• I’m writing a paper.
• I’m still taking that Science Journalism course, and working on a final ~3,000 word piece, which I’ll put up–here or elsewhere–when I’m done. I can tell you it’s about some brilliant research coming from the lab of Dianne Newman, an MIT Professor.
• As usual, I’m banging drums in an MIT jazz combo. This term we’re playing, among others, the James Brown song “Mother Popcorn,” and it’s sooo funky.
• Other miscellaneous debris.

To tide readers over until a more steady stream of original content appears, I am posting something I wrote three years ago, when I was a wet behind the ears first year graduate student. The Department of Biology has a wonderful class, only for the first year grad students, called “Methods and Logic in Molecular Biology” (colloquially known as “seven-fifty” or “Methods”), an intense paper reading course led by several faculty. (Actually, eventually I should probably write some posts about these classes for potential students or others who are interested?)

Anyhow, our section for Methods became somewhat tight, and occasionally we exchanged emails about the current week’s assigned papers. Around 2am on the day of the last class of the semester, I sent the following email to my section. Clearly I was high on something–not a controlled substance; possibly a couple beers; likely joy at almost being done with the class/semester; as likely rebellion against being told what to read, instead choosing to read what I wanted to. Most of my classmates had already exhibited in spades dysfunctional behavior, it was my turn. I still think it’s a stimulating read:

On the eve of our last class, instead of re-reading the papers I did some Internet research into the fascinating area of honeybee genetics. Topic is more interesting than heat maps or MALDI experiments. Some things I found:

In a bee colony, there are three types of bees: few female queens, hundreds of male drones, and thousands of female workers. Females are diploid and males are haploid. Females develop from fertilized eggs. Haploid male drones develop from unfertilized eggs, and therefore they have no father! Sex determination is made at a single locus, the csd gene, of which at least 19 alleles are known. It seems that all alleles can be found in males and females. It was also shown that once activated, csd remains active throughout development. RNAi inactivation of csd causes diploid female eggs to develop male gonads, but does not affect haploid male egg sexual development. Therefore it has been hypothesized that 2 different alleles of csd somehow result in two protein products that can interact together to direct a specific step in the sex determination pathway towards female development. Hemizygous csd eggs cannot make this product, and thus the default state is male.

Female queen and worker bees develop from queen bee eggs fertilized by drone sperm. Females must be heterozygotes for csd alleles to survive. Diploid flies homozygous for a csd allele develop into sterile males, but soon after these larvae hatch from the comb, they are selectively removed and destroyed by worker bees (not sure how workers can recognize these larvae). (This also makes it difficult to develop inbreed stocks of honey bees, colonies die out quickly due to loss of csd homozygotes.) Since both queens and worker females come from fertilized eggs, what distinguishes them is that between larvae and pupa stages, queens receive a hormonal mixture called the “royal jelly”, whereas workers arise from larvae that have been denied this. Workers are sterile because they don’t develop ovarioles, and only live a few weeks. Queens usually mate once in their life and then live for years.

Queen bees must mate with many drones at one time early in there lifetime, and must do it 50-100 meters in the air and kilometers from their colony! (This makes it difficult for bee breeders to maintain isogenic stocks of bees, an intensely studied research problem in bee genetics.) The drones die after mating, and the queen returns to hive and doesn’t need to mate again. She will produce thousands of offspring from eggs fertilized from perhaps 5-15 drones. From an evolutionary perspective, the fact that she usually mates with multiple partners once early in life, and far from the hive prevents her mating with her own son, reducing the chances of producing half inviable progeny homozygous for csd allele, (which means fewer worker bees to support the colony). Pretty cool, huh.

Oh yeah, consider this my contribution to Thursday’s discussion.

Sorry, but I can’t remember my references.

Give this man a Nobel Prize. Give it to him.

I know before I’ve made clear my affection for Harry Noller, and that affection still remains strong like a peptide bond, but lately I’m head over heels (head over sneakers) for Venki Ramakrishnan. Last week in his lecture for the MIT Biology Colloquium, Venki Ramakrishnan charmed me and several hundred other people with his humor, smarts, and beautiful structural work.

The scene of Venki’s lecture, titled “How the ribosome facilitates selection of the right tRNA during decoding of the message,” was quite a spectacle. There was an electricity in the air. Never had I seen room 32-123 so packed. Every seat was taken, of course, and there were at least one-hundred other people huddled in the back of the lecture hall, down the stair aisles, in front, everywhere. Some professors were seated on the concrete floor.

Venki’s faculty host had warned the audience before the lecture began that the aisles had to be clear (for fire safety reasons), and so they were cleared. But sure enough, ~10 minutes into Venki’s lecture, the honorable MIT campus police unkindly entered the room and, temporarily, ruined some beautiful science.

It was quite funny: Venki was captivating us from the lectern, as he faced a projection screen to his left. To his right, a plump MIT police officer sauntered in, unbeknownest to Venki, but knownest to everyone else in the room. The copper reached his arm out to the lectern to capture Venki’s attention, Venki stopped talking, and the officer motioned to follow him outside the lecture hall. Totally perplexed, Venki obliged and left the room, to a chorus of boo’s directed at the police. Moments later, Venki emerged calm as a clam, and succinctly directed movement of his audience into a fire-escape safe arrangement so that his lecture could continue.

Imagine what Venki’s story could sound like: “I won the Nobel Prize, went to MIT, and was accosted by the campus police at my own lecture!”

Venki gave a beautiful introduction, even making a jab at Jim Watson (Watson the man, not Watson the scientist). (He also later hilariously and appropriately mocked Tom Steitz.) He then proceeded to give the best structural biology talk I have ever seen.

He described how proper base pairing between the tRNA anticodon to the mRNA codon induces subtle structural movements between that end of the tRNA and small subunit RNA that are transmitted up through the tRNA toward its aminoacyl end, inducing residue movement in EF-Tu leading to GTP hydrolysis–a cascade of events leading to EF-Tu release and aa-tRNA incorporation. (For more, see Venki’s recent review.)

He ended by narrating an incredibly cool animated movie of all the ribosome structural movements he had just described in detail, and then reprised the movie with a version set to a soundtrack of snippets of classic pop tunes (e.g. by The Clash, David Bowie, etc.), arranged by his lab. The lyrics spoke to the molecular movements spotlighted in the movie. It was very entertaining.

I realize I could have proclaimed Venki the “Rolls-Royce” of ribosome investigators, since he’s at the MRC. But no. He’s American; he’s a Caddy.

Bartel focused on “ribozymes,” a class of molecules composed entirely of ribonucleic acid (RNA) that catalyze chemical reactions. Starting from a synthetic large pool of different RNA molecules of random sequence, he used a technique called in vitro selection to isolate a new type of ribozyme that performed a specific chemical reaction, known as an RNA ligation, otherwise only attributed to natural protein based enzymes. He then evolved these ribozymes by introducing random mutations into their sequence and re-isolating more active versions that outcompeted earlier variants—in effect modeling evolution in a test tube.

Naturally occurring ribozymes are rare—a host of enzymatic activities they may have once possessed have been appropriated to faster and more accurate proteins during evolution—but owing to their structural similarity to DNA, and catalytic potential, they are attractive candidates for the original self-replicating macromolecules, predating more complex life forms.

While generation of a self-replicating ribozyme remains elusive, and would not prove that catalytic RNAs directly preceded life, Bartel’s study was paramount in demonstrating the ability to isolate ribozymes with new activities. He notes, “We’re left with lots of gaps in knowing how RNA based life forms might have emerged, but it is still useful to know what RNA can do.”

Appreciating the highly mutable nature of evolution, Bartel also recognizes that the particular ribozyme he isolated was only one possible outcome of the experiment. “Evolution is the sum of a huge number of chance events, and the biological world is the culmination of all these chance happenings,” says Bartel. “If you replayed a tape of biological evolution, we would all look different.”

Reference: Bartel and Szostak, Science 261, 1993.

I wrote the piece above for another assignment for my science journalism class. The assignment was a “front of the book” magazine piece, written for an audience interested in science, but probably not very knowledgeable about RNA or ribozymes.

We were instructed to pick a paper we are (were, have been, etc.) excited about, and interview one of the authors or an expert in the field. Ribozymes are what first sparked my interest in RNA back in college, so I picked my favorite ribozyme paper. And conveniently, being at MIT, I had the opportunity to interview Professor David Bartel, the first author of the paper.

To accomplish this task, miRNAs chemically pair with messenger RNAs (mRNAs), the molecular intermediates between a gene’s DNA sequence and its protein product. This pairing event essentially sequesters the mRNA, preventing it from being decoded to produce protein.

Recognition of an mRNA by a miRNA is imparted by complementary molecular motifs present in both molecules. These molecular motifs are encoded in the DNA sequence of a genome.

MicroRNAs compose a class of gene regulatory molecules widespread and evolutionarily conserved throughout plants and animals. While miRNAs’ effects on gene expression are modest compared to other modes of gene regulation, en masse their total contribution to gene expression in the cell is significant.

The ~150 word definition of microRNA above I wrote for an assignment in a science journalism class I am taking this semester. (Incidently, you may have noticed the frequency of my contributions to this blog has slowed in recent weeks, resulting from an increased non-blog workload. Sorry.)

The assignment was to write an “explainer” for a key scientific term, like you might see set aside from a longer article (in a Scientific American or Discover type magazine) concerning some scientific idea, that references the term. Thus it is meant for a audience interested in science, but not necessarily familiar with the scientific details of whatever topic is under discussion.

My class had mixed reaction to my explainer. The paragraphs were deemed somewhat disconnected in content from each other, perhaps resembling more of a list of facts in the form of sentences. Some found the voice dry and boring (like that of a scientist’s?).

My goal in writing it was to define microRNAs: (1) What they are; (2) What they do; (3) How they function; (4) Why they are important. I could have started with the importance, but the way I wrote it is the way my brain organizes thoughts. I should learn to think like my reader, and also make the language a little more, um, charming.

Weinberg recounted one story about how, in the early 70′s, he crossed paths with a fraud who had shared his intense interest in the mechanisms by which infection with DNA tumor viruses could lead to transformation of mammalian cells. Weinberg had thought of several experiments to address the question, only to find out soon after that another investigator in Toronto had beat him to it, becoming a minor celebrity while giving talks up and down the east coast. However, the unthinkable quantity of data this investigator had amassed would soon end his career. (This outcome is assumed as Bob did not provide a name.) The data was fabricated. The first hint resulted from a back of the envelope calculation by a journal editor showing that the number of petri dishes required to generate the data far exceeded the number available to all scientists in the Toronto area during the relevant period.

Hot fields are of course vulnerable to fraud. Recently there was Woo Suk Hwang in the stem cell field, and Luk Van Parijs ran a lab studying RNAi technologies and immunology. Back in the late 80′s/early 90′s there was the notorious Imanishi-Kari/Baltimore affair. I suppose it’s possible that immunology was hot back then. Stranger things happened.

So how ’bout the small RNA field? No big scandals I can recall, (although I’m fairly new to the field). A search for the terms “miRNA” and “retraction” in Pubmed yields 3 results: one Science paper and a couple Nature papers. However, retracted papers and scientific fraud are not one and the same.

The controversies in the small RNA field seem to encompass smaller battles, say a partially disputed and highly visible paper with data that is not totally rebuked by others, but where the authors make overly great leaps in their interpretations. And as a colleague of mine pointed out, the field is so new and moves so quickly that there hasn’t been enough time for researchers to confirm all previously published results.

A paper claiming miRNAs can up-regulate translation under certain cellular conditions generated a lot of controversy. A recent presentation of this work by Joan Steitz at the Keystone meeting in April, more than a year after its initial publication, was still met with some underlying skepticism in the questions that were asked.

Bob Weinberg himself published a miRNA paper that generated some discussion in journals (1), and at conferences (2). But these last two examples are really cases of a healthy scientific discourse, openly discussing results and interpretations.

While in the literature I regularly encounter dubious data, it’s more often the dubiousness of some of the interpretations that bothers me. For novel published data I think it is beneficial to temper skepticism with belief. Considering alternative explanations for what one may consider questionable data is a much better intellectual exercise than discounting an entire paper, and just ragging on it.

But hey, what do I know. Listen to a guy like Bob Weinberg. He said he doesn’t even read the literature. He just talks to people.

(1) http://www.nature.com/nature/journal/v455/n7216/full/nature07362.html; http://www.nature.com/nature/journal/v455/n7216/full/nature07363.html

(2) Witnessed at Keystone RNAi, MicroRNA, and Non-Coding RNA 2008 meeting

MIT Faculty Open Access Policy FAQ

PURPOSE AND AIM OF THE POLICY

Why are we doing this?

In the eyes of many faculty, the goal of disseminating research is best served by using the unified action of the faculty to enable individual faculty to distribute their scholarly writings freely.

This view of the importance of open access to the faculty’s writings is especially apposite in the face of increasing efforts by some commercial publishers to further close access to the scholarly literature they control. Other organizations with a vested interest in scholarship are independently supporting such efforts as well. For instance, the Wellcome Trust requires any scholarly articles on research they fund to be made openly accessible. The National Institutes of Health, by congressional legislation, have recently instituted a similar requirement, mandating posting in the open-access PubMed Central repository.

Isn’t this unprecedented?

No. Harvard University’s Faculty of Arts and Sciences and Harvard Law School, as well as Stanford School of Education, have similar policies. As mentioned above, the Wellcome Trust mandates an open access requirement for their grantees. NIH also has a policy mandating open access.

What’s in it for me?

The Internet and web have enabled individual faculty to make their articles widely, openly, and freely available. Research has repeatedly shown that articles available freely online are more often cited and have greater impact than those not freely available, and this trend is increasing over time. Consequently, many faculty already make their writings available on their web pages, sometimes in potential violation of copyright law and sometimes through individual copyright negotiations with publishers. This policy will allow you to make your writings openly accessible, and it will enable MIT to help you do so.

COMPLYING WITH THE POLICY

What do I have to do to comply with this policy?

The policy operates automatically to give MIT a license in any scholarly articles faculty members complete after its adoption. This is true for both existing and later-hired faculty members. MIT will establish procedures for confirming this license and obtaining copies of articles to post in the repository, as well as for granting waivers of the policy when informed by an author of a decision to opt out.

In connection with the grant of this license, MIT faculty are recommended to communicate this policy to their publisher and add to any copyright license or assignment for scholarly articles an addendum stating that the grant is subject to this prior license. That way, you will avoid agreeing to give the publisher rights that are inconsistent with the prior license to MIT permitting open-access distribution. MIT provides a suitable form of addendum for this purpose. However, whether you use a suitable addendum or not, the license to MIT still will have force.

What if a journal publisher refuses to publish my article because of this prior license?

You have a number of options. One is to try to persuade the publisher that it should accept MIT’s non-exclusive license in order to be able to publish your article. Another is to seek a different publisher. A third is to consult with the Scholarly Publishing & Licensing Consultant or the Office of General Counsel about taking steps to address the publisher’s specific concerns. A fourth is to obtain a waiver for the article under the policy.

SCOPE OF THE POLICY

What kinds of writings does this apply to?

Only scholarly articles. Using terms from the Budapest Open Access Initiative, faculty’s scholarly articles are articles that describe the fruits of their research and that they give to the world for the sake of inquiry and knowledge without expectation of payment. Such articles are typically presented in peer-reviewed scholarly journals and conference proceedings.

Many of the written products of faculty effort are not encompassed under this notion of scholarly article: books, popular articles, commissioned articles, fiction and poetry, encyclopedia entries, ephemeral writings, lecture notes, lecture videos, or other copyrighted works. This is not to denigrate such writings. Rather, they are generated as part of separate publishing or distribution mechanisms that function in different ways and whose shortcomings, if any, the present motion does not and is not meant to address.

Does the policy apply to articles I’ve already written?

No, it doesn’t apply to any articles that were completed before the policy was adopted, nor to any articles for which you entered into an incompatible publishing agreement before the policy was adopted. Of course, the policy also does not apply to any articles you write after leaving MIT.

Does the policy apply to co-authored papers?

Yes. Each joint author of an article holds copyright in the article and, individually, has the authority to grant MIT a non-exclusive license. Joint authors are those who participate in the preparation of the article with the intention that their contributions be merged into inseparable or interdependent parts of the whole.

LEGAL ASPECTS

Is MIT taking the rights to my writing?

No. This policy grants specific nonexclusive permissions to MIT. You still retain ownership and complete control of the copyright in your writings, subject only to this prior permission. You can exercise your copyrights in any way you see fit, including transferring them to a publisher if you so desire. However, if you do so, MIT would still retain its license and the right to distribute the article from its repository. Also, if your article arises, in whole or in part, from NIH-funded research and was accepted for publication after April 7, 2008, you must retain sufficient rights to comply with NIH’s Public Access Policy.

What happens if I get into legal difficulty when attempting to comply with the new policy?

Staff in the Office of General Counsel and the Libraries will be available to support the new policy and to supply guidance to faculty.

OPTING OUT (OBTAINING A WAIVER)

How do I opt out?

To opt out, you simply send an email or other written notice to email address informing MIT of the following:

* Name of MIT author
* Title of article (expected or working title)
* Journal you expect to publish in
* Reason you are opting out

Why does the policy include an opt out? Doesn’t that undermine the policy?

The majority of faculty in our fall discussions stated a preference for an opt out. One of the concerns frequently raised was the importance of an opt out for junior faculty who do not want to jeopardize their ability to work with certain publishers. Another was the desire to comply with a certain society’s policies even if in conflict with this policy.

Even with opt out, the policy changes the default for author’s rights. The new given is that MIT has rights to openly share MIT faculty work and can extend rights to the authors for their use as well.

What happens if I do not opt out, but assign exclusive rights to a publisher anyway, mistakenly signing a publisher’s agreement that conflicts with the policy?

MIT’s license would still have force, because it would have been granted (through this policy) prior to the signing of the publisher contract.   If the publisher expresses concern that cannot be remedied, you could opt out for a given article.   You could consult with the Office of General Counsel or Libraries staff for a review of your options.

What happens when co-authors disagree on whether or not to opt out?

Each co-author in a jointly written article owns the copyright. Under US copyright law, any co-author has the right to grant a nonexclusive permission to others. It would be up to the co-author to decide whether to opt out of the policy for a given article to accommodate a co-author.

Could I retract a paper later if a publisher required me to do so?

It would be possible to remove a paper, particularly in cases involving a legal dispute.  The specifics would depend on what procedures are worked out by the Faculty Committee on the Library System to implement the policy.

WHAT WILL BE DONE WITH THE ARTICLES

What will MIT do with the articles it has license to?

MIT will continue to operate its open-access repository, DSpace, to make available the scholarly articles provided by its faculty members. This repository has MIT standing behind it to ensure its availability, longevity, and functionality, to the extent technologically feasible. The repository is backed up, mirrored, and made open to harvesting by search services such as OAIster and Google Scholar.  Adjustments will be made to the interface and deposit processes, under the guidance of the Faculty Committee on the Library System, to make it as convenient as possible.

The DSpace open-access repository is run by the MIT Libraries. The Faculty Committee on the Library System will provide advice on the implementation of this policy in the DSpace environment.

Through the transferability provision, MIT may further allow others to distribute the content, provided that the articles are not sold for profit. For instance, faculty at other institutions could be given permission to make copies for free distribution directly to their students. However, MIT does not have– and cannot grant to others –the right to sell the articles for a profit or to sell a book containing the articles for a profit.

Does this license preclude all activities that involve payment?

No, not necessarily. An activity will not cease to be permitted under the policy merely because a charge is imposed to cover some or all of the costs of the activity, provided that articles are not sold for a profit. Hence, for example, MIT’s selling course packs at cost would be permitted. The Faculty Committee on the Library System will provide advice on what licensed uses of repository material are appropriate and consistent with the purposes of the policy.

Can others distribute my work, for instance, placing it in a course pack?

This policy would grant MIT the right to license others to distribute the work, so long as the work was not sold for a profit.  For example, MIT could give permission for an article to be used in a coursepack, so long as the course pack was not sold for profit.  (As author, you could obtain permission from MIT for free use of your articles in coursepacks as well, if you transferred copyright to the publisher and therefore needed permission to use an article in this way.)  To take another example, MIT also could authorize others to make your articles available online (for example, in another repository), provided that they were not sold for a profit.

No one would be able to sell your articles for profit without getting permission from the appropriate rights holder, whether that were you or a publisher to whom you have assigned such rights.

Can my articles be used to provide search or other services by companies such as Google?

Yes, consistent with the goals of open access and ensuring wide visibility and availability of scholarly articles, the license allows MIT to enable both commercial and nonprofit entities to use the articles to provide search or other services, so long as the articles are not being sold for a profit. This is true even if the services generate advertising revenues or the company charges for the services. For instance, the license allows MIT to enable the articles to be harvested and indexed by search services, such as Google Scholar, so that they can more readily be found, and to be used to provide other value-added services that don’t involve selling the articles themselves for a profit. MIT also could authorize use of the articles in a commercial service that provides information extracted from the articles (but not the full text itself), such as bibliographic data or citation lists.

Will MIT be able to take advantage of future changes in technology to provide open access to the articles?

Yes, if new technological means of distributing or making the articles available evolve during the lengthy term of copyright, the license is intended to give MIT the flexibility to use those means to advance the purposes of the policy, provided always that the articles are not sold for a profit.

Who will monitor implementation of the policy?

The Faculty Committee on the Library System (which is a standing committee of the MIT faculty) has agreed to they would work with the MIT Libraries to develop an implementation plan that has faculty interests in mind.  They anticipate developing processes and procedures by soliciting input from the faculty and providing progress reports to the faculty as the implementation plan is developed. The Faculty Policy Committee would be responsible for presenting a report regarding the policy to the faculty in five years.

IMPACT OF POLICY ON SCHOLARLY PUBLISHING ENVIRONMENT

Will this policy harm journals, scholarly societies, small friendly publishers, or peer review?

There is no empirical evidence that even when all articles are freely available, journals are canceled. The major societies in physics have not seen any impact on their publishing programs despite the fact that for more than 10 years an open access repository (arXiv) containing nearly all of the physics literature written in that time has been available and successful.  If there is downward pressure on journal prices over time, publishers with the most inflated prices – which tend to be the commercial publishers – will feel the effects sooner.   Journals will still be needed for their value-added services, such as peer review logistics, copy editing, type setting, and maintaining web sites.  In a capitalist system like ours, if there’s value added, someone will find a way to pay for that value, particularly as there is already a great deal of money tied up in the existing scholarly publishing system that could potentially be repurposed.

Will this policy harm those in tenure processes who need to show publication in high quality journals?

The opt out option would protect authors who need to publish in journals that will not cooperate with the policy.

How will this policy affect other universities, particularly small ones?

We expect that as similar policies are passed at more universities, the overall climate for scholarly communication will improve, to the benefit of all institutions of higher education.  Smaller universities may not have the resources to build their own repositories, but shared repositories are starting to become available for such cases.

What’s in it for MIT?

The policy would increase the impact of MIT research by making it more widely available.  Studies show a very large citation advantage for open access articles, ranging from 45% to over 500%, but restrictive publisher business models limit wide sharing through onerous terms in contracts with university libraries and individual authors.  For example, many publishers prohibit authors from posting their work openly on the web, and publishers have attempted to exclude access at MIT to various groups or locations; they commonly ‘rent’ access to their content, putting access at risk following cancellation of subscriptions.  Performing systematic searching, advanced indexing, or analysis are prohibited in virtually all contracts.

The policy would give MIT a means of negotiating for more attractive terms with publishers, an effort needed in a context of dramatic inflation and market consolidation: the 5 largest journal publishers now account for over half of total market revenues, and over the past 15 years, the price of scholarly journals has grown roughly three times as fast as the Consumer Price Index.

MORE INFORMATION

What is Open Access?

Open access as discussed in relation to this policy refers to free availability of journal articles on the public internet, permitting any users to read, download, copy, distribute, print, search, or link to the full texts of these articles, crawl them for indexing, pass them as data to software, or use them for any other lawful, noncommercial purpose, without financial, legal, or technical barriers other than those inseparable from gaining access to the internet itself.

How is granting a nonexclusive license to MIT compatible with MIT being able to exercise ‘all rights under copyright’? Don’t those rights include the right to grant exclusive rights?

The legal framework for copyright is that you can’t give away what you don’t have. MIT will have been granted nonexclusive rights, and will not be able in turn to grant exclusive rights.  MIT, however, will be able to exercise all of the other rights under copyright, including reproducing, displaying, distributing, and making derivative works of articles covered by the policy, as long as these activities are not done for profit.

Why make this an automatic license? Why not just suggest that faculty individually retain a license for open-access distribution?

First, experience has shown that mere exhortations have little effect on authors’ behavior. For instance, before Congress made it a requirement, participation in the NIH Public Access Policy was optional. During that period, there was only a 4% level of compliance. Second, experience in many areas has shown that opt-out systems achieve much higher degrees of participation than opt-in systems, even while remaining noncoercive. Third, by making a blanket policy, individual faculty benefit from their membership in the policy-making group. MIT can work with publishers on behalf of the faculty to simplify procedures and broaden access. Without a blanket policy, the unified action benefit of the policy would be vitiated.

Why isn’t student work, or that of research staff, covered by the policy?

The policy applies only to faculty because as a first step, it seemed clearest to focus on faculty work. MIT already receives a license to Ph.D. theses; in addition, many student articles will be co-authored by faculty and will be subject to this policy.

Why aren’t images covered by the policy?

Images are created by faculty in such a wide range of contexts and for such a wide range of purposes that it was too complex to include images in the policy.  To the extent that images are contained in the articles, however, they would be covered by the policy.

Why isn’t data covered by the policy?

Data is not generally covered by copyright, apart from its particular expression in an article or supplement to an article, which would already be covered by this policy.  It therefore seemed clearest not to include “data” in the policy apart from the article.

Why aren’t Ph.D. theses included in the policy?

MIT has already made a commitment to making Ph.D. theses openly available worldwide.  MIT’s DSpace makes more than 20,000 selected theses and dissertations from all MIT departments, dating as far back as the mid 1800′s, openly available to the world at http://dspace.mit.edu/handle/1721.1/7582. Since 2004, all new Masters and Ph.D. theses are being scanned and added to this collection after degrees are awarded.  Regardless of whether copyright is held by the student or the Institute, the MIT Libraries publish the thesis electronically, allowing open access viewing. While there are fees charged for scanning and printing theses, these fees are used to scan more theses and make them available for open viewing.   Without these fees, only a very small subset of theses (those submitted electronically) would be available online.  While the current system is not open access as defined by the Budapest Open Access Initiative, which would include no limits on printing, it’s a significant step in that direction.

Why should MIT or any university take the lead here? Why not the scholarly societies or an organization like the American Association of University Professors?

Given the wide range of publishing operations and viewpoints, and discipline-specific missions, concerted action from scholarly societies is not likely to be the nexus of needed change.  Other organizations that support university faculty have very specific missions that don’t necessarily position them to take an early lead in this area.

Wouldn’t collaborative action among universities be more effective? Couldn’t some commitment to this be included in the policy?

This policy is the first step; once a group of universities has similar policies, there may be opportunities to work collaboratively.

Who will pay for this?

MIT already has the technical infrastructure in place to store the articles, in the form of the open access repository DSpace@MIT.  In addition, the MIT Libraries have experience supporting access to faculty research such as technical reports and working papers, and for the past several years have maintained an Office of Scholarly Publishing and Licensing to assist faculty who wish to retain rights in their published works.   Once an implementation plan is developed, it will be possible to assess what other staff or technical support might be needed, if any, and to reassess priorities in light of those needs.

What are publishers saying in response to the Harvard policy?

It is too soon to have a broad and clear sense of publisher responses.  Some publishers are asking Harvard authors to opt out, but the policy has nevertheless changed the default for openness to Harvard research.  The policy doesn’t solve all problems immediately, but is expected to be a first step in a long process of change intended to redress imbalances in the scholarly publishing system.

How is this policy related to the NIH Public Access Policy, and how is that policy working?

The NIH Public Access Policy applies only to NIH funded research – about 1/3 of MIT’s funded research dollars.  It requires authors to deposit their peer-reviewed articles in the open access repository PubMedCentral where they must be accessible within 12 months of publication.  Making the policy mandatory has had a dramatic effect on deposits: the rate has increased from under 10% to an estimated 60%.  The policy makes tax payer funded research available to taxpayers.

A particular article could be subject to both this policy and the NIH Public Access Policy, if it is peer reviewed and arose, in whole or in part, from NIH-funded research and is accepted for publication on or after April 7, 2008.

If an NIH-funded article is covered by this open access policy, the author would use the MIT amendment to publication agreements to cover NIH’s obligations and accommodate the MIT policy.  Even if the author decides to opt out of the policy for an article, the author must reserve rights sufficient to comply with the NIH policy when entering into a publication agreement for the article.

For more information on the NIH addendum and compliance with that policy, please see the MIT web page about complying with the policy.

Why doesn’t the policy express support for open access journals?

This policy takes only a first step towards re-balancing the scholarly publishing system, giving MIT a means of negotiating for faculty and allowing wider sharing of their research.  Other steps will no doubt make sense in the future.  Some universities, for example, have begun supporting open access journals by creating funds authors can use for publication fees.

This seems like a small step. Shouldn’t we be doing more?

It’s an essential first step, allowing the faculty to be able to speak with one voice, and the Institute to negotiate on its behalf. Other steps may follow.

Why doesn’t MIT just go into the business of publishing open-access journals?

MIT already has a publishing enterprise, the MIT Press, which is experimenting with open access journals and business models. The Press’s experience and knowledge will be an asset in supporting the new policy.

Why doesn’t the policy offer a delay before posting the articles, so that the MIT version doesn’t show until after journal publication?

Different disciplines have very different “half lives” for journal articles, making it very difficult to include a particular time period in an overall policy.  Instead, the opt out exists for authors whose publishers require a delay before posting.

Won’t this lead to the proliferation of versions and confusion over citation?

With or without this policy, the academic community will need to work on the problem of version control in digital scholarship.  There are technical and standard-based solutions that will address this problem.  The Association of Learned and Professional Society Publishers (pdf) has been examining this issue, along with the International working group of scholars, scholarly societies, and publishers (pdf) and the AAAS, among others. Nomenclature and modeling efforts have been begun by the National Information Standards Organization (pdf) and the Version Identification Framework.