Government does not fund peer-reviewed journal articles—publishers do

[D H]

Here is a nice (where is that tongue-in-cheek smiley?) piece of bipartisan legislation: The Fair Copyright in Research Works Act. It would remove restrictions on federally-funded research that requires researchers to make their publications publicly available (e.g. PubMed, arXiv). The galling title of the thread is a statement by Allan Adler, Association of American Publishers VP for government and legal affairs to the House Judiciary Committee last week.

 

Some reading:

http://paulcourant.net/2008/09/17/fair-copyright-in-research-works/

http://www.libraryjournal.com/article/CA6595774.html

http://scienceblogs.com/gregladen/2009/02/open_access_the_time_to_act_is.php?utm_source=sbhomepage&utm_medium=link&utm_content=channellink

 

Googling the Fair Copyright in Research Works Act results in a lot more hits, none of the favorable to the publishing industry.

 

I made my opinion known by calling Adler's statement galling. Any other opinions?

[npts2020]

What other possible "value" could the fed get besides rights to use/publish the work it funds? I wonder if this law would supercede eminent domain and who actually "owns" the results of government financed research?

[CharonY]

Well our opinion in restriction the dissemination of our results should be pretty obvious, shouldn't it?

I assume the point the publishers make is not that they own the research data, but rather the finalized paper. You are essentially free to post your data all over the net (once it has been published) but woe you if you post the article as a whole somewhere.

Truth is, of course that many people do regardless and that this kind of copyright violation is rarely, if ever, persecuted.

I have no idea how likely it is that this bill will pass, but given the fact that quite a number of the big science journals allow the authors to at least make their unedited manuscript publicly available (even though sometimes with as much as a 6 month delay) it does not really reflect the science publishing reality anymore.

Even if the NIH is not allowed to force you to put your paper in Pubmed Central, many journals will allow you to do so. And not allowing you to propagate you results will harm the journal itself in the long run (due to declining impact factors).

[swansont]

What other possible "value" could the fed get besides rights to use/publish the work it funds? I wonder if this law would supercede eminent domain and who actually "owns" the results of government financed research?

 

The concept being promoted with open access is that if the public funds the research, they should have access to the results of it. And nobody "owns" the results, because works of the government are not subject to copyright. I'm not sure of the details of copyright law that allow someone to take work that would be in the public domain and slap their own copyright on it. I know that when I've submitted papers they all include a transfer of copyright page, and in my current position I always include the phrase "work of US government, not subject to copyright" on it, but those are invariably conference proceedings, which have a different dynamic to them.

 

I also find Adler's statement to be galling. So, no other opinion here.

[Severian]

The most galling thing about publishers in my field is that they don't actually add value to research. Nowadays, researchers write papers in a format which is perfectly readable and clear without the need for publishers to re-typeset anything at all. They still do of course, but no value is added.

 

Furthermore, all the papers are available on line, and no-one reads them in the journals at all. In fact, they are often a bit out of date by the time they are published in journals. One might argue that the journals provide peer review, but the reviewers are all academics who peer-review for free (myself included).

 

Why do we need journals at all? The only reason I still publish in journals is because journal publications are the metric by which funding bodies rank our productivity.

[Pangloss]

The purpose of journals is to pad one's curriculum vitae, expand the professional development section of one's annual review, and make the school look good to alumni and philanthropists.

 

Edit: Almost forgot one: Giving postgrads something to reference!

[iNow]

I just wrote my representative to oppose this, because another purpose of these journals is to help Joe Nobody's like me to still be able to learn and enrich our lives despite the fact that we're rats running through the economic maze of corporate america... to stay connected to new and interesting ideas and research. I relax and distract myself by reading this sort of thing, and I fail to see how ANYBODY in their right mind could argue that the benefits of this act outweigh the costs.

 

So, yeah... write to your congress critter and ask them to oppose this. It's stunningly backwards and wrong on so many levels.

[Mokele]

Why do we need journals at all? The only reason I still publish in journals is because journal publications are the metric by which funding bodies rank our productivity.

 

It's rather odd that science, so advanced in so many respects, seems to be so backwards with respect to publishing.

 

The problem is, PLOS is...well, it's got a lot of problems, including some pretty egregious failures of peer-review and quality control.

[mrburns2012]

It's rather odd that science, so advanced in so many respects, seems to be so backwards with respect to publishing.

 

The problem is, PLOS is...well, it's got a lot of problems, including some pretty egregious failures of peer-review and quality control.

 

All PLoS journals or specifically PLoS ONE? I like PLoS ONE, but I'm not sure of it's progress.

[Mokele]

I've seen papers in both PLoS ONE and PLoS Biology that are, well, utter crap.

[swansont]

I've seen papers in both PLoS ONE and PLoS Biology that are, well, utter crap.

 

And that's one area where journals could be making a difference. If they served no purpose, people could just upload their work to ArXiv and their own site and be done with it. But that doesn't do much screening for garbage (not sure how much ArXiv's sponsorship program helps or hurts things in this regard).

 

Having a paper available on a website isn't ultimately much use, since links come and go, and doesn't ensure access. Journals at least represent a centralized repository, and their prestige level does some sorting as to the importance of the work.

 

But this protectionist legislation is reminiscent of the RIAA's efforts of years past, blind to the reality of information dissemination.

[ajb]

Why do we need journals at all? The only reason I still publish in journals is because journal publications are the metric by which funding bodies rank our productivity.

 

From what I can gather getting something out there on SPIRES/ArXiv is considered the same as being published as far as people on the "ground are concerned". To some extent it is "self-reviewing" as people who read the first version may well contact you with their thoughts and ideas.

 

Though, there is some rubbish on the ArXiV and I question some of the advancement to knowledge gained by the masses of phenomenological papers. (How many of they really contain new ideas and honestly extend our understanding?)

[CharonY]

I've seen papers in both PLoS ONE and PLoS Biology that are, well, utter crap.

 

That is true. But the same is true for PNAS (though admittedly they it is kind of an unusual journal).

[Mokele]

Yes, but a paper with an N of 1 wouldn't make it through any sort of peer review at all.

 

The problem is, PLoS doesn't have true peer review - it says right in the documentation that they can publish articles immediately if an editor signs off on it. And it shows - "sexy" articles have been published with gross flaws such as totally inadequate knowledge of prior work or the aforementioned N of 1.

[CharonY]

Actually some PNAS papers are published completely without peer review. Members of the academy can publish a certain number of publications without going through the process. Also I recently read in a environmental health journal a study that had a n=2 with enormously low p. Which technically actually is not possible. Sometimes reviewers (especially if they do not find some specialist on the particular field) just miss obvious flaws, unfortunately.

Edited February 13, 2009 by CharonY

[Mokele]

PNAS does that? Damn, my faith in the system just gets lower and lower all the time. Obviously peer review isn't perfect, but it's sure better than the alternative.

[Baby Astronaut]

Actually some PNAS papers are published completely without peer review. Members of the academy can publish a certain number of publications without going through the process.

All research papers published in PNAS are peer-reviewed [2]. (Wikipedia)

 

But don't know how accurate that is.

[CharonY]

There is a fast track for academy members (I knew one who had fast-tracked his papers that way). The trick is that you need to secure favorable reviews, but you can choose the reviewers yourself. "Hey Dan, could you just write up that you like the MS?". So, in theory there is a peer-review, only in a way that makes rejections very hard.

 

The good thing is that those papers have something like "Contributed by Member x" on it.

Generally it can give young scientists quite a boost if the advisor is an academy member, because the article is published with a rather high profile.

Just to add, a while back a student of the above mentioned advisor got his paper into PNAS (using that inside track) and he would not stop bragging about it in front of my PhD students (we had a common brown bag meeting once in a while). So I made a rough calculations and somewhere between 70-90% of all inside track submission get published, whereas the rate for the "normal" way is below 20%.

That shut him up and two days later there was a cake with my name on it in the break room. I shared it with the others, just in case it was poisoned.

Edited February 16, 2009 by CharonY

[Severian]

From what I can gather getting something out there on SPIRES/ArXiv is considered the same as being published as far as people on the "ground are concerned".

 

That is definitely not true. For many many purposes, peer reviewed journals are insisted upon. The Research Assessment Exercise is one example that springs to mind.

 

Though, there is some rubbish on the ArXiV and I question some of the advancement to knowledge gained by the masses of phenomenological papers. (How many of they really contain new ideas and honestly extend our understanding?)

 

They are better than all that String Theory conjecture drivel. Phenomenology papers are at least making predictions that can be tested.

[ajb]

That is definitely not true. For many many purposes, peer reviewed journals are insisted upon. The Research Assessment Exercise is one example that springs to mind.

 

Should preprints be included in your list of "publications"?

 

I was thinking for more "unoffical" things than RAE's and similar. Anything on the ArXiv is "effectively published" as it is free for the scientific community to see. I can't imagine something will be completely discredited (assuming it is of worth) if it is not published in a peer-review journal and only appears as a preprint.

 

Surely your read preprints?

 

They are better than all that String Theory conjecture drivel. Phenomenology papers are at least making predictions that can be tested.

 

I can't say that my experience of attending phenomenology seminars really reflects that. I see many nice colourful plots that say that what they have done agrees with what we have observed if said parameters lie in this (often large) range. How many of these papers really point to something new?

 

I don't mean to belittle phenomenology, it is important. I just question some of the originality and value in the volume of papers.

 

Again, I am sure there are a lot of string theory papers out there that don't really extend our understanding. The same question applies, how many are really original and of value?

 

Of course not every paper is "earth shattering", but I do sometimes wonder how much "extra knowledge" is really in an average paper.

[ecoli]

 

The problem is, PLoS doesn't have true peer review - it says right in the documentation that they can publish articles immediately if an editor signs off on it. And it shows - "sexy" articles have been published with gross flaws such as totally inadequate knowledge of prior work or the aforementioned N of 1.

well i don't know about that... I wrote a paper that just got accepted to PLoS 1, and they seem to be reviewing it... and they're taking their sweet ass time about it too.

 

 

But think about it this way... I've read articles in the big journals that are utter crap. Sometimes, in journal clubs, we'll just sit around tearing holes in papers published by big labs in big journals. You know they just got in because their big labs, or the research is new. How is this any better than open content journals with smaller peer-review standards?

I think it's more a fact that they're just more open about their easier standards, no?

 

Edit: I should mention that the only reason why we had to go "down" to the level of PLoS 1 is that we got a cranky (and lazy) reviewer at I&I.

[Mr Skeptic]

I made my opinion known by calling Adler's statement galling. Any other opinions?

 

Not here.

 

I think journals still add value by selecting interesting and important-looking articles out of the many available. Of course, it might be easier to do that nowadays by allowing a large group of scientists to select interesting articles from publicly available sources, which could probably be done easily over the internet for free.

[Mokele]

well i don't know about that... I wrote a paper that just got accepted to PLoS 1, and they seem to be reviewing it... and they're taking their sweet ass time about it too.

 

It's not that they never review, it's that they don't *always* review.

 

 

The way is see it is this: a fundamental law of human output is Sturgeon's Law - "90% of everything is crap". Politics, art, forum posts, all are mostly crap. The peer review process doesn't make everything non-crap, but merely reduces the level of crap to, say, 40%, which is a huge improvement.

 

I think journals still add value by selecting interesting and important-looking articles out of the many available. Of course, it might be easier to do that nowadays by allowing a large group of scientists to select interesting articles from publicly available sources, which could probably be done easily over the internet for free.

 

One of my most reliable sources for new developments is a weekly posting of new articles on a mailing list for my field, done out of the kindness of the list owner's heart.

 

Also, much of what is done to an article by the journal is just layout. And while some journals are more lax than others, all have some level of formatting standards. So why not just ask that the article is laid out (according to guidelines) in PDF form prior to submission? Then reviewers could comment on the science, the writing, the graphics, the layout, etc, and once it's all done, it's just posted online.