Slashdot Mirror
Wellcome Trust to Require Open-Access Publishing
Lars Arvestad writes "The Wellcome Trust, one of the worlds largest research funding agencies, will require results from research funded by the Trust to be available in public repositories six months after publication. The Trust's policy advisor Robert Terry writes in
an article in PLoS Biology that the Trust plans to start its own public access repository where authors are expected to deposit their published works. The repository is modeled after NLM's PubMed Central and is called UKPMC. Terry's article also mentions that a recent Wellcome report found that an author-pays business model has the opportunity for a saving of 30 % on publishing costs alone compared to reader-pays. This contrasts the recent IEEE report (Slashdot story last week) where it was claimed that some universities will face higher costs using author-pays."
I am not sure what is the best answer.
Maybe differtent strokes for different folksis the way to go.
What is undeniably good is that people are trying to do something about access to the information and the cost of accessing it.
The Singularity is closer than you think
Quant
I've had that problem... Randomly recently, I googled my father, and found some articles he had authored in post-doc work, with his mentor/professor and someone else who's name I did not know... I tried to access the article unfortunately the publication had only the listing of the articles of its back volumes online, and even that seemed partially incomplete... Its sad... unless I can find that article some day in the future in our things... I may never get to read the paper... its the kindof thing that can get too easily lost among one's personal things after 20 years... moving from state to state and country to country...
Gravity Sucks
In Astronomy, all the major US journals are author pays. We also have a preprint server that is free that most astronomers post their articles to (except for Nature articles because Nature won't let them). The one problem I have with author pays is that you have to come up with the grant funding, and a lot of the grant funding is project specific, so if you do a side project that isn't funded (something real common when working with students), you've got to get creative and beg, borrow, or steal the funding to pay for it. For instance, I had a 29 page paper as a grad student that didn't fall under my advisor's grants, and had to beg from department sources (finally getting the $4000 I needed from our observatory budget).
Professor of Astronomy, Author of Spider Star & Star Dragon (Tor)
Ultimately, the best thing that could happen to research publications is to separate the peer review process from the publishing process. This would facilitate "just-in-time" publishing while maintaining the credentials for a peer-accepted work. Then, other interested parties can download the paper and read it from their computers, or print it out to a hardcopy (as school libraries might do).
Yes, it takes the whole aspect of "profit" out of the equation, but this is science, not entertainment.
Things like this are the best way to force open access for scientific publications.
Why is this such an important thing?
Imagine the follwing business plan:
1) Make people PAY to incorporate their computer programs into your project.
2) Make people give you their copyrights to accept their program into your project.
3) Make people contributing code to your project also debug other peoples code. For free.
4) Profit!
Who would put up with such a kwaaazy system? We scientists. Why do we put up with this exploitation? Because we have no other choice if we want to remain competitive.
However, if there is enough external pressure for the system to change, it will.
You think I'm a Krazy Krackpot? I present you with the following:
1) My lab publishes ~ 2-3 papers a year, in journals like Biochemistry and J. Biol. Chem. It costs us ~ $2,000/publication.
2) Although we PAY the pulishers money, we still give them full copyright. (Recall: we formatted, created graphics and edited the documents).
In case you are worrying about the poor publishers, remember the following:
1) Few people read printed journals these days, most download the articles in PDF format. How much can that cost?
2) The process of editing and reviewing papers is done by other scientists, such as myself - for FREE.
Let's hope the trend is towards liberating the information that is paid for by taxpayers.
I think, therefore I thought.
This is one of the main concerns, but I think it will be mitigated as the publishing model starts to take off. If a paper is important, then it will always get reviewed. Hopefully there will be less of an emphasis on the volume of papers published, with a greater emphasis on quality. The current model is terrible for this. An author may incur few costs when publishing a paper, so there is no disincentive to publish. People start publishing lots of papers to pad their CV, and then it becomes the norm. This is bad for science.
One concern is that the author pays model will replace this phenomenon with something worse. One possibility is that the prestige of the journal will be inferred from the amount that they charge authors. Not so much inferred by scientists, but inferred by funding agencies. I don't think this will be too much of a problem, since funding agencies are generally cheap.
So what are the other problems with this model ? Well, the integrity of all the participants is much more critical than it was under the old model. There will be pressure on journals to publish things - money changing hands will create that expectation. There is far more potential for major corruption and scandal. I can see scientists and funding agencies threatening to take future papers elsewhere if X is not published, because journals will be directly beholden to scientists for revenue. One way around this is to have funding agencies help fund journals as well as scientists, but this creates the opportunity for collusion. There is probably no good solution to this problem, so we should hesitate before making paper submissions the sole source of revenue for journals.
In Soviet America the banks rob you!
On further reflection, I think that the only way to ensure the integrity of the process is to have a third model: "everyone pay".
Under the author-pays model, libraries get a free ride. Before we give them all that money back, we should ask them to do something small to help the process. Libraries should cover the distribution costs. A few terabytes here and there isn't going to cost a lot of money, probably the amount that they spend on a single journal now. And putting a copy of all this information in every campus library will be pretty economical - no internet charges for local connections. Libraries could make a vital contribution by simply doing what they have always done, amalgamating information and making it accessible to their local community. Journals wouldn't have to worry about paying thousands of dollars in internet access fees each month, and I wouldn't have to scour through forty different poorly designed web sites to get information.
Scientists pay for review, libraries cover distribution, and journals are the middlemen with few recurring costs. What do people think ?
In Soviet America the banks rob you!
One of the things I really like about PLoS (Public Library of Science) is that they don't just make their articles free to access, they actually release them all under the Creative Commons license. You can do pretty much anything you want with released content, including derivative works and commercial use, so long as you give the original author credit.
Hopefully the new repository that the Wellcome Trust is setting up will use something similar.
Hopefully the author pays thing isn't run like the crooked 'we will publish you' sci fi ripoffs out there.
The big question to ask is, "Who is the customer?" If the customer is the author, rather than the reader, the publisher bends over backward to make sure the author is happy. Want happy authors? Publish any drivel they spew.
I got my Linux laptop at System76.
"Terry's article also mentions that a recent Wellcome report found that an author-pays business model has the opportunity for a saving of 30 % on publishing costs alone compared to reader-pays."
Some papers take it to the ultimate level. The IOVS (and ophthalmology research paper, huge readership) figured out how to mix all models.
- The reader pays, subscription is compulsory if you are a member of the international society of ophthalmic research.
- The paper is full of advertisements
- It's peer revieved, revievers get nothing
- Authors pay for publication (and the paper puts a footnote for each article, indicating that since the author paid, the article legally is paid advertisement), colour pictures are extra.
- Authors also assign copyright to the paper (don't know the exact terms).
So, the best of all worlds: author, reader, advertiser - they all pay!
3: Profit!!!
The issue with Author pays type systems are that when you start a lab, you have a small startup budget, from which you pay for equipment, consumables, staff, etc. You are then expected to bootstrap your research programme, go out, and get grants to continue its existance.
,that the paper journal is an anachronism, and we'll just nicely format everything for printing in PDF, and move to virtual journals hosted by a distributed network of NSF funded sites. It would be easy to add to the stipulations in the grant process that a university of > N scientists will host a publication repository built along bittorrent lines. Authors will publish to a 'journal', and you'll receive a notice by email that a new 'issue' is available, but it will exist entirely in the ether, where you can read it there, or print it out later, if you're more traditionally minded.
So far, so good, except that you need publications to get the grants, and it's possible that publishing those papers is going to prevent you from having the resources to actually do the research. A couple $4K papers on a theorist startup budget, and I'm out of business.
It may be time to admit, in an era of cheap desktop publishing
Of course, we could also admit that maybe there are just too many scientists out there, everything we need to know has already been discovered, and downsize the great hordes before they discover something else that's going to challenge some long-cherished, utterly wrong, belief.
the more accurate the calculations became, the more the concepts tended to vanish into thin air. R. S. Mulliken
The way to solve this whole problem is that whoever's name is on the copyright notice is the one that pays the cost of publication. If the magazine wants to own the copyright, it cannot charge the author; it can only charge if the author retains all rights. This will solve the problem nicely.
The lessons of history teach us - if they teach us anything - that nobody learns the lessons that history teaches us.