In case you haven't heard, a new law is now in effect that requires all recipients of NIH funding to deposit copies of their peer-reviewed manuscripts into PubMed Central. There's just one problem - if your work is published in an ACS journal, it's the ACS - not you - who owns the right to copy, reformat, and redistribute the manuscript.

Setting aside the issue of whether the new law (PL 110-161) is good public policy, PL 110-161 raises some difficult intellectual property issues. This site has highlighted some of them, and the resources section at the end of this article contains some leading references.

The NIH has solicited public comment on its implementation of PL 110-161, and the ACS has responded. To make the ACS comments more accessible, they have been reformatted and reproduced below in their entirety:

GENERAL OBSERVATIONS & OVERVIEW OF COMMENTS

In the Federal Register notice announcing the Request for Information (RFI) the National Institutes of Health (NIH) poses three questions regarding the new mandate. The American Chemical Society (ACS) is concerned that the questions posed are insufficient to garner the type of input necessary to lead to a comprehensive and meaningful evaluation of the mandate. Further, the format for submitting comments discourages the submission of comprehensive, cohesive and thoughtful input. An examination of input received at the NIH web-site as of the date of this submission seems to substantiate the fact that much of it is far from comprehensive and may be of limited use towards producing a meaningful analysis by NIH at the conclusion of the comment period on May 31, 2008.

The following ACS comments broadly cover all three questions posed by NIH but by extension also cover many issues attendant to a larger more expanded response to each of the questions.

In short, we believe that:

• NIH missed an opportunity to make its 2005 voluntary initiative a success – and has failed to address the underlying problems associated with the voluntary policy and those impediments to success will now become exacerbated under a mandatory policy.
• NIH has not implemented the mandate pursuant to Congressional intent to ensure consistency with copyright law, and instead is forcing the research community to divert time and effort away from advancing the frontiers of knowledge to perform this task in NIH’s place.
• NIH should have followed the federal Administrative Procedures Act, not an RFI, to solicit public comment on the implementation of the mandate. The APA, the federal rule-making standard for over six decades provides a more structured and meaningful process and could have been accomplished within the same timeframe as the RFI.
• NIH must proactively address a number of copyright, intellectual property, and other concerns raised by ACS and other publishers to make the mandate a success (as outlined on pages 5-9)

ACS AND SCHOLARLY PUBLISHING – WHY COPYRIGHT IS IMPORTANT

The ACS is the world's largest scientific society with more than 160,000 members. We care deeply about the advancement of scholars and scholarship and pursue these goals through advocacy, publishing, conferences, information resources and professional development efforts. We have been doing so ever since publishing our first journal – The Journal of the American Chemical Society – in 1879.

Our 36 peer-reviewed scientific journals are distributed globally in print and electronic media and showcase the world’s finest research in chemistry and related sciences. Articles that appear in our journals are widely regarded having received recognition of excellence and the visibility that content in ACS journals receives not only helps scholars achieve new scientific breakthroughs but also leads to practical applications that directly benefit human health and welfare and the world’s economy.

Collectively our peer-reviewed journals form an informal but widely recognized hierarchy used by funding bodies and the academic community itself to assess research quality, impact, and priority—key factors used to allocate funding resources, evaluate levels of personal achievement, and determine professional advancement.

We believe that it is in the public interest to foster this beneficial publishing activity and toward that end we invest heavily in staff and technology resources required to be successful in this endeavor. Copyright creates the opportunity for us to do this by sustaining our publishing enterprise. This is why, we trust, Congress directed NIH to implement its Public Access Policy in a manner consistent with copyright law and respect for its underlying principles and why the Senate Appropriations Committee, in Report 110-107, directed NIH to “seek and carefully take into account the advice of journal publishers on the implementation of this policy” and “to ensure that publishers’ copyright protections are maintained”. We also believe that Congress considers, as we do, that the integrity of intellectual property is an essential criterion for the advancement of science as well as for innovation and creative activity.

IMPEDIMENTS TO THE SUCCESS OF THE VOLUNTARY POLICY STILL EXIST UNDER THE MANDATE

The American Chemical Society supports public access to the results of federally funded research but asserts that the implementation plan for the NIH Public Access Policy mandate does not abide with the law creating the mandate or with the sentiment and direction of the U.S. Congress, particularly as outlined in the Senate Appropriations Committee report that directed NIH to work with scientific journal publishers in implementing the new policy mandate.

The NIH missed an opportunity to make its May 2005 voluntary public access policy a success by not proactively including scientific journal publishers as it developed its procedures and policies for the deposit of manuscripts reporting on NIH-funded research into PubMed Central. Consultation with publishers is critical in 2008 to prevent the agency from embarking on a similar collision course as it proceeds to implement the new mandate. Key to success will be NIH taking an active role, one based on openness and inclusiveness, to resolve the outstanding copyright and intellectual property issues that cut across a very broad and deep swath of the scientific journal publishing community.

The ACS publishes annually approximately 4,000 articles that acknowledge NIH as a research funding source. ACS has tried to resolve outstanding copyright and intellectual property issues with NIH in connection with the Society’s efforts to deposit manuscripts directly with NIH PubMed Central on behalf of ACS authors who have elected to “opt in” to have the Society do so on their behalf. Despite ACS’ efforts, the NIH instructed ACS in December 2005 to suspend article deposition into PubMed Central; that prohibition has prevented the Society from depositing more than 3,000 manuscripts on behalf of ACS authors. During the 2005-2008 time period, NIH PubMed Central has accepted unauthorized postings of ACS copyrighted material, and repurposed and openly displayed such postings without adhering either to NIH’s own policy guidelines, or terms and conditions as set forth to NIH by ACS as rights holder. Issues of concern to ACS remain unresolved and will continue to be problematic and exacerbated under the new mandatory policy as outlined in NIH’s implementation plan issued on January 11, 2008.

ACS hastens to point out that the vast majority, if not all, the Society’s copyright and intellectual property concerns could be resolved if NIH would abide the original intent of the Public Access policy and post without alteration or modification the unedited author versions of peer-reviewed manuscripts on PubMed Central—without any reformatting, repurposing or modification or any mirroring of content to third-party websites—and simply link back to the final published article as the authoritative version for readers on the Society’s own website.

NIH HAS NOT IMPLEMENTED THE MANDATE PURSUANT TO CONGRESSIONAL INTENT

The American Chemical Society (ACS) expresses concern that NIH has not abided by the law in creating the new mandatory public access policy as stipulated in Division G, Title II, Section 218 of Public Law 110-161 (the Consolidated Appropriations Act of 2007)

In enacting Section 218, ACS believes that Congress was aware that flawed implementation of a mandatory public access policy could create serious problems for the scientific publishing community which is why it included the statutory proviso directing “That NIH shall implement the public access policy in a manner consistent with copyright law.”

In its implementation plan published on the NIH website on January 11, 2008, NIH placed the burden of ensuring copyright compliance on the individual researcher or institution, a directive that ACS asserts does not comport with the Congressional intent or guidance.

Shortly after P.L. 110-161 was enacted, ACS submitted a letter to NIH Director Zerhouni wherein we proposed a constructive path forward for implementation of the new mandatory policy in consultation with publishers as rights holders. That letter, sent several hours before NIH posted its implementation plan, asked that NIH seek broad input into the formulation of its implementation plan, and recommended the appropriate method to do this is through the rulemaking procedures under the Administrative Procedures Act (APA), an Act that has guided federal regulatory activities for over 60 years. Following the APA would assure that all stakeholders have an opportunity to provide input into the implementation process and the oversight of NIH’s administration of the policy.

ACS asserts that following the APA would be consistent with Senate Committee Report 110-107 which conveyed with P.L. 110-161. The report directed NIH to take the following course of action when implementing the new mandatory policy. The Report reads, in part:

“…The Committee highly encourages collaborations with journal publishers that would enable them to deposit manuscripts on behalf of the funded investigator, if all parties agree. The committee directs the NIH to seek and carefully take into account the advice of journal publishers on the implementation of this policy.

In particular, the Committee directs the NIH to ensure that publishers’ copyright protections are maintained…”

Following an APA process would also have been consistent with the approach NIH followed when it published its proposed voluntary public access policy on September 17, 2004, in the Federal Register and the public was invited to offer comment. NIH noted in publishing its final voluntary policy in the February 9, 2005 Federal Register that it was not required to follow the APA because of the voluntary nature of the policy, but had done so in order to obtain public comment on the proposed policy. NIH noted that it received over 6,000 public comments at that time. The public comments received were quite thoughtful and provided value to the process, as NIH modified its original proposal and increased the timeframe for manuscript deposition into PubMed Central from 6 to 12 months, citing the need to “ensure that peer review of scientific articles is preserved.”

It would only seem logical and fair that since the policy is now mandatory, and carries with it full enforceability of federal law that its implementation should now be subject to the APA process. That process would allow comment from all concerned parties to ensure the policy is implemented fairly and comports with the Congressional intent of complying fully with the protections that rights holders are afforded under copyright. It is hard to see how NIH’s implementation announcement on January 11, 2008 – 16 days after enactment – followed either the statutory language or the above referenced Senate Committee report language. ACS is unaware of any scientific journal publishers that were consulted in this 16 day window.

By contrast, the RFI process initiated by NIH on March 31, 2008 is a much more informal process than the APA and has only attracted about 1% of the responders to the NIH notice seeking public comment on its voluntary policy as outlined above. ACS believes that the lack of response to the RFI is due primarily to the fact that NIH has already gone ahead and fully implemented the mandate on April 7, 2008 just seven days after announcing the RFI leaving scores of potential responders wondering what value their time and effort in offering thoughtful input might yield.

ACS is unaware of any other federal mandate being implemented in such a disjointed fashion. A policy as important as this should be initiated only after a period of public comment and those comments have been carefully weighed and considered and any amendments found necessary have been made. ACS believes that an APA process would best abide the Congressional intent in both that statute and committee report and does not feel that the RFI will yield the input that an APA process could generate. In fact, an examination of input received at the NIH web-site as of the date of the ACS submission seems to substantiate the fact that much of the input is far from comprehensive and may be of limited use towards producing a meaningful analysis by NIH at the conclusion of the comment period on May 31, 2008.

COMMENTS AND CONCERNS RELATIVE TO COPYRIGHT AND INTELLECTUAL PROPERTY

Since 2005, ACS has made voluntary and good-faith efforts to facilitate the deposit of NIH-funded research into PubMed Central in a manner consistent with the Society’s interests in copyright. We have been prevented from doing so by NIH-generated intellectual property and process-related roadblocks that remain unaddressed by the agency to this day.

We are concerned that such problems will remain unaddressed, and may even be exacerbated under the new mandatory policy. By way of example, ACS has in excess of 3,000 unedited peer-reviewed author manuscripts pending deposit with PubMed Central, as a consequence of NIH’s having refused to accept such deposits from ACS. NIH has rejected ACS’ right, as copyright holder, to establish reasonable safeguards on use of this material.

Instead, NIH has sought to appropriate copyright for itself - reformatting and altering submitted author manuscripts; “repurposing” deposited manuscripts in connection with their display in PubMed Central; and expatriating versions of that repository to countries elsewhere around the globe. In our view, implementing the Public Access Policy in a manner consistent with copyright law, and the intent of that aspect of its Congressional mandate, would mean that:

A. NIH would respect the integrity of the copyrighted content it receives and ensure that any revisions to copyrighted materials such as reformatting, enhancing, linking or otherwise changing the articles are undertaken only when consistent with copyright. Not only are there no mechanisms in place to do this, but also the range of uses outlined in NIH’s terms and conditions for manuscript deposit take substantial liberty with content to create unauthorized derivative works. ACS questions how NIH can proceed in this manner, as doing so would seem to disregard the intent of Congress.

B. NIH would respect ACS’ right, as the copyright holder, to stipulate what it will or will not allow related to third-party use of its works. Instead, NIH has rejected ACS terms and conditions, designed to protect the integrity of the scientific record and, without permission or consultation, has linked content to a variety of online resources (or seeks the latitude to do so). For example, nothing in NIH’s implementation guidelines explicitly prohibits the licensing, selling, or distributing of links or access to content deposited within the PubMed Central database.

C. NIH would acknowledge and support ACS as the copyright holder in the works deposited. Instead, the NIH website directs users to information which we assert undermines ACS’ copyright. In some cases, ACS’ copyright notice is not displayed or NIH’s site links to its own copyright information rather than that of ACS as rights holder.

D. NIH would respect the trademarks and branding of the ACS. Not only has there been no affirmation of these markers of quality, but all too often branding information is missing ? potentially misleading users to the erroneous conclusion that the NIH is claiming copyright, or that the content is in the public domain.

E. NIH would take steps to ensure copyright compliance for material deposited into PubMed Central. No mechanisms to do this are in place, even for content that that has been erroneously deposited by authors without authorization, or that does not fall within the scope of the NIH Policy, and thus should not have been made publicly available. This causes potential economic harm to ACS as publisher.

F. NIH would provide a mechanism to incorporate the concerns of publishers as the policy evolves over time. No mechanisms to do this are in place or have been proposed ? indeed, the implementation guidelines in connection with the mandated Public Access Policy were announced almost immediately after enactment of P.L. 110-161, without consultation with publishers, even as the NIH proceeded to inform other stakeholders.

G. NIH would respect ACS’ right, as copyright holder, to decide how its content will be disseminated. Instead, NIH, without permission or consultation, has made arrangements to mirror ACS content deposited on its site. A mirror site for PubMed Central has been established in the United Kingdom, and our understanding is that other mirror sites are planned or proposed to be located internationally. This raises important questions and concerns regarding copyright protections that would prevail in such circumstances for content located outside the borders of the U.S. We question also how such international mirror sites serve the needs of the U.S. taxpayer and the intended purpose of the NIH Public Access Policy as directed by Congress.

H. NIH would support the integrity of the scientific record. Instead, NIH has declined to use a link to the final published article at a specified URL on the ACS’ own website as an alternative to the deposit and display of the unedited author’s version on PubMed Central. Furthermore, NIH has also chosen to use its own system of article identification (PubMed Identifiers) rather than adopt the widely- accepted Digital Object Identifier (DOI) as a means of identifying authoritative material and associating it with the rights holder of record. This adds to reader confusion as to the definitive version of the article, and by diverting web traffic from ACS’ final published article, poses economic harm in the process. It is unclear to us how NIH will manage and maintain such an identifier system, or the value that this system adds on behalf of the taxpayer.

I. NIH would seek only the deposit of final, peer-reviewed manuscripts upon acceptance for publication. Instead, under the scope of the policy NIH allows, and even encourages, the deposit by authors of the final published version (article) ? without any provision for distinguishing the two versions, or for compensation in recognition of this federal taking of the publisher’s investment in the peer-reviewed version of the manuscript. In so doing, NIH is placing authors in potential conflict with publishers and their copyright or other publishing policies, or (at worst) steering authors to favor journals with policies consistent with a particular business model. ACS questions whether the intent of Congress was to enable NIH to engage in this interference with the private sector and authors’ right to assign and transfer their copyrights in an unfettered manner.

In addition to the issues summarized above, we note that the NIH has not implemented its current voluntary Public Access Policy in a manner consistent with its own self-created guidelines.

For example, articles, including those from ACS, falling outside the NIH Public Access Policy implementation date of May, 2005 have been posted on the PubMed Central repository. Those and other posted articles are made openly available that should have been embargoed for 12 months. Final published articles in journal formal and with content copyrighted by ACS have been converted into NIH’s XML format and posted regardless of publication date. One of our own journal editors has expressed surprise that PubMed Central includes open access to articles he published prior to the policy implementation date ? articles that were posted by others without his knowledge. These experiences indicate that NIH lacks adequate control over the posting of manuscripts on its own website. This must be addressed.

Mindful of these unaddressed implementation problems that affect protections provided under copyright, ACS is concerned that the new mandatory NIH Public Access Policy leaves key policy and implementation questions unaddressed, such as:

1. By what process will NIH establish criteria to ensure that publishers’ copyright protections are maintained? Why has NIH refused to engage in a notice and comment rulemaking that would help to assure publishers that such protections can be put in place?

2. By what process will NIH seek and take into account the advice of journal publishers in determining if it is implementing its Public Access Policy in a manner consistent with protections under copyright law, and the spirit of NIH’s Congressional mandate?

3. If deposited content will be “mirrored” to other sites outside the U.S., what process will be entailed for the establishment of such sites, and how would national and international copyright considerations be addressed to protect rights holders?

4. What limitations, if any, would be imposed on PubMed Central as an “aggregator” of content from sources such as HHMI, Wellcome Trust, other U.S. government agencies, etc?

5. Will NIH negotiate terms and conditions with publishers for the use of NIH grant funds to enable the deposit of copyrighted works on behalf of authors? Will NIH make such payment directly to publishers on behalf of its grantee authors?

6. How will NIH identify grant funds allowable to be used for the payment of publication fees? Will supplemental funds be made available to support author compliance with the mandate?

7. What steps will NIH take if it is found that its Public Access Policy is harming publishers?

8. Why has NIH refused to work with publishers to gather and share PubMed Central usage statistics on copyrighted content? Should this information not be considered as in the public domain, as it is funded with taxpayer monies? Why should articles be freely available, but information about the usage of those same articles be hidden?

9. How will NIH ensure that articles on PubMed Central meet ACS requirements, such as the access-control period, and that the policy actually applies to the articles that it is posting?

10. How will NIH prevent piracy of the articles from PubMed Central? At present, publishers are not protected from systematic downloading that could occur from the NIH website; pirates also could disseminate paper copies of article content, and undermine publishers’ economic interests. What will happen if piracy is discovered as a result of downloading of content from PubMed Central?

11. What provisions will be made to evaluate whether the policy is effective in achieving its intended purpose of promoting public access by US taxpayers to NIH-funded research? As announced, the policy makes no mention of mechanisms for oversight of NIH’s implementation efforts, to ensure that the policy’s scope and operational costs are contained. What “sunset” provisions will be made so that the policy mandate may be amended or phased out if it proves to be ineffective, too costly to maintain, or too disruptive to the peer-reviewed scientific publications on which ultimately it relies?

Regardless of the questions and serious nature of the issues raised above, we choose not to believe that NIH is willfully disregarding copyright law and Congressional intent in the implementation of its Public Access Policy.

However, ongoing consultation with publishers such as ACS is needed to ensure that NIH does not misapply its Congressional mandate and do irreparable harm to the very fabric of scholarly publishing that supports scientific research and our nation’s competitiveness. ACS is willing and able to work with NIH and other key stakeholders to establish the kind of productive ongoing dialog that we feel will truly maximize the sustainable dissemination and discoverability of knowledge in chemistry and the allied sciences, and fulfill the intended purpose of the Congressional mandate in serving the public interest.

Thank you for this opportunity to share the views of the American Chemical Society with you.

Brian Crawford President, Publications Division NIH RFI Comments Results

Resources

• Open Access: Think Globally, Act Locally Short video explaining the core issue.

• A New Beginning or More of the Same? Be careful what you wish for.

• NIH Hears Publisher Feedback on Open Access Mandate A preview of the ACS position.

• Crunch Time: Can NIH Grant Recipients Still Publish in ACS Journals? Something's gotta give.

• ACS and the NIH Public Access Policy: Clarification at Last Policy enabling ACS authors to comply with PL 110-161.

A new law that introduces major changes in the way many U.S. scientific papers are published and redistributed is about to go into effect. Late last year, President Bush signed into law H.R. 2764 (now Public Law 110-161), part of which gives a broad new mandate to the NIH to intervene in the scientific publication system:

SEC. 218. The Director of the National Institutes of Health shall require that all investigators funded by the NIH submit or have submitted for them to the National Library of Medicine's PubMed Central an electronic version of their final, peer-reviewed manuscripts upon acceptance for publication, to be made publicly available no later than 12 months after the official date of publication: Provided, That the NIH shall implement the public access policy in a manner consistent with copyright law.

A new NIH Public Access Website describes how the agency intends to implement the law. Recipients of NIH funds have two options for complying:

1. If you choose to publish your article in certain journals, you need do nothing further to comply with the submission requirement of the Policy. See http://publicaccess.nih.gov/submit_process_journals.htm for a list of these journals.

2. For any journal other than one of those in this list, the author must:

   a. Inform the journal that the article is subject to the Public Access Policy when submitting it for publication.

   b. Make sure that any copyright transfer or other publication agreement allows the article to be submitted to NIH in accordance with the Policy. For more information, see the FAQ Whose approval do I need to submit my article to PubMed Central? and consult with your Institution.

   c. Submit the article to NIH, upon acceptance for publication. See the Submission Process for more information.

The new policy becomes effective April 7, 2008.

In other words, all recipients of NIH funds will soon have an obligation under Federal Law to disclose to journals not on the NIH's list that their work is subject to PL 110-161.

The question is: what will the journals, some of which represent the most prestigious in their field, do with this information?

What Will the ACS Do?

For an organization making a lot of noise recently about its new Web site and focus on communication with its members, the ACS has been very quiet on what what position, if any, it will take regarding the new law.

In fact, from the ACS Homepage, one might get the impression nothing has changed. Looking at the home pages for flagship journals with a large amount of NIH-funded content provided no insights, either; J. Med. Chem, J. Org. Chem., and Org. Lett. have nothing to say on PL 110-161 that I could find.

The ACS author copyright release form doesn't appear to have changed. In other words, when you agree to publish your article in an ACS journal, you're still handing over copyright in your work to the ACS, who has the right under Copyright Law (and presumably PL 110-161) to prevent NIH grant recipients from depositing their manuscript into PubMed Central.

Even the ACS Office of Policy and Legislative & Government Affairs has zero guidance, as of this writing, to offer prospective authors who may have questions about complying with PL 110-161.

Misplaced Burden of Compliance

One of the many problems with PL 110-161 Section 218 is that it places the burden of compliance on authors themselves, not publishers. The law states very clearly that implementations must be "consistent with copyright law." As I wrote previously, this provision gives all the latitude needed to continue business as usual, which is exactly what we're seeing so far.

Two critical questions remain unanswered:

• What obligation, if any, does the ACS have to reject manuscripts from NIH-funded authors, given that it remains ACS policy to take copyright from its authors and with it the right to deposit the accepted manuscript into PubMed Central?

• What obligation, if any, do NIH-funded authors have to avoid publication in journals that strip copyright from them and thereby prevent their ability to comply with PL 110-161?

In partial answer to the second question, the NIH offers this FAQ:

Whose approval do I need to submit my article to PubMed Central?

Authors own the original copyrights to materials they write. Consistent with individual arrangements with authors' employing institutions, authors often transfer some or all of these rights to the publisher when the journal agrees to publish their article. Some publishers may ask authors to transfer copyrights for a manuscript when it is first submitted to a journal for review.

Authors should work with the publisher before any rights are transferred to ensure that all conditions of the NIH Public Access Policy can be met. Authors should avoid signing any agreements with publishers that do not allow the author to comply with the NIH Public Access Policy.

Federal employees always may submit their final peer-reviewed manuscript to PubMed Central, because government works are not subject to copyright protection in the United States.

But even here the language is garbled. Saying that an author "should avoid signing any agreements with publishers that do not allow the author to comply with the NIH Public Access Policy" is not the same as saying authors "shall not sign any agreements with publishers that do not allow the author to comply with the NIH Public Access Policy."

The former describes a suggestion; the latter describes a punishable offense.

Regardless of whether or not PL 110-161 is good public policy, far greater clarity will be needed from both the NIH and scientific publishers if the new law is to be enforced effectively.

Image Credit: wili_hybrid

Disclaimer: I am not a lawyer.

Searching the primary literature on the Internet leaves a lot to be desired. For example, nothing puts the brakes on a search faster than finding a reference to a key paper, but being unable to download it due to the access policies of the hosting journal.

Enter ChemRefer. The site works very simply: type in a keyword and get a list of papers matching it somewhere in the article. Click on one of the search results links, and you'll get the entire article as a PDF file.

What makes this remarkable is that most of ChemRefer's holdings appear to come from copyrighted, subscription-only journals. For example, searching for "taxol" produces links to the full PDFs for papers published in Tetrahedron Letters., Org. Letters, and J. Med. Chem., among others.

It's not clear whether ChemRefer is legally or illegally redistributing these articles. The results are not comprehensive, indicating some form of pre-selection. The journals in question are known for their tough copyright stance and high prices, so there may be a problem regardless of how the material was originally obtained. ChemRefer's founder, William James Griffiths was interviewed on Reactive Reports about his service.

Regardless of the legal status of its holdings, ChemRefer offers an intriguing window into another world. What wonders might result if independent Internet services could mash-up the primary literature at will?

The appearance of chemical blogspace is just one indication that the amount of new content created by scientists but appearing outside of scientific journals is set to increase in coming years. This new content will likely take the form of written work, images, movies, and digital audio. With this situation comes a new problem for authors - managing all of their new content.

By way of Web 2.0 Magazine's Top 100 Web 2.0 Sites, I found a service called Numly that may offer a solution to this problem. Numly enables individual authors to assign unique identifiers that can be used to establish copyright on a work, make licensing to a work known (for example, Creative Commons), provide a way to aggregate all of an author's copyrighted work, and create a permanent URL through which a work can always be accessed.

Numly works by associating a copyrighted work with a Numly Number, which is a unique numerical identifier not unlike a DOI or ISBN. The number can, apparently, be assigned to any work - perhaps even blog comments. Numly may also offer a solution to the problem of self-archiving Open Access articles. As an experiment, I'll be assigning a Numly number to this article.

As with any new service, it's difficult to predict how Numly will ultimately be used. All that can be said for sure is that the future of scientific communication is open. With this move will come new, unforeseen problems that services like Numly may be able to solve.

74817-070104-665306-45



This work is licensed under a Creative Commons Attribution 2.5 License.

Selecting an open-source license is a minefield for which few are prepared when they need to be. There are a plethora of licenses under which open-source software can be released. Selecting a license at the initiation of a FOSS [Free and Open Source Software] project is likely to be a low priority, as there is no initial value to the project. Without a line of source code written, wading through the legalese and nuances of distribution licenses seems unimportant. In reality, the irrevocable nature of the license makes this the most critical time if authors wish to eventually exercise control over derivative works. ... Unfortunately, even the most carefully selected and restrictive license may not afford complete protection from unanticipated and undesired uses.

-Matthew T. Stahl, Drug Discovery Today

Few subjects cause as much confusion and as many heated debates as Open Source licensing. The Open Source Initiative has approved over 50 licenses compatible with their ten-point definition of "Open Source". Whenever that many solutions to a problem exist, it's a sure sign that one size does not fit all. In this article, I'll introduce some of the key concepts in Open Source licensing.

Disclaimer

There is a phrase used so often in discussing the legal aspects of Open Source software that it has its own acronym: I Am Not A Lawyer (IANAL). Clearly IANAL, and chances are that you are not one either. Yet the very acts of writing and using Open Source software require basic familiarity with licensing terms and concepts. My aim in this article is not to provide legal advice, but rather to relate what I've found useful in trying to understand Open Source licensing for my own work. When in doubt, hire a lawyer.

One Good Book

The best writing on the subject of Open Source licensing I've read can be found in the book Open Source Licensing by Lawrence Rosen. An intellectual property attorney, Rosen also served as general counsel and secretary of the Open Source Initiative. His book is remarkably clear and easy to read. If you'd rather not pay for a hardcopy, it can be viewed in its entirety online.

The Good News

Fortunately, all Open Source licenses share some common features, if you know what to look for. For example, most licenses can be divided into one of two major categories:

• Academic Licenses These licenses, named for their original use in universities, allow unlimited freedom to distribute binaries based on altered source code without making these changes public. Examples of widely-used academic licenses include the Apache License, the BSD License, and the MIT License.

• Reciprocal Licenses These licenses require, to varying degrees, the developer of a derivative work to release his or her modifications to the public if their work is distributed. The question of what constitutes a "derivative work" varies from license to license, but most generally involves the modification of the files of a software package. Examples of widely-used reciprocal licenses include the GNU General Public License (GPL), the GNU Lesser General Public License (LGPL), the Mozilla Public License (MPL), and the Common Public License (CPL).

The Importance of Copyright

A frequently-encountered misconception equates Open Source licensing with release into the "public domain." Nothing could be further from the truth. The difference is in the ownership of copyright.

Software in the public domain has no owner. All enjoy unrestricted freedom to copy and otherwise use public domain software. A well-known example is David Megginson's SAX XML toolkit. Megginson, by placing his software in the public domain has forfeited all rights to control how his work is used. Sun Microsystems incorporated SAX into their Java Development Kit without any obligation to Megginson whatsoever. SAX is not Open Source software; it is public domain software.

In contrast, software distributed under an Open Source license remains the intellectual property of the copyright owner. The license is simply a mechanism for the software's creator to give some (or all) of their rights to a licensee, usually in exchange for conditions that must be met. Ownership remains with the creator, who is free do distribute his or her work simultaneously under commercial and Open Source licenses if they so desire.

As you can see, copyright gives a license its legal legitimacy. Far from placing software in the public domain, Open Source licenses use copyright law in the same ways as commercial licenses. This is why understanding Open Source licenses is so important for developers and users alike.

Reciprocity: Share and Share Alike?

Critics of the GPL frequently cite its "viral" nature. The debate essentially boils down to the following paragraph:

You must cause any work that you distribute or publish, that in whole or in part contains or is derived from the Program or any part thereof, to be licensed as a whole at no charge to all third parties under the terms of this License.

-GNU General Public License Section 2a

Like a virus that spreads through replication, the GPL spreads by forcing licensees to release their modifications under the GPL. There are at least two other terms that describe this concept. The Free Software Foundation (FSF) uses the term "copyleft." Lawrence Rosen prefers the term "reciprocity" because of its neutral tone and greater descriptive ability. It's the term I'll also use. Reciprocity is such a fundamental concept in the GPL and other licenses that Rosen's book dedicates an entire chapter to the subject.

Developers distribute their software under reciprocal licenses for a variety of reasons. Two of the most common are:

• To limit "freeloading", or the use of the software by those (typically companies) who contribute nothing back to the developer community.

• To prevent "forking", or the establishment of a competing software package based on the original package.

In reality, Open Source licenses are limited in their ability to prevent either freeloading or forking. For example, provided that a company distributes no modifications to a GPLed package, they are under no obligation to release any of their own source code. Forking happens whenever one or more developers feel strongly enough about a subject to go in a different direction; an Open Source license does nothing to change this.

Given the limitations (and complexities) of reciprocity provisions, one might ask "why bother?". This is an excellent question, the answer to which will depend on your specific goals for your software. And as Stahl points out, the time to make this choice is before a line of code has been written.

Conclusions

Although Open Source licensing may appear to be a minefield, there is nothing mysterious about it. A lot of good writing is available on the subject, with Lawrence Rosen's book being a prime example. If you plan on creating or using Open Source software, learning the basic ideas behind Open Source licensing is a wise investment.