high in this ratio tend to be more general and more difficult to get published in.

However, just because a young scientist (like myself) may not be able to get an article published in a "major"/prestigious journal, if the work is at least somewhat novel and is done rigorously, it will get published in perhaps a more specialized publication.

dgruss23
2004-Jan-21, 04:07 AM
Peer review is vital. Seriously flawed papers are usually weeded out. Referee's can use their expertise to give solid advice on how an author should improve a paper. Most researchers that have put papers through the review process have stories that illustrate the review process working well and other stories where it hasn't worked so well. Peer review is subject to all the flaws that people are subject to.

There are odd things that can happen within the peer review system:

I've had a referee criticize a method of presenting results that was the exact method suggested by a different referee in an earlier paper. :o

I've had a referee recommend a paper for publication with several suggested changes. After making the changes and resubmitting the paper went to a different referee who rejected it. :lol:

I've had a paper under review for almost 10 months at a journal. It doesn't take that long!

I've corresponded with a researcher that not to long ago had a paper rejected on the grounds that the paper was missing a specific calculation. When the author pointed to the spot in the paper where the calculation had been done the editor rejected the paper on the grounds no changes had been made in response to the referee. :lol:

A referee criticized a Hubble Constant calculation I had made on the grounds that my sample of ~300 galaxies was too small. A week later a paper came out in the same journal that determined the Hubble Constant from only 2 galaxies. :-?

I had a referee return a paper after 5 weeks excusing him/herself from "formally reviewing" the paper saying that it was too controversial for the journal I submitted it to, but recommended a different journal. :lol:

I had a referee criticize the size of a calibrator sample when one of the referee's own papers used a calibrator sample only 1/3 the size of the sample I had used!

This is all fun, funny stuff - and really annoying at the same time! You have to learn to deal with it. If the study you're conducting and the ideas your testing have merit you'll eventually get it through. Being persistent is the key - and I've benefited from the critiques of referees.

Sparks
2004-Jan-21, 07:08 AM
I've had a paper under review for almost 10 months at a journal. It doesn't take that long!
Actually, that's about normal for IEEE Transactions journals.

Mind you, what really gets my goat isn't paper review - it's grant proposal review.
I've had a grant proposal review come back that not only contradicted itself, but said that my project wasn't really novel because it would use GPS, whiich several other projects already had for that application.
I wouldn't have minded if I hadn't actually spelt out that the novelty in the project was that it wouldn't use GPS...

Argos
2004-Jan-21, 01:11 PM
João Magueijo, number one enemy of peer-review, talks to New Scientist:

http://www.newscientist.com/opinion/opinterview.jsp?id=ns23811

milli360
2004-Jan-21, 01:19 PM
I think I got you all beat. I talked to two reviewers of a grant proposal--one rejected it because he thought it was not feasible, the other rejected it because he thought I'd already done it.

Eta C
2004-Jan-21, 02:05 PM
In physics (at least) papers tend to go through two stages before publication: the pre-print and review. A pre-print is the preliminary version of the paper and often gets a wide distribution before it's ever published. Most university and lab libraries maintain a pre-print collection. In my grad school days at the Stanford Linear Accelerator Center (SLAC) they would put a new batch out a couple of times a week and maintained a large database (called SPIRES as I recall) of incoming pre-prints. These days, pre-prints are distributed electronically as well as on paper and can be accessed from several central sites such as SPIRES. This all occurs before and without peer review, so there has to be a certain "buyer beware" when reading one of these. Also, they're not citable until reviewed and published. Nonetheless, an idea that might fail peer review can get a large distribution through pre-print. If it turns out to be worthy, it will eventually get published.

Most physics journals are done both electronically and on paper. APS journals such as Physical Review Letters (http://prl.aps.org/) are published on-line and articles are added incrementally. This may or may not have sped up the publication process. A quick glance shows times between submittal and publication ranging from a year to two months with five being about average. It does, however make for a wider distribution. Electronic subscriptions are cheaper, you can download only the articles you want, and you don't need your own library to handle the ton of paper a normal subscription would produce. I've seen collections of PRL that make most people's National Geographic collections look small. :)

This isn't to say that peer review is perfect. Like any human activity there are opportunities for bias and suppression that occasionally occur. A few papers may float around for years as pre-prints. They can, and often are read in this form, so it's not as if they're totally suppressed. Cream will rise and truely valid and new ideas will eventually be published.

To argue that peer review is a way of enforcing some scientific orthodoxy, however, is wrong. As many posters have noted above, scientists like nothing more than to be able to tear down (for valid reasons) an existing theory and putting up a new one. As the story goes, someone once walked into a physics lab and found the physicists therein happily working away at something. When asked how things were going the reply was "Wonderful. Everything we knew yesterday is wrong!"

wedgebert
2004-Jan-21, 02:25 PM
João Magueijo, number one enemy of peer-review, talks to New Scientist:

http://www.newscientist.com/opinion/opinterview.jsp?id=ns23811

Gotta say that that article didn't convince me at all. Magueijo came off as insulting and petty. Because he couldn't get his theories published, when he even admits is had a a major hole in it, the peer review process is broken.

He also uses the common tactic of shifting the burden of proof from himself to others. Specifically :

So why should the speed of light vary?

It's more useful to turn that round. The issue is more why should the speed of light be constant? The constancy of the speed of light is the central thing in relativity but we have lots of problems in theoretical physics, and these probably result from assuming that relativity works all the time. Relativity must collapse at some point, at the beginning of the Universe, for example

All evidence we have points to a constant speed of light. Magueijo is the one with the extraordinary claims, so he must be provide the evidence.

Guess I'll have to look up why relativity must collapse, because from everything I've seen, it doesn't.

gethen
2004-Jan-21, 02:47 PM
João Magueijo, number one enemy of peer-review, talks to New Scientist:

http://www.newscientist.com/opinion/opinterview.jsp?id=ns23811

Some comments from a lay person:

-It's interesting that Magueijo says he himself receives up to three papers a week to review, and implies that he doesn't read them carefully. I'm wondering why he even accepts the responsibility for review if he feels so strongly that it's pointless.
-When asked what those in charge of funding are supposed to do when some new proposal for grants is offered and they don't know whether the ideas are good, he says they can "ask someone in the know." That sounds like peer review, doesn't it?
-Perhaps if Magueijo wants to be taken seriously, he should avoid the kind of juvenile statements from his book that the interviewer quotes, which I won't repeat here, as the BA frowns on that kind of language.

tusenfem
2004-Jan-21, 04:02 PM
Peer-review is not working properly because it is undergoing a total breakdown. It simply can´t keep up with the mass production of science material. What happens is that people (e.g. popular science reporters) take journals like “Nature”, from the start, as a reference of quality. This is hardly true. Articles are not separated by quality when they go through the peer review process. Peer review is a sociological response. It’s a rite. It´s not guaranteed that referees read the papers. The result is that people are continuously fooled when they use science journals as reference of quality (or maybe they are only fooling themselves in search of an authority argument to justify commonly accepted opinions).


I have to object to the statement here as a regular referee of papers for JGR, GRL and Annales Geophysicae. I do read the papers, comment on them extensively, and will allow for new ideas being brought in.

If, as stated the referees take out any new ideas or questionable ideas, the papers on small comets by Frank et al would never have been published and never have received a revival. (though is was later clearly shown to be an artifact of bad image processing).

Internet sites, like e.g. the preprint libraries, are indeed a good way of getting your work out before refereeing, but the referee system is still the best way we have of controlling the quality of journals, albeit a good editor needs to be in charge.

Astrobairn
2004-Jan-21, 04:47 PM
I've got something against the peer review process, its lengthy and picky. But thats probably just as well as I'm far too gung ho when it comes to publication.

I know of papers that have appeared on the astroph server when they had a few flaws in them that needed clearing up in the refereeing process. I've also seen stuff thats so far against the mainstream its not funny. These ideas do get an airing. I know of one astronomer who questions an accepted idea and has papers published recently in two top journals.

Argos
2004-Jan-23, 06:10 PM
I compliment the fellow posters for the thoughtful and valuable comments.

I don’t necessarily agree with all that João Magueijo thinks, though I appreciate his rebel attitude towards the academy – part of me is still a young boy. Magueijo´s in a juvenile hype, maybe dazzled by his own condition of member of one the most prestigious institutions of the world, the Imperial College. I understand his desire of being under the spotlight. He’s young and bright, and he’s at his moment. Sometimes he gets too boring with his mannerisms and flamboyant style. Sometimes he looks as though he´d taken some stimulating medication.

However, it’s difficult to dispute his claims, from where I stand. Universities, MBA courses, Science Journals, they pop all around, in every corner, in every provincial settlement. Everybody wants to be a philosophy doctor. Everybody wants to be a (renowned) scientist (and appear on the cover of Time). Everybody is making money out of the academic paper industry. In the environment I live, to publish a science paper became a rite of passage for graduating students. And at some point the old structures of society get in the way, contaminating the process. To publish, once there´s a plenty of material, becomes an exercise of social empathy. The publishers and persons of the highest position in the scientific community are invited to Country Club parties. They are paid court to like artists are in the mundane side of life. And the name of the researcher counts (“oh, the second duke of Marigold has written a lovely article”). People without the lesser vocation for science are writing science articles, for the sake of social position - dreadful articles, with poor form and almost null content (though accepted by the peers). They advertise “on-demand” Master and PhD papers, on newspapers classified sections and on the Internet - and many of them are published in “B” journals (which also endorse peer-review process). I’ve seen identical articles in different journals, bearing slightly different names, reporting tests of variables of previous tests, made under slightly different conditions, as if there was nothing new on the horizon.

Can real science breathe in such a nasty environment? Can we take the scientific literature generated this way as reference, whatever the level? You may argue that the scientific community is capable of recognizing this pattern and acting to insulate the junk stuff. But it does not seem to be happening around for the time being. In fact, I see the deviations increasing. Or you may say that this does not represent the totality of the scientific universe. But I say that this is all the science that 80% of the people, including the specialized people, will ever get in touch with (to read Nature is quite a luxury, even for 80% of the skilled people of some places south of the Equator). Does not the digital connectivity offer an excellent opportunity for a new kind of scientific debate, enlarging the scope of the classic peer-review process?




Guess I'll have to look up why relativity must collapse, because from everything I've seen, it doesn't.

Does not relativity collapse (in the sense of to fail) at the singularities and at the big-bang?

wedgebert
2004-Jan-23, 06:38 PM
Maybe João Magueijo should try taking on the corrupt journals and not try to cast out the entire system. More importantly, he shouldn't try to replace it with a system that doesn't work.

For every journal that publishes articles by the Second Duke of Marigold just because of his name, I'm sure there are at least 10 that don't. Professionals are the biggest suscribers of journals in their respective fields, and most professionals won't suscribe to a journal that prints articles without merit.

Ian Goddard
2004-Jan-26, 11:33 AM
Some interesting links here (http://cognews.com/1074756011/index_html) on peer-review publication.

Argos
2004-Jan-26, 01:07 PM
Thanks Ian. That´s another aspect of peer-review, which hasn´t been discussed here. So far we focused on the implications of peer-review to the process of acquiring knowledge. But peer-review also affects public affairs, as it informs the decisions of those in charge of public policies.

A good BBC article on the limitations of peer-review:

http://news.bbc.co.uk/1/hi/sci/tech/3140261.stm

(*) Disclaimer: I´m not aiming a political discussion. In fact, my intention with the original post was to address Magueijo´s “ugly universe”. But I think we can keep going with this, bearing in mind BA´s restrictions.