My blogging has been curtailed to maximize my personal time to prepare for the Medical Library Association annual meeting in May and my visit to the Republican Scientific Medical Library in mid-June to teach my Armenian librarian colleagues about identifying and applying health information research techniques to improve Armenian health care outcomes.  I am preparing a workbook and website which requires much time outside of my MLA and Olympic National Park visit.

Rest assured that I should return to active blogging when all my preparations are ready for a successful training activity.  You can read about my previous expedition to the Republic of Armenia here.

Singing Fountains, Yerevan Armenia, June 2008

NIH chief Francis Collins speaks to high school students (SOURCE: USA Today)

On April 20th 2012  NIH Director Francis Collins issued clarification and support for the eventual public release of revised H5N1 journal manuscripts by Dr. Ron Fouchier and Dr. Yoshihiro Kawaoka:

After careful deliberation, the NSABB unanimously recommended the revised manuscript by Dr. Yoshihiro Kawaoka be communicated in full. The NSABB also recommended, in a 12-to-6 decision, that the data, methods, and conclusions presented in the revised manuscript by Dr. Ron Fouchier be communicated fully after a number of further scientific clarifications are made in the manuscript. The recommendation to communicate the research was based on the observation that the information in the revised manuscripts has direct applicability to ongoing and future influenza surveillance efforts and does not appear to enable direct misuse of the research in ways that would endanger public health or national security.

The HHS Secretary and I concur with the NSABB’s recommendation that the information in the two manuscripts should be communicated fully and we have conveyed our concurrence to the journals considering publication of the manuscripts. This information has clear value to national and international public health preparedness efforts and must be shared with those who are poised to realize the benefits of this research.

What Dr. Collins did not share was the reality that there are are already at least 60 articles deposited in PubMed Central that have been funded from US and non-US research grants. Anyone with an email account can establish MyNCBI Collections of research results. I combined a search for Influenza in Birds appearing in the same article with Influenza, Human/virology and limited the result to a variety of publication types for funded research.

View my collection, “Open Access Supported Research on Avian Flu and Human Infection” from NCBI

The precedent has already been well established by NIH that national and international public health preparedness efforts and must be shared with those who are poised to realize the benefits of this research.

Here are video instructions on how to host and build these collections with your MyNCBI account:

http://www.flickr.com/photos/jmsmith000/3590382389/ (CC BY-ND 2.0)

Video has become the medium of choice for education and entertainment,  young and old. Take those staggering youtube statistics:

• 60 hours of video are uploaded every minute, or one hour of video is uploaded to YouTube every second
• Over 4 billion videos are viewed a day
• Over 800 million unique users visit YouTube each month
• Over 3 billion hours of video are watched each month on YouTube
• More video is uploaded to YouTube in one month than the 3 major US networks created in 60 years
• 70% of YouTube traffic comes from outside the US
• YouTube is localized in 39 countries and across 54 languages
• In 2011, YouTube had more than 1 trillion views or almost 140 views for every person on Earth

I see similar staggering statistics, tens of thousands of views annually for relatively mundane instructional videos on library tools, compared to few if any attendees for equivalent medical library drop-in classes.

So it is not really surprising that the most recognized and successful  open access biomedical publisher, Public Library of the Science (PLoS), decided in 2011 to publicize their articles through a new video collaboration with Kikim Media, PBS Video, and the The Alfred P. Sloan Foundation to produce Science Bytes. These are a series of short videos, based on recently published studies from PLoS, which highlight discoveries that can naturally engage a general audience. The executive producer of this  series, Micheal Schwarz, says, that this is “ next step in the natural evolution of science communication.”

Science Bytes, like PLoS articles, is an open access resource. You are free to to place this video on any site you wish, with credit for where you found it. This is the one I discovered tonight, the third in the series, on diabetes, diet, and reversing kidney damage in mice (and perhaps humans):

This is the article:

Poplawski MM, Mastaitis JW, Isoda F, Grosjean F, Zheng F, et al. (2011) Reversal of Diabetic Nephropathy by a Ketogenic Diet. PLoS ONE 6(4): e18604.doi:10.1371/journal.pone.0018604

It also dawned on me that this episode was published back in September 2011, more than 6 months ago.  Was the plug pulled?  I hope not.

There are more than 50 questionable open access publishers on Jeffrey Beall’s List of Predatory Open-Access Publishers.   Some questionable journals publish independently of any publisher.  How has this disease spread? Here are my thoughts and evidence:

1. Plug-and-play content management such as Open Journal  Systems (OJS) provides a no-cost easy way to set up a professional journal publishing platform.  For better or worse, this respected attempt by the Open Knowledge Project aimed at ” improving the scholarly and public quality of research” has simultaneously offered a fast-track opportunity for publisher copycats to launch an “open access” scholarly publishing operation. To be fair to OJS, there are more than a dozen publishing software packages listed in the Open Access Directory launch by Simmons College.
2. Nearly every journal concerned with their reputation in fields such as medicine makes considerable effort to adopt a credible peer review mechanism, but the traditional method of author-blinded reviewing does not provide a way for the submitting author to see the identity of the reviewer who detected deficiencies or judged their research deficient.  A journal admitting the lack of sufficiently knowledgeable reviewers may ask an author to provide names and contact information of potential reviewers in her/his specialty, creating a different kind of competing interest.
3. When one of the questionable publishers recently sent me an email solicitation to publish in one of their journals, I also noticed that they charged an author-processing charge (APC) of $400-$600.  That amount is very low, considering that open access journals with strong peer review reputations have a business model that requires 200%-500% of that amount for an APC.  Even if publication costs other than peer review can be reduced to lower the APC, maintaining a system of credible reviewers that provide careful unbiased analysis is an ongoing expense of time and effort.
4. A code of conduct published by the Open Access Scholarly Publishing Association (OASPA) is voluntary, and the OASPA itself has not taken fellow publishers to task about the predatory evidence that seems to exist for more than 50 questionable open access publishers
5. The stagnant global economy has created the largest unemployment rates in modern history, and would-be entrepreneurs with computer skills and time on their hands could find opportunity by simulating a closed system of peer review and rapid publishing turn-around, satisfying  the need of researchers to find a publisher when competition for a place in important journals is very competitive.

So the problem is that there are plenty of would-be biomedical open access publishers, but on closer inspection many do not meet a trust-able standard.  Perhaps there are genuine publishing efforts and a desire to be removed from the category of questionable or predatory.  I think there is something to do to create trust that even goes beyond aligning with the OASPA code of conduct:  adopt careful and scientifically-based peer review that is open and available for authors, readers, and the institutions that are providing APC subsidies as part of the Compact for Open-Access Publishing Equity (COPE).

I encountered open peer review more than 9 years ago, as the editor of a new open access journal on the Biomed Central (BMC) platform.  BMC announced that journals could consider publishing the peer reviews as part of an article’s history, signed by the review authors. The editorial staff I initially led declined to make open peer review mandatory, but at least one BMC journal has embraced it. Here is their own description:

Biology Direct offers a novel system of peer review, allowing authors to select suitable reviewers from the journal’s Editorial Board; making the peer-review process open rather than anonymous; and making the reviewers’ reports public, thus increasing the responsibility of the referees and eliminating sources of abuse in the refereeing process.

To see how this works in practice, take for example an article from 2011: Not all transmembrane helices are born equal: Towards the extension of the sequence homology concept to membrane proteins. The reviewers’ comments are signed and in the public view. Here is one:

Reviewer’s report 1

Shamil Sunyaev, Division of Genetics, Dept. of Medicine, Brigham & Women’s Hospital and Harvard Medical School

This manuscript investigates differences in sequence complexity and hydrophobicity between transmembrane helices serving purely structural role and transmembrane helices with additional functional roles. On the practical side, the analysis of sequence complexity and hydrophobicity is able to identify transmembrane helices that are responsible for spurious sequence search hits as opposed to functional transmembrane helices that are useful for homology search. The manuscript is a follow up on an earlier manuscript by the same authors that suggested suppressing transmembrane segments in domain models to increase sensitivity and specificity of remote homology search. The results are of interest both in terms of protein evolution and in terms of practical utility for sequence similarity searches.

Authors’ response

We thank the reviewer for highlighting these points. Even if the straightforward, silent extension of computing homology via sequence similarity from globular domains to membrane proteins is probably correct in a number of cases, it is important to identify this uncertainty as blind assumption and to derive criteria aimed at excluding situations when assigning homology in this simplified manner is not justified.

I have two minor comments:

1) Since sequence complexity and hydrophobicity seem to be correlated, I wonder why is not the inverse co-variance matrix included in the Z-score.

Authors’ response

If the distribution of the points (x̃Φ,x̃c)(we use the normalized forms x̃Φ=xΦ−μΦ∕σΦand x̃c=xc−μc∕σc) is considered in the hydrophobicity/complexity plot (Figure 3A), we find that most of the points representing real TMs are either of high hydrophobicity/low complexity (simple TMs or of less hydrophobicity and higher complexity. The two alternative quadrants (low hydrophobicity/low complexity or high hydrophobicity/high complexity) are not much populated and, from the viewpoint of the concept developed in this article, it is not very clear how to deal with them meaningfully. Of course, one can develop analytically more sophisticated expressions for the Z-score as proposed by the reviewer and, maybe, it might prove useful in the future. Here, we wanted to have a very simple form of the Z-score that is symmetric across the four quadrants. At the end, we are only after TMs with low Z-score.

2) On a more philosophical note, I would not state that “homology” equals to common evolutionary origin but rather than ‘homology” is a surprisingly high level of similarity that cannot be explained by a functional constraint and, therefore, indicative of potentially common evolutionary origin.

Authors’ response

We fully agree with the reviewer’s view. The problem is in the practical detail: How do we know that there is no functional constraint for query sequences (whether there is convergent evolution or common ancestry)? From the sequence analytic perspective, homology is usually inferred from easily derived similarity measures of aligned sequences because there is no other direct measure of common ancestry, not because this criterion is especially elegant or rigorous. The practice becomes questionable when this caveat is forgotten.

It may occur to you that a well-written review becomes a scholarly publication, much in the way a book review does in other disciplines.

One potential downside is the same concern raised by my own editorial staff. In a relatively small community of like-minded professionals, would an honest review that appeared harsh not be appreciated in the spirit of improvement and/or considered a threat to friendship? I guess we have to ask ourselves why more open access journals did not adopt open peer review, once someone as respected as NCBI Director David Lipman became the editor of Biology Direct.

You can see from the example above that editorial guidance for the reviewer can also filter out harshness and really create an opportunity for dialogue that improves the overall process of research in a particular discipline, perhaps even pleasant exchanges after publication.

Today I met the Chief Medical Information Officer for Yale-New Haven Hospital, Allen L. Hsiao MD, who is also an Assistant Professor for the Yale School of Medicine specializing in Pediatric Emergency Medicine.  Dr. Hsiao was was talking to the Yale Medical Library staff about the implementation of an electronic medical record (EMR) system for our hospital network, including background on the incentives (carrots and sticks) the federal government is offering for health care organizations that adopt effective EMR systems and practices to become more efficient and improve patient outcomes.

I figured that as an active academic clinician, Dr. Hsiao had written some articles, so I checked for his scholarship in PubMed. Dr. Hsiao is a co-author on 14 articles in PubMed. Half of them are either published in open access or offered publicly as free access by their publisher.  Using Google Scholar, I decided to take a peak at how many times each article had been cited.  Here’s my survey:

• Adding up the times cited for the first seven open or free access articles and dividing by seven, Dr. Hsiao’s average rate of being cited is an impressive 30.3 times per article.
• Adding up the times cited for the second seven subscription or toll access articles and dividing by seven, Dr. Hsiao’s average rate of being cited is less impressive:  only 13.4 times per article.

Seems like there is an open access advantage, even considering some nuances of bias. Maybe its time to take a look at Alma Swan’s The Open Access citation advantage if you still have a hard time accepting my conclusion.

I wonder if other faculty at my school want to be put through a similar assessment.

I saw an account of three major publishers’ copyright infringement law suit filed against open education publisher Boundless Learning (BL) in the April 5th Chronicle of Higher Education blog.   If you launching an open education startup, viral word-of-mouth marketing is the inexpensive alternative to advertising.  But then you attract venture capital funding and a law suit in the same week,  articles show up in The Chronicle and even TechCrunch.  Free publicity…. priceless.

What’s the beef the publishers have with BL?  They accuse BL of:

• Stealing the creative expression of others
• Violating intellectual property rights
• Exploiting and profiting from Plaintiffs’ successful textbooks by making and distributing the free versions of those books
• Build a business upon Plaintiffs’ intellectual property rights

The example cited in The Chronicle article is about a popular biology textbook:

To illustrate this claim of intellectual theft, the publishers’ complaint points to the Boundless versions of several textbooks, including Biology, a textbook authored by Neil Campbell and Jane Reece. The Boundless alternative, the complaint alleges, is guilty of copying the printed material’s layout and engaging in what the complaint calls “photographic paraphrasing.” In one chapter of the printed book, for instance, the editors chose to illustrate the first and second laws of thermodynamics using pictures of a bear running and a bear catching a fish in its mouth. Boundless’s substitute text uses similar pictures to illustrate the same concepts—albeit Creative Commons-licensed images hosted on Wikipedia that include links to the source material, in accordance with the terms of the open license.  (source)

source: http://www.flickr.com/photos/aptmetaphor/2316773059/

This bear is not running or catching a fish. This bear is bushed. I think this argument is too. At least two general categories of material are generally not eligible for federal copyright protection, according to the government’s Copyright Basics (pdf):

• titles, names, short phrases, and slogans; familiar symbols or designs; mere variations of typographic ornamentation, lettering, or coloring; mere listings of ingredients or contents
• ideas, procedures, methods, systems, processes, concepts, principles, discoveries, or devices, as distinguished from a description, explanation, or illustration.

The idea of a bear’s physical activity being used as an analogy for the laws of thermodynamics is not copyrightable.  And if the bears are different and the other diagrammatic expressions integrated with the bear not identical, is this any different than the similarity that already takes place between competing commercial textbook publishers trying to illustrate principles of thermodynamics?

Anyway, this is all speculation based on the outlines of a lawsuit and  a copies of an open textbook that is still effectively gated for currently in invite-only beta. BL is happy to have all this attention and the roll of David against a Goliath consortium.   There are also decent questions about the quality of the BL products, and I have my own opinion about certain free open access textbook publishers around quality and the opportunity to exploit scholarly authors.

On March 29, 2012, at 9:30am, the US House of Representatives Committee on Science, Space, and Technology, Subcommittee on Investigations and Oversight held a public hearing to examine Public Access and Scholarly Publication Interests.

Two pieces of legislation provoked this hearing.  In early February House Representatives Mike Doyle (D-PA), Kevin Yoder (R-KS), and Lacy Clay (D-MO) reintroduced the Federal Research Public Access Act (FRPAA), a bill that would ensure free, timely, online access to the published results of research funded by eleven U.S. federal agencies.  This legislation frames the issue of public access as a taxpayer expectation.  The Committee also expects to receive a report in the coming weeks on this issue from the Office of Science and Technology Policy, as directed in the America COMPETES Reauthorization Act of 2010, which addresses U.S. research and development  to improve the competitiveness of the United States. This frames public access as either an enabler of research and development or a threat to the competitiveness of US Publishers, in particular scholarly society publishers.

Although the distinguished speakers had only about 5 minutes to summarize their perspective, each was allowed to submit written remarks that become part of the hearing record.  Here are the speakers providing testimony and a link to their submitted remarks.   I also took the liberty to copy the main remarks of each speaker  and use the tag cloud site TagCrowd to create a word visualization of the each speakers 50 most used words (eliminating common words, and a word has to be used a minimum of 5 times).  Here is the testimony links (click on the witness name for the PDF of the submitted remarks)  and results:

Witnesses

This statement is a direct quotation from a 2006 article “Scientific Journals are ‘faith based’: is there science behind Peer review?” by a group of public health upstarts suggesting a lack of scientific method in the peer review process could be remedied by alternative methods of scholarly quality control and new forms of data-driven review.  Predictably, a immediate respondent to this article questioned the authors’ research rigor, yet also sought common ground to “agree on the objectives of peer review and develop appropriate validated tools that can measure its effects.”

Oh, there is one other thing I noticed. The respondent admitted her competing interest as an employee of  a journal publisher, hired to conduct research on peer review and publishing.  Nice that we know it.  In most scenarios of journal peer review, there is no ability to see or know if  there would or had been a competing interest among a particular article’s peer reviewers.  A few experiments with and discussion of open peer review have not led to a trend or shift in any large measure.

The most promising game-changer in bringing a scientific method to scholarly publishing today was probably the decision of PLoS ONE to offer two-stage peer review consisting of initial determination of scientific methodological soundness, then allowing the post-publication usage metrics to provide  usage-data-driven evidence of article significance to a discipline.  In their 2009 debut of the metrics, the PLOS blog explained:

First, we are focusing on articles rather than journals. The dominant paradigm for judging the worth of an article is to rely on the name and the impact factor of the journal in which the work is published. But it’s well known that there is a strong skew in the distribution of citations within a journal – typically, around 80% of the citations accrue to 20% of the articles. So the impact factor is a very poor predictor of how many citations any individual article will obtain, and in any case, journal editors and peer reviewers don’t always make the right decision. Indicators at the article level circumvent these limitations, allowing articles to be judged on their own scientific merits.

PLoS ONE’s decision to implement the scholarly “wisdom of crowds” has attracted both authors and readers less concerned with the journal brand and more concerned with immediate access, methodological credibility, and usage. Take for example a 2011 article in PLoS ONE: NSAID Use Selectively Increases the Risk of Non-Fatal Myocardial Infarction: A Systematic Review of Randomised Trials and Observational Studies.   Here are the metrics for the article as of today:

Click to View Original Page

The metrics show positive trend, times cited from multiple sources, and impressive usage statistics.  SPARC will host Building New Measures for Impact: Article Level Metrics,  a webcast of April 12th that will focus on information on article level metrics,  with Peter Binfield.  Yes, I know what you are thinking.  His competing interest is that he is merely the publisher of PLoS ONE. At least we know it and can take it into account.

Here are the top 10 clinical genetics journals for 2011,  according to SCImago Journal Rank for today, limited to journals with at least 100 articles over the last three years:

1. Nature Genetics is the highest ranked journal by a wide margin.  In a reputation strong position, Nature Publishing Group (NPG) does not offer optional gold open access to authors of articles in Nature Genetics, as is done for some other NPG journals. NPG does offer publisher processing for NIH-funded articles bound for PubMed Central.  Authors do not assign copyright for NPG articles, though they also grant NPG an exclusive license to publish. NPG also offers guidance and support for author self-archiving after a 6-month embargo.  Stevan Harnad took them to task for their resistance to immediate self-archiving and self-labeling their efforts as “liberal.”  I would give them an Openbiomed open access grade of  C.
2. PLoS Genetics, published by the Public Library of Science (PLoS), offers the a gold open access business model which funds the enterprise with an author fee.  For PLoS Genetics, the current publication fee is US$2250. Authors who are affiliated with one of our Institutional Members are eligible for a discount on this fee. PLoS applies the Creative Commons Attribution License (CCAL) to articles, and authors retain ownership of the copyright.  Articles are automatically eligible for immediate self-archiving. An Openbiomed open access grade of  A is well-deserved.
3. Nature Reviews Neuroscience also comes from the NPG, part of the Nature Reviews family.  According to this page, In 2009 NPG rebranded the Nature Clinical Practice journals into Nature Reviews, for eight clinical specialties. For authors, no particular open access enhancement in comparison to Nature Genetics. The Openbiomed open access grade of  C also applies.
4. Circulation: Cardiovascular Genetics is published by the American Heart Association and is representative of journal literature publishing the results of translational science that seeks to move laboratory discoveries aggressively into clinical practice. All AHA journals use the same author copyright transfer agreement, which causes the author to give away control and publication rights,with the one exception that  the author may post an accepted version of an article in a non-profit funders web archive and/or an author’s institutional repository, no earlier than 6 months after publication.  Nothing said about a personal web site. The Openbiomed open access grade is  C- .
5. Molecular Biology and Evolution (MBE) is published by Oxford University Press under the umbrella of Oxford Journals.MBE is one of the more than 100 journals that offer the Oxford Open option for a publication charge of £1700/$3000/€2550.  Authors from developing countries receive a significant publication charge discount. Post-Print version of articles can be placed on personal websites, your institution’s website and in institutional or subject-based repositories such as PubMed Central, as long as a 12 month embargo is respected. Authors do not assign copyright, though they also grant OUP an exclusive license to publish. The Openbiomed open access grade is  B .
6. DNA Repair is a subscription journal from Elsevier. Optional sponsored open access is available for this  journal. Only 2 articles in 2010 from this journal were sponsored open access. The charge for the sponsored option is us$3,000 plus tax where applicable.  Elsevier is fully green, which may be a reason why the number of sponsored articles is so low. Elsevier is also a full partner in Research4life, (as is OUP) and most of their article  reach developing nations that way.  The Openbiomed open access grade is  B+.
7. The Orphanet Journal of Rare Diseases(OJRD) is an open access journal on the Biomed Central(BMC) open access platform. An article processing charge of £950/US$1500/€1145 is required, though an institutional membership in BMC can provide a discount. Authors retain the copyright to their work, licensing it under the Creative Commons Attribution License which allows articles to be re-used and re-distributed without restriction,provided the original work is correctly cited. BioMed Central is owned by Springer Science+ Business Media. With a APC fee substantially below PL0S, this journal rates an Openbiomed open access grade of  A+.
8. BMC Medical Genomics is another open access, peer-reviewed journal on the Biomed Central platform. Blessed with all the open access attributes of BMC, with an APC of £1230/$1945/€1485.  Being a little more than OJRD without a clear compelling reason, his journal rates an Openbiomed open access grade of  A.
9. The American Journal of Medical Genetics Part C: Seminars in Medical Genetics is a subscription journal from Wiley-Liss, a imprint of Wiley-Blackwell.  Wiley-Blackwell offers a hybrid OnlineOpen access option for over 500 journals, including this one. The OnlineOpen APC fee is fixed at US$3000 for most journals.  Wiley-Blackwell expects to receive either a signed Copyright Transfer Agreement  or an Exclusive License Form before publication; in most cases this is the copyright transfer. Only OnlineOpen submission get explicit permission to post the  final PDF of the contribution on a personal website or in a personal repository. For non-open articles, only a pre-print copy is allowed in a personal website or repository. The Openbiomed open access grade is  B.
10. Evolution, the International Journal of Organic Evolution, is a subscription journal from Blackwell, acquired by Wiley.  The open access opportunities and constraints are similar to  The American Journal of Medical Genetics Part C: Seminars in Medical Genetics  , including the OnlineOpen option. Same open access grade: B .

Only 3 journals rated a C or less, and there were a few A‘s and several B‘s.   A guess I am an easy grader.

Michael P. Taylor,  one of the Sauropod Vertebra weekly image-posters that recently emerged as a commentator on scholarly communication from his academic vantage point,  made a worthy attempt this week at telling readers of TheScientist  about the dysfunctional system of academic publishing that continues to be idealized, resuscitated, and defended, albeit mostly by publishers and their lobbyists.   No competing interest there.   

Mr. Taylor suggests (and I agree) that there probably is more ignorance than fear and loathing among faculty ranks.  This is something that librarians can fix.

Taking a look at the March 5th anti-FRPAA letter being trotted around by lobbyists, there are probably about 50 scholarly societies and associations on the signatory list. Each of these organizations represent dedicated faculty providing the labor-intensive peer review and editing activity in what little time can be gleaned from teaching, mentoring, grant management, and their own scientific research.  Senior faculty provide credibility, authority, and status for journals published by a scholarly society. Junior faculty are concerned about their own academic standing and success that become evidence for a permanent academic appointment.

Every faculty member is connected to libraries and librarians that provide resources and services.  Library liaison and outreach activities increasingly include NIH open access policy publication support, knowledge management education for institutional translational science programs, and essential literature searching for evidence-based systematic review development. In my opinion,  faculty collaboration with  information professionals during the scholarly processes of discovery, organization, writing, and publishing  is more interactive and appreciated.

Librarians can also educate their faculty constituencies on scholarly communication issues, the risks of substandard peer review, and the costs of providing knowledge to their academic community. Here is some of the scholarship from the library and academic medicine community that cites the growing influence of librarians on academic publishing and scholarly communications.

Brower SM. Medical education and information literacy in the era of open access. Med Ref Serv Q 2010 Jan;29(1):85-91.(subscription required)

Koehler BM, Roderer NK. Scholarly communications program: force for change. Biomed Digit Libr 2006 Jun 21;3:6.

Plutchak TS. Searching for common ground: public access policy and the Scholarly Publishing Roundtable. J Med Libr Assoc 2010 Oct;98(4):270-272.

Sperr EV,Jr. Libraries and the future of scholarly communication. Mol Cancer 2006 Nov 7;5:58.

Koehler and Roderer speak most directly to the type of initiative that I envision for every academic medical center library:

• Conversations and forums to raise scholarly communication issues,  educating faculty about the issues and encouraging  appropriate practical actions.
• Web sites to promote awareness of scholarly communications issues, initiatives, and practices while offering an interactive space for users’ responses.
•  Detailed information conveniently arranged about a journal’s publisher, impact factor, local usage statistics, and opinions of local authors who have published in particular journals to comment on their own experiences with publishers.
• Using all types of student and faculty encounters to make sure that scholarly communications brochures were available and that these meetings provide another opportunity to educate constituents to the changing world of publishing.
• Consistent and persistent faculty engagement.

The Koehler and Roderer article concludes:

Scholarly communication is approaching a crossroads. If the current system is unsustainable, as many believe, and if technology has changed the landscape of publishing, then a time for serious decision-making is at hand. There is a role for the libraries and librarians in this enterprise – continuing to support authors and to disseminate scholarly communications information until there is an economically sustainable system that provides the widest possible access to scholarship. The goal of the Hopkins’ scholarly communications initiative has never been to undermine the world of scholarly publishing. It is not necessary to make everything free for libraries or to put publishers out of business. Indeed, our goals are to ensure that Hopkins authors know what their rights are, that they manage their own work in a way that benefits science as well as their own needs, that they understand the business plans and philosophy of the journals they work for, and that they take control of their own publishing destinies.