Slashdot Mirror
White House Plans Open Access For Research
Hugh Pickens writes "Currently, the National Institutes of Health require that research funded by its grants be made available to the public online at no charge within 12 months of publication. Now the Office of Science and Technology Policy in the Executive Office of the President is launching a 'Public Access Policy Forum' to determine whether this policy should be extended to other science agencies and, if so, how it should be implemented. 'The NIH model has a variety of features that can be evaluated, and there are other ways to offer the public enhanced access to peer-reviewed scholarly publications,' OSTP says in the request for information. 'The best models may [be] influenced by agency mission, the culture and rate of scientific development of the discipline, funding to develop archival capabilities, and research funding mechanisms.' The OSTP will conduct an interactive, online discussion that will focus on three major questions: Should this policy be extended to other science agencies and, if so, how it should be implemented? In what format should the data be submitted in order to make it easy to search and retrieve information? What are the best mechanisms to ensure compliance? 'It's very encouraging to see the Obama Administration focus on ensuring public access to the results of taxpayer-funded research [reg. required] as a key way to maximize our collective investment in science,' says Heather Joseph, executive director of the Scholarly Publishing and Academic Resources Coalition."
My opinion was always if the taxpayers pay for it, the taxpayers own it. Research, patents and discoveries and even software. At a minimum the government should be able to transfer licenses from one branch to another. If your research is that valuable, don't take federal money. A lot of universities are taking federal money for research and then selling those discoveries to companies that sell them back to the taxpayers. It's not always that clean but it just doesn't seem right. If you don't like the restrictions, don't sell to the government. I love the way so many institutions, lately including banks, are acting like they're doing us a favor taking federal money. And there's always someone who will yap about government wouldn't be able to get access the best software tools. I doubt that. I'm not talking about making anything the government buys open source, just that government can move software licenses around based on need.
We're still screwed, if it takes no less than a president to enact something as commonsensical as this.
Still, though, go for it big O!
If you don't know what AltaVista is (was), get off my lawn.
Just mandate that all Federally funded research papers be submitted to Arxiv.org. In many fields (e.g., astrophysics), that happens routinely. In others (e.g., geophysics), it is rare. I see no harm arising in the fields where this is routine, and making it universal would mean that the entire scientific world would gain access to all of our scientific research.
Here's the way things work right now in my field, astrophysics: a scientist has an idea. He writes a grant proposal to the NSF and receives money. He uses the money to (hire a grad student, travel to telescope, build an instrument, etc.). He writes a paper on the results. In order to have the paper published in one of the big journals -- which is necessary to gain credit for tenure, promotion, reputation among peers -- he PAYS THE JOURNAL ~$110 PER PAGE. The journal makes the information available only to subscribers, who pay around $50-$100 for individuals or $1500-$3000 for institutions.
If you don't publish in the big peer-reviewed journals, you don't get recognition.
So, suppose that the government changes things: now the journals must make government-funded research available to the public without charge. The journals will lose money from their subscriber base; after all, who would bother to pay for the articles when they are free? Where do the journals make up the money? My guess: they increase the page charges. Now it might cost $200 or $250 per page to publish an article in a journal. Whence comes that extra money? From the government grant.
Result: the scientific papers are now available freely to the public, but scientists must ask for more money from the NSF in order to pay the higher page charges.
Michael Richmond "This is the heart that broke my finger."
mwrsps@rit.edu http://stupendous.rit.edu
...what arguments Springer-Verlag, Elzevier and the like will find to fight this. Since in my (humble, but competent) opinion there can be no honest reason to oppose this, they will need to be very creative. And yes, as a scientific researcher, I have very often been hampered in my search for references by the unjustifiable monopoly held by those vultures, and the hefty subscription prices that go with it. Heck, they even made me transfer the copyrights on my own publications to them. I may not even cite myself extensively!
Like you can believe anything they say anyway.
While I have no doubt that the first part of the statement will come true, with regards to journals trying to ramp up publishing costs, I have doubts that the scientific community will bear that much more extortion. As it stands, we're practically at the breaking point, where scientists would almost rather not publish than do so, but have no other choice as you have stated, their livelihoods depend on it. As it stands, you already have to pay, give up an irrevocable copyright license... you just about have to give over your first born child to get any notoriety at all.
Once the government grant money for publishing is increased, you'll likely see scientists push to keep some of that money for themselves and their research, and more alternative services such as arXiv will being to pop up and grow and replace the stodgy old journals. The fact is, scientific journals are just as close to becoming obsolete with the invention of the internet as newspapers are, and it's very likely they'll try the same exact tactics as newspapers to hold on: increasing fees on both sides of the equation, going to governments and schools to try to ill-legitimize other services, form exclusive contracts with various schools, etc.
You can argue about journals being more peer reviewed and such, but even the internet serves as a better tool for those purposes as well: being able to instantly expose anyone in your field to your work and have anyone criticize or improve upon it immediately rather than going through various journal's information back-channels or blackbox review committees will vastly speed up science.
So, while things may look more bleak in the short-run with such a plan, in the mid- to long-term, it should help restore the US's place in science.
If so, I'm not holding out much hope...
An, excellent proposal, long overdue, and the whines of Academics, Librarians and Learned Journal publishers, which will all be anti must be stoutly resisted, and can easily be solved by the Presidential Task Force.
.doc and OOXML must be prohibited.
This will be kind to trees, Library Space, shelving and reading and good for students, adademics and the general public.
Peer Review must ensure that Data and Methods are published or Publically available, which would stop repeats of Climategate in their tracks.
Employed, tenured Full Professors in the age 40 to 60 should be required to Peer Review min->max or % of papers to retain tenure.
Formats should be inclusive, certainly TeX, SGML (used in house at Elsevier), PDF, HTML and ODF. Proprietary formats like Flash,
Google can (will) do the indexing as can M$.
'It's very encouraging to see the Obama Administration focus on ensuring public access to the results of taxpayer-funded research [reg. required] as a key way to maximize our collective investment in science,' says Heather Joseph
Ms. Joseph should thank the Bush administration for starting the ball rolling by opening up the NIH. Going forward, as long as the government applies the same peer review and quality standards to publishing the results that reputable journals do, the policy makes sense. But what happens if the researcher's peers don't like the quality of the work? Today it's quietly buried, will the government still publish it but with some kind of a caveat/stigma?
As a professor myself, I hope that the unintended consequence will be that we move away from the restrictive, expensive, academic journal publishers like Elsevier and toward an open model of academic publication where your recognition and peer review come from broad, open, dissemination.
I, for one, would like to see a peer review system where articles are posted on-line and evaluations (i.e. referee reports) are also posted in an open, strongly authenticated, way. I don't know about you, but one thing that really annoys me is to receive a referee report on a paper where it is obvious that the referee hasn't even read past the introduction. I believe that forcing the evaluations to be open, and strongly-authenticated (so that everyone knows exactly who is writing it) would improve the quality and credibility of research.
I suspect that some people would claim that if referee reports aren't anonymous, then they won't be honest. But, a referee report should not be about opinions, it should be a straight forward analysis of the results reported in the paper. If it's really science, then it should be completely objective, thus opinion and personality should have nothing to do with it. Hence, there should be no need for anonymity. When I grade my students' papers, it certainly isn't anonymous, but it doesn't need to be because I am giving them objective feedback (e.g. "this is wrong because you said cos(x+h) = cos(x) + cos(h) which is not true.").
Using an open system would allow articles to receive recognition and ranking based upon the open discussion of their merits. Individuals doing the ranking could also receive recognition for the quality of their work, which is important because it can sometimes take weeks of work to thoroughly understand a new result. That work should receive more acknowledgment in the academic system than it currently does. (I suspect it's the current lack of acknowledgment for refereeing which makes many people into lazy referees. After all, why bother putting much effort into that referee report when it won't count toward promotion. You are better off spending that time writing your own papers.)
Finally, using an open system gives the public greater credibility in the system. When people want to know why paper A is considered correct and paper B isn't, the analysis and discussion will be available, too.
It's my understanding that people in congress have considered before the question of expanding open access requirements to other disciplines. Obviously publishers will oppose such a move because it cuts into their bottom line. How far it eats into the bottom line depends on the reasons people subscribe and just how the opening of access works: You can make new papers closed and older papers open, or you can do the reverse. Additionally, you can make papers totally open or you can institute some half-way measure, like something similar to Google Books or Amazon book previews, which are designed with the aim that you can read the content but not easily save a copy of it.
In Physics, the APS (the professional organization for physicists) publishes the Physical Review journals, which are some of the most influential in the field besides Nature and Science. Apparently the APS relies on subscription fees from the journals in part to subsidize many of their other (worthwhile) activities, e.g. scientific conferences. As a result, it's my understanding that they opposed open access requirements (though they might have been willing to accept them in some form). This is especially interesting because the Physical Review journals have relatively friendly policies that allow one to post a pre-print to the ArXiv (which physicists generally do) and host a copy of the paper on your own website, so most of the papers they publish (at least more recently) are already available for free one way or another.
I generally have a very favorable opinion of the APS, but I would very much like to see more openness in scientific journals, at the least for taxpayer funded research. If this means that the APS will have to raise dues and conference fees to more accurately reflect the cost of their activities, I think that's something we'll just have to accept.
"You call it a new way of thinking; I call it regression to ignorance!" -- Operation Ivy
The fact that they are in the UK is irrelevant; they should still have published their data and algorithms instead of fudging the data and hiding it.
...if the government decides to create a new, open journal will be developing a 'prestige factor' for it. If it isn't impressive to publish in it, the best works simply won't be published there.
Since there's little chance of this potential journal assailing Nature or Science, we may end up with free average quality papers and expensive (to both publish and consume) high-impact ones. That is, unless there's a scientific culture change, and academia starts to value open-access and transparency. Maybe if a few big works are published openly we may have a chance, and with all the climate-gate noise this would be especially good for science on the whole.
If the government simply adds a clause that all papers must be available to the public, we might instead see a new market force driving the big journals to open up all or some of their articles to the public, because if they don't, high quality government-funded works won't be available to them. Not a bad prospect either!
Meanwhile, if you're looking for a paper that your institution (if you belong to one) doesn't have access to, I highly recommend google'ing the title in parentheses - it's amazing how many are posted on ill-protected course pages.
Hey mate, spare a sig?
You'd need to implement something like that in a hierarchical manner, not unlike slashdot. The number of submissions would dramatically increase due to its free nature while the quality would surely decline, and nobody wants to sit there and read a large percentage of questionable work to determine if it's valid and if so if the results are correct. New submissions could be subject to quick reviews for validity testing (with moderation of course, people who troll by negatively reviewing and voting down new papers without actually reading them or considering their results, or for any bias should be barred from such reviews), and once a paper has been verified it can move on to a stage where people who don't want to sift through garbage to find the gold can really scrutinize them and see if they stand up.
:)
Why not make such a website?
Some provision should also be made for making data available as well. For controversial and issues of practical importance, the benefit is obvious. There could also be unexpected benefits for esoteric subjects. The experiment I worked on for my physics dissertation cost around $10^7, and the data is sitting on shelves at three universities, if it hasn't been chucked yet. Admittedly, in my case the only people who would concievably care are the hundred or so people in the subfield, and they probably care more about working on their own experiments. But making the data available would also preserve it for potential inquiries that have not yet been concieved.
Of course there are practical problems such as the volumes of data and organizing it to make sense, but I suspect that any work done to make the data interpretable by others would improve scientist's own analysis.
Nice summary of why the "Free Market" isn't always the solution to everything.
Over-the-top Response Guy! Giving "Over-the-Top Responses" since 1970.
You'd need to implement something like that in a hierarchical manner, not unlike slashdot. The number of submissions would dramatically increase due to its free nature while the quality would surely decline, and nobody wants to sit there and read a large percentage of questionable work to determine if it's valid and if so if the results are correct. New submissions could be subject to quick reviews for validity testing (with moderation of course, people who troll by negatively reviewing and voting down new papers without actually reading them or considering their results, or for any bias should be barred from such reviews), and once a paper has been verified it can move on to a stage where people who don't want to sift through garbage to find the gold can really scrutinize them and see if they stand up.
Why not make such a website? :)
Yes, the description you're providing is very similar to what I have in mind. I would love to create such a website, but it will have to wait until after I have tenure... and I'll need some sizable grants to get it up and running to begin with. I wonder if NSF would fund the development of such a system:-)
Is that where you say nih to the scientist until they're forced to do their jobs right.
Two related items I've written on this:
"An Open Letter to All Grantmakers and Donors On Copyright And Patent Policy In a Post-Scarcity Society "
http://www.pdfernhout.net/open-letter-to-grantmakers-and-donors-on-copyright-policy.html
"On Funding Digital Public Works "
http://www.pdfernhout.net/on-funding-digital-public-works.html
The executive summary from the first (the second is a longer version of the first):
"""
Foundations, other grantmaking agencies handling public tax-exempt dollars, and charitable donors need to consider the implications for their grantmaking or donation policies if they use a now obsolete charitable model of subsidizing proprietary publishing and proprietary research. In order to improve the effectiveness and collaborativeness of the non-profit sector overall, it is suggested these grantmaking organizations and donors move to requiring grantees to make any resulting copyrighted digital materials freely available on the internet, including free licenses granting the right for others to make and redistribute new derivative works without further permission. It is also suggested patents resulting from charitably subsidized research research also be made freely available for general use. The alternative of allowing charitable dollars to result in proprietary copyrights and proprietary patents is corrupting the non-profit sector as it results in a conflict of interest between a non-profit's primary mission of helping humanity through freely sharing knowledge (made possible at little cost by the internet) and a desire to maximize short term revenues through charging licensing fees for access to patents and copyrights. In essence, with the change of publishing and communication economics made possible by the wide spread use of the internet, tax-exempt non-profits have become, perhaps unwittingly, caught up in a new form of "self-dealing", and it is up to donors and grantmakers (and eventually lawmakers) to prevent this by requiring free licensing of results as a condition of their grants and donations.
"""
A 21st century issue: the irony of technologies of abundance in the hands of those still thinking in terms of scarcity.
So you get a $100,000 government loan on something too risky to get enough private funding and then want to pay it back X years later so you retain all rights?
In other words, you want a risky zero interest loan for R & D for X years? The X being variable and the payback possibly never happening? Or additional money will be sunk in until something fruitful comes out of it.
I'm just using your example and fairly phrasing it; I'm not saying government shouldn't ever give out risky zero interest loans or grants to private projects for an indirect greater good. Stadiums these days are all doing it; except that the loans don't seem to ever get paid back - on rare occasion they'll even call it a grant or claim ownership for a while until they sell it off cheaply when the "right" politicians come to power.
China doesn't do jack for quality control other than execute a few people when things get really really bad. China still has to get past our FDA.
Our FDA costs money to maintain a minimum level of quality control; despite corruption and the fact it is hopelessly underfunded. Sure you pay some fees for FDA services but I doubt they reflect costs involved in running the FDA. The EU has MORE people and the BIGGEST economy in the world with strong standards and many times the barriers for businesses; they are proof that our half measures are nowhere near as damaging as the misleading business peoples' claims.
Democracy Now! - uncensored, anti-establishment news
I mostly agree with an open review system, but there are a few issues. Reviews probably need to remain anonymous in some fields to prevent the scientific equivalent of "log rolling" : providing good review to people who gave you favourable reviews. Some fields are small enough that the pool of available reviewers is tiny.
I am very much in favor of making government funded scientific research available to the public but again there are a few problems: In many fields your career depends on publishing papers. You don't want someone to spend a lot of effort developing an experiment and then someone else publishes first based based on the data (without full credit going to the original experimenter). Of course you also don't want an experiment to sit on data for years before it is released.
Making the data public can have different meanings: You can provide the "raw" data to the public - this is usually useless - "here, have 100 TB of unlabelled binary data". But, the alternative might be to require the scientists to process the data and provide it in a easily interpreted form - this could be a very large added workload.
I'm fortunate enough to work in a field (accelerator physics) where publishing is not particularly important to my career, so everything I work on is available. I do work with X-ray experimenters who feel they need to keep their work quite until they publish - they are worried that someone else will publish first.
But, a referee report should not be about opinions, it should be a straight forward analysis of the results reported in the paper. If it's really science, then it should be completely objective, thus opinion and personality should have nothing to do with it. Hence, there should be no need for anonymity. When I grade my students' papers, it certainly isn't anonymous, but it doesn't need to be because I am giving them objective feedback (e.g. "this is wrong because you said cos(x+h) = cos(x) + cos(h) which is not true.").
Your idea is interesting, but I guess opinion matters more than you think it does. I recently recommended rejection or major revision on a paper about "Economic forecasting". I think the math was all fine and dandy. In your view, I should have recommended publication. But I simply could not stand the way terms like prediction or forecasting were used. I asked the author to use terms like "data extrapolation". If economic prediction were possible, why publish? The function of economic forecasting is to make astrology look respectable; and all those clich'es.
My point is, something could be totally right inside the little confines of a model, while the premises of the model become a matter of opinion and philosophy. So I don't see how you can actually separate the two; what's opinion and what's objective, so cleanly like that.
http://tech-net.sba.gov/tech-net/public/dsp_search.cfm is the search engine. The Small Business Administration doesn't make SBIR or STTR awards, but Congress has charged the SBA with tracking them. Every year, all agencies that awarded SBIR or STTR awards in the previous Federal Fiscal Year are required to report those awards to the SBA, by March I think. The TECH-Net search engine allows you to capture search results in mail merge format for import into spreadsheets, for example. You can drill down to awards and even phases within an award. (As a result of Phase I, the abstract of Phase II could change, for example, and you can see both.) I see that its keyword search capability is finally back after a long hiatus.
This search engine has been on the Web for over 10 years.
Including medical research? You know, transparency that would have shown up the tobacco industry far earlier. Or would have shown the downsides of use of aspartame as a sweetener?
Or is it going to be for research that may negatively impact commercial interests?
You didn't read the code, did you. The bit saying "this is fudged" put the "fudged" data (mind you, why label it "fudged" if you're working a conspiracy???) into an array that was NEVER USED.
Commented out.
Anulled.
Not fudged.
And if you still demand this is proof, to be of any use in misinforming, the results of that fudged data has to be used and displayed in some paper somewhere. Where is it?
PS note too that the code that prints the fudged data also prints on the axis of the graph that this IS fudged data. Hardly a good idea if you want to conspire secretly, is it...)
I mostly agree with an open review system, but there are a few issues. Reviews probably need to remain anonymous in some fields to prevent the scientific equivalent of "log rolling" : providing good review to people who gave you favourable reviews. Some fields are small enough that the pool of available reviewers is tiny.
My point is, something could be totally right inside the little confines of a model, while the premises of the model become a matter of opinion and philosophy. So I don't see how you can actually separate the two; what's opinion and what's objective, so cleanly like that.
Yes, these are certainly valid objections. Perhaps the review system needs multiple mechanisms for assessing articles.
An open refereeing system could establish correctness (or, in less rigorous disciplines, could judge how likely it is that the presented data and arguments support conclusions). This part would not impart value judgments (such as how significant the results are).
A "significance" mechanism for determining the value of a work could begin with a slashdot/digg style rating scheme that raises worthwhile results above the inevitable noise of crank submissions. Those works could then be evaluated for significance (within a field) openly by experts in that field. Significance evaluations would need to be cumulative, so it would take many experts agreeing that a work is important within a field before it gets elevated to the equivalent position of a top-level journal paper. Much like slashdot, the experts might even have a limited number of (randomly and secretly assigned) mod-points that they can use for significance. This provides a polite exit from a complicated quid-pro-quo political situation:
"Hey John, I was just looking at your article on homeomorphic fig-tree inversions, and it seems pretty significant. I've got a related result on peach-trees that I recently submitted, could you take a look at it?"
"Sure Dave, but I'm all out of mod points this year, so all I can do is check the correctness for you."
The significance measure could also be fluid over time and incorporate measures like how often the work is cited, etc. Certainly the system would have many wrinkles to be worked out in its initial years. Perhaps, though, in the long run we would have a superior system to the current mess.
I suggest they use the PDF format for interchange so that redactions can be worked around!
how is babby formed?
http://www.opencrs.com/
I'd like to buy homeland for our 10 million people. http://twitter.com/mahadiga
The trouble with most moderation systems is that they tend to be based on the fundamentally flawed idea of objective values on subjective criteria. Correctness is one thing, but one persons significant might not be another persons significant. As fields grow and diverge you'll end up with significance for the divergent branches being subjective and depending on which branch you are pursuing. Eventually you end up with either too much significant material, too heavy culling, patchwork meta-fixes or being forced to split off into somewhat arbitrary groupings.
A more scalable model would build upon a social networking model of moderation; anyone and everyone can assign moderation values as they like, but the weight of a moderation value that the viewer sees is dependent upon how closely a specific moderator follows the viewers own moderation. That way, if someone rates an article 'significant' and you and they often have the same idea of what's 'significant', that article will be moderated as 'significant' for you.
Such a system would sidestep the whole problem of quid-pro-quo politics and, in fact, most other political problems tied to hierarchial system views. Of course, it may also result in some level of fracturing of fields that may or may not be entirely postive; it's hard to tell, I have yet to see such a system implemented in reality outside certain small subfields (last.fm neighbours and music recommendations seems to work with a somewhat similar concept, but is still fairly far away).