As usual, the last couple of months have seen lots of talk (on lists, weblogs, e-sources and in print) and maybe some action, although the action’s not final as I write this.
Sometimes it’s hard to take statements at face value. Consider “How To Access Medical Information,” a two-page August 2004 statement from the Professional/Scholarly Publishing Division of AAP, the Association of American Publishers. This statement informs us that “publishers and their library partners have invested hundreds of millions of dollars in the past decade to improve access to the biomedical journal literature.” “Library partners”—what a wonderful turn of phrase! The publishers create electronic services to “deliver this information directly to the desktop of physicians, researchers, and other health professionals” and “other communities get access” because publishers kindly make those services available through libraries, “either under license or via free access.” (It’s that “under license” that accounts for much of that “hundreds of millions of dollars,” of course.)
Eight bullet points follow to show how committed publishers are to making “medical research results widely and readily available.” Publishers “actively participate in literature retrieval systems”—and you have to read that one carefully indeed. It highlights PubMed, “a free web-based service with data about the biomedical journal literature” [emphasis added] and “web links that enable both medical research professionals and the general public to locate the full text of the articles, which are made available from the publishers’ own web sites.” Note that “made available” may and typically does mean “at the price publishers choose to charge.”
“Publishers enable electronic access to their journals via
flexible licensing arrangements…” and “most licenses let libraries give free
access to any member of the public who is permitted to use the library on a
walk-in basis. In the
“Publishers endorse the practice of interlibrary loan”—but
the bullet point is silent as to whether electronic licenses allow for
The cooperative and aggressive actions of publishers to improve access to the medical literature means that in contrast to the situation a decade ago, where access was limited to the hundreds or thousands of paper copies in circulation, tens of millions of researchers and physicians now have desktop access—and the latest advances in medical research are made available rapidly to the interested public through their libraries or the publishers themselves.
A wonderful statement—but its timing, shortly after the NIH proposal to mandate OAI archiving for all that medical literature funded by NIH grants, strikes me as a little too convenient. If you read the statement quickly, you’d think all that medical information was readily available to everyone for nothing or almost nothing. You’d be wrong. All in all, this strikes me as a cleverly worded attempt to establish that all’s right with the world, and those government bureaucrats and OA meddlers are just trying to solve a problem that doesn’t exist. Maybe I’m paranoid, and this is actually nothing more than a sincere attempt by AAP/PSP to publicize their services. If so, I apologize in advance.
For the rest of this essay, I’m going to do something I should have done back in June: Provide a numbered key to the standard arguments against OA publishing (as opposed to unique arguments such as “it distracts attention and money from OAI archiving”), so I can simply list the numbers used in specific pieces. For this issue at least, here’s a subset of those arguments:
Ø 1. STM publishing has developed over centuries and works just great as it is.
Ø 2. $1,500 (or $500 or $525) can’t possibly pay the real costs per article; OA isn’t sustainable without charging ($3,000, $4,500, whatever).
Ø 3. OA publishing weakens or undermines peer review.
Ø 4. Research grants don’t include publication funding.
Ø 5. OA/article-fee publishing gives well-funded scientists advantages over others.
Ø 6. OA/article-fee publishing will prevent scientists in developing nations from publishing.
Ø 7. OA publishing undermines professional societies that subsidize their activities through journal profits.
I’m qualifying 5 and 6 because not all OA publishing involves article fees; quite a bit is sponsored in some other manner.
Martin Frank, Executive Director of APS (American Physiological Society in this case) published “Open Does Not Mean Free!” in The Physiologist 47:4. He offers arguments 1, 2, 4, and 5. Additionally, he suggests that it’s unrealistic to expect NIH to come up with the “full cost of publication at a time of budgetary restraints.” Interestingly, Frank cites the scientific journal arena as “over 5,000 scientific journals,” one of the lowest numbers I’ve seen. I tend to agree with Frank’s final statement, but I’m not sure that it has much to do with NIH’s proposal to require the equivalent of OAI archiving for government-funded research results, which if done in government laboratories would automatically be in the public domain:
We believe that a free society allows for the co-existence of many publishing models, including an author pays model, and therefore believe that it would be foolish and dangerous to do away with one model for another that remains largely unproven.
Now if I could only find the dragon that Frank’s trying to slay—the powerful advocate who calls for immediately shutting down all traditional journals.
Fred Spilhous, another professional society Executive Director (American Geophysical Union this time), sent a letter to The Economist in August objecting to their article on Open Access (which Spilhous puts in scare quotes). He calls it a “utopian vision” containing “fatal flaws.” He uses argument 3 and adds suggestions of government interference with publishing and some other odd questions. He calls the results of OA “scavenging in a huge garbage heap.” Peter Suber’s quick commentary includes the note that arguing government interference at the point of publication is odd—since most funded research is already funded by governments. Suber also notes that “upfront funding” (i.e., article-fee funding) is not the only funding model for OA journals.
Remember Springer’s disingenuous “Open Choice” initiative, where it offers free access if you pay a mere $3,000 per article? Derk Haank “blasted critics” of the initiative and, of course, emphasized Argument 2. His response to the objection that Springer still insists on taking copyright? Don’t laugh: “Copyright is not that important to us, but we are using it here as a mechanism to protect the author from having articles taken by other commercial publishers.” Right.
Speaking of Martin Frank (a few paragraphs back), he and two other APS officials wrote “A not-for-profit publisher’s perspective on open access,” which will appear in Serials Review 30:4. It’s an invited paper, available as a 16-page manuscript. The article describes “A decade of progress” in “how far STM publishing has come in terms of providing electronic access to information” (a variant of #1), including APS’ own experience; includes a section on “Government-run scientific publishing” that somehow manages to include PLoS; and continues with a bunch of reasons that OA is a bad thing. I would say that the article is valuable as a history of APS and non-profit experience—but in fact, only about four paragraphs (less than one page) are about APS. This is mostly another anti-OA screed. It’s a different one, though: I only recognize #1 and #2 from the standard list, although #2 is driven into the ground. Other arguments include flat assertions that PLoS and BioMed Central institutional memberships are paid for by libraries (certainly not true of national memberships); that somehow allowing a tradeoff between prepaid membership fees and per-article processing charges is directly comparable to (or at least no less objectionable than) “using subscriptions as ransom for access”; and a direct attack on the NIH/centralized repository approach based on the idea that modern searching means it doesn’t matter where documents are deposited.
Some elements of the article are simply strange, such as the early statement that “ten years ago…the era of online publication had not yet begun,” which for a 2004 paper is truly ahistorical. (I just looked up the Public-Access Computer Systems Review special issue on e-journals, which included essays relating to at least six of the e-journals already in existence: It appeared in early 1991. E-journals go back at least to 1987.) Somehow, the fact that STM content is far more accessible now than it was in the past (true) is offered as the answer to those who say that government-funded science should be fully accessible. I see some confusion, I think deliberate, between OA publishing and OAI archiving. The escalation of claims for the true cost of online publishing is escalated once more, with a claim that the cost per article of Journal of Clinical Investigation is around $6,000—that’s expensive processing! And the numbers involved with APS’ experiment in “author-pays” publishing seem a little odd. Physiological Genomics will make papers immediately available for a $1,500 fee; otherwise there’s a one-year embargo, Only 10% of authors have paid the fee. But the institutional online subscription price for Physiological Genomics is $205. That price raises the question: What are the true costs of article processing for that reasonably priced online journal? $1,500 seems high—but I don’t have access to the full set of numbers.
At the same time, much of the article is reasonable, at least to my mind. If Michael Eisen of PLoS really did call it “morally superior” to Nature, Science, and others, you can count me out of that particular crusade. The authors say “Not-for-profit journals are not generally seen as the source of the cost increase problem,” and I believe that’s true—noting that some journals issued under the aegis of professional societies are most definitely profitable, whatever their tax status.
The Creative Librarian commented on this article in a
Another in our parade of Society Executive Directors Against Open Access Publishing, (SEDAOP?), John H. Ewing offered his “point of view” in the October 1, 2004 Chronicle of Higher Education: “Open access to journals won’t lower prices.” He does admit that journals publishing is in crisis, then asserts that OA arguments represent “misdirection” of the sort magicians use. Further, he says it’s a mistake, based on “information must be free” ideology.
I see versions of arguments 1, 2, 3, 5 and 7. He deals with
access issues, in part, as follows: “Of course, e-mail makes it possible for
another scholar to ask an author for a copy of an article and receive it the
same day.” Problem solved—as long as you’re part of the inner circle and
therefore (a) know of the article, (b) know or can find out the author’s email
address, and (c) are yourself of a stature such that you can assume the author
will respond to your email instead of deleting it unread.
Part of me wants to buy into
Scholars and librarians have to stop dealing with high-priced journals, as authors, editors, referees, or subscribers. Soon the publishers of less-expensive journals will grow, and those of more-expensive journals will decline. The less-expensive journals will publish more papers, making them more efficient, and society publishers will earn slightly more profit, which they can reinvest in their disciplines.
If only it were that simple. If every ARL library simply stopped all of its subscriptions to journals published by Elsevier, Springer, and others of their ilk, that would certainly solve the STM-related budget problems of those libraries. Let’s not mention the problems that would be caused by that solution, particularly for scholars at those institutions that have substituted access for ownership and don’t have back print runs of the journals involved. Would the libraries survive the campus political firestorm to enjoy their improved budget status?
Finally (for this section), here’s “Electronic cultures and clinics: Reasons to be hysterical (and hopeful),” the 2004 Elsevier Library Connect medical library lecture, given May 25, 2004 at the Medical Library Association Annual Meeting by Dr. Richard Horton, editor-in-chief of The Lancet. It’s a transcript of what must have been an engaging talk. Indeed, I found the first nine pages (of 15 total) fascinating, and was taken aback to read the claim that one of the key OA declarations (“Berlin II”) apparently calls for the replacement of conventional scholarly communications, which is overreaching. Unfortunately, after that, we get arguments 1, 3, 7, 2, and 5 (in that order), with an astonishing $10,000 per paper offered as a realistic number. Additionally, Horton makes a statement that I will assert is untrue and am certain is unprovable. He quotes a statement from a Wellcome Trust report, “Open Access means that for learned societies they have quote, nothing to fear.” To which he says: “Not one person who works in a learned society believes that.” Not one? There is not a single learned society in the world with one employee who believes OA can’t harm the society? There are no learned societies that have adopted OA and can’t be harmed by its progress? Even the mighty Elsevier editorial offices don’t have that kind of competitive intelligence.
The Empire Strikes Back in Cites & Insights 4:7 discussed seven of the first 25 (or so) essays in an ongoing Nature “Web focus: Access to the literature.” That discussion has now apparently concluded, given the unsigned 35th essay that’s unsigned and seems to comment on the forum as a whole. You can get to the whole set of essays at www.nature.com/nature/focus/accessdebate/. A few comments on the last six essays, in numeric order:
“What do societies do with their publishing surpluses?” That
question introduces the results of a survey of society publishers—admittedly
These three, from the Centre for Information Behaviour and the Evaluation of Research (you can guess it’s British!), offered “Journal publishing: what do authors want?” It’s based on a huge survey—91,500 authors who had published in ISI-indexed journals over the past 18 months, with 3,787 fully completed responses. While the responses are interesting, they mostly support the sense that most scholars still don’t pay much attention to library budget problems or, in fact, the outrageous prices charged for the journals in which they publish. Those are someone else’s problems—and maybe that’s how scholars should react.
Most scholars want to “narrowcast”—they want to reach researchers in their own fields. Most (but only 74%) want to reach researchers in other fields, and a slim majority (56%) want to reach education professionals. Only 40% care about reaching policy makers and 18% care about reaching the general public. Scholars want “the imprimature of quality and integrity that a peer-reviewed, high-impact title can offer, together with reasonable levels of publisher service.” What other results would we expect?
Some other numbers are double-edged swords. The essay says authors are “generally happy with their access to the journals literature”—but only 61% say they can “currently get hold of most or all of the titles they need.” That leaves 39% who are shy of access. Sure, there’s more access than five years ago (although 11% say it’s worse). Then there are the “author-pays” possibilities, limited to the 18% of authors who knew something about OA and worded rather nicely: “If all journals were Open Access, what do you consider would be a reasonable payment to have your paper published in the best journal in your field?”
49% of authors still said “nothing,” with another 46% offering less than $1,000; only 6% of the largest group of respondents (medicine, allied health and veterinary science) would be willing to pay more than $1,000, and no field showed more than 19% (earth and planetary sciences, but that’s really 13 respondents!). Overall, only 16% were willing to pay more than $500. The authors add commentary suggesting that most scholarly authors don’t really think that publishers add much value—which may help explain their disinclination to see author payments.
The proponents of OA are not just offering one more good idea; they are promoting the one true faith, and they demand that we all become converts.
He quotes the Budapest Initiative, PLoS, and Harold Varmus. I
would say he reads quite a bit into their statements, but—more importantly—there’s
a lot more to OA than PLoS/Varmus and
Ross entitles his essay “Electronic publishing models and the public good” and comes to an unusual early conclusion: He believes that authors are on the side of OA and don’t care about publishers, while librarians find themselves more closely allied with publishers. I believe the survey cited above (Rowlands et al) suggests fairly forcefully that most authors know nothing of OA and that, of those who do, most won’t support the economic model. On the other hand, I agree with some of his reasons librarians might understand the concerns of publishers: librarians know that publishing isn’t free, appreciate publications, believe that publishers add value, and understand that electronic publishing can be complicated. I suppose librarians “have shared similar anxieties about being disintermediated along with publishers”—but some publishers seem as willing to dismiss the contributions of libraries and librarians as are some within the OA camp. (Whenever someone calls the 70% of academic library budgets that goes for salaries and the like “overhead” I want to scream, but who would hear me?)
On the whole, I think this is a good essay, worth reading and thinking about. He uses the higher range of cost estimates for articles ($1,500 to $5,000), but contrasts that with estimates of the actual research cost per article published: $50,000, $150,000, or $250,000 to $300,000! Given the amount of least-publishable-unit publishing that happens, those are truly astonishing numbers. I wonder about this comment: “As Open Access costs shift away from the user to the producer, scientists find themselves becoming publishers.” I don’t understand: By that logic, libraries are currently the publishers, and I don’t believe that to be true.
A declaration of interest at the end raises a touch of argument #7, but only a touch.
At its start, this essay—“The pros and cons of Open Access”—appears fairly even-handed, but as it continues I note that arguments against OA are seldom refuted, while statements for OA seem to carry direct counters. Arguments 2 (at great detail, but in a form that biases the discussion hugely toward traditional publishing) and 7 predominate, and Worlock throws in the association of OA with Stewart Brand’s silly “information wants to be free.”
Nature is a traditional publisher. Why would we expect that the publisher’s summary of this discussion would be even-handed? Arguments #2 and #7, and an indirect but strong whiff of #1, show up along with direct attacks on (Nature’s interpretation of) the NIH proposal and a pretty good indirect roundhouse on the UK study. Most of this essay is a checklist of “how publishing adds value.” It’s a good list, worth reading as a reminder of why effective refereed publishing and dissemination will have non-trivial costs, no matter what the publication model.
Action and discussion on the NIH open-access plan continue. A few points along the way:
August 24, 2004, a press release announced the formation of the Alliance for
Taxpayer Access, “an unprecedented coalition of public interest groups” that
will urge NIH and Congress “to ensure that peer-reviewed articles on
taxpayer-funded research at NIH become fully accessible and available on line
and at no extra cost to the American public.” The new group does represent a
broad range of groups, including
Ø Two days later, a group of 25 Nobel laureates sent an open letter to Congress “to express our strong support for the House Appropriations Committee’s recent direction to NIH to develop an open, taxpayer access policy requiring that a complete electronic text of any manuscript reporting work supported by NIH grants or contracts be supplied to the National Library of Medicine’s PubMed Central.” That lengthy sentence does state the precise support in full; it’s followed by several paragraphs about the importance of science and the need for consumers to have access to current research. It cites the same $30 article fee as the ATA statement, but this time it’s “or more” rather than “as much as.” The letter also explains why PubMed Central access “will not mean the end of medical and scientific journals at all” and notes that mandated open access would only apply to NIH-funded research. The laureates include 18 winners in Physiology or Medicine and 7 in Chemistry.
Ø In early September, Peter Suber offered a first take on the September 3 plan from NIH and how it differs from the July 14 House Appropriations report language. The September plan drops the requirement for immediate access if NIH paid any part of the article’s publication cost, substituting OA within six months or sooner. It details what gets deposited at PMC and what NIH funding triggers the OA plan—notably including articles whose underlying research “was supported in whole or in part by NIH funding,” a potentially tricky requirement. The September plan offers a range of specific goals from NIH, including improving the health of Americans, sharing and supporting public access to results of NIH-funded research, and balancing the need for access with the ability of publishers to preserve peer review, editing and quality control.
Ø A September 13 news report from Library Journal notes that NIH director Zerhouni recently met twice with stakeholders, but that some lawmakers were backing off their call for immediate access—and that Senator Arlen Spector said that he would not add a call for public access to the Senate version of the appropriations bill.
Baum, editor-in-chief of Chemical & Engineering News, attacked the NIH
plan forthrightly in a
September 21 editorial at Data Conversion Laboratory has a nice way of telling
part of the story, as in its lead sentence: “Government committees in the
Ø September 23 brings a letter from BioMed Central’s Jan Velterop to NIH’s Elias Zerhouni. Velterop notes that roughly 15% of BMC’s articles indicate some form of NIH funding—and that all BMC articles are deposited immediately at PubMedCentral. He argues BMC as a counter-argument to “the reservations expressed by traditional publishers as to the economic sustainability of an open access publishing model.” He also endorses the six-month delay as “a sufficient and appropriate help” for traditional publishers to adjust to a new model.
Ø Barbara Quint cheers on the NIH in her “Up Front” column in the October 2004 Information Today, calling the plan “the day of liberation” and “only the first of many.” The column is typical Quint, with strong opinions and strong language to state them.
Ø The October 2, 2004 SPARC Open Access Newsletter (issue #78) leads with “A busy month of action on the NIH open-access plan.” You’ll find loads of links to various documents and statements in the essay. Interesting points: the U.S. Chamber of Commerce supports the NIH plan, as do the American Association of Universities and National Academy of Science—but the CC endorsement is the most startling. The New England Journal of Medicine has an endorsing editorial—but still calls for journals to hold copyright “in order to block the redistribution of mangled copies of the text” (a rationale for copyright transfer that I’ve never quite understood). John Regazzi of Elsevier gave a typical Elsevier “yes, but” response: “No one can argue against giving the public access to NIH information…but…the NIH proposal is moving too fast.” Since Elsevier now allows OAI archiving, which differs from the NIH plan primarily in using distributed rather than centralized archives, things are already complicated. Suber notes that the current NIH plan “requests” rather than “requires” article deposit at PMC—but that there’s reason to believe failure to do so would endanger future grants. There’s also a preliminary estimate for the cost of the larger PMC digital library: $2.5 million (per year, I assume), not the $100 million suggested by some critics.
The deadline for a full plan is December 1. I see nothing in the NIH plan that calls for OA publishing, which makes the tenor of some criticisms a bit odd. What’s currently planned is a publisher-friendly modified version of OA archiving, differing from OAI archives in two key respects: There’s a six-month “toll access” wall, and papers are either deposited in a central repository or appear in publisher archives with pointers from that repository.
That same October 2 SOAN includes fascinating notes from 1974 about the dangers that photocopying poses for STM journals—remember the Williams & Wilkins suit? He also offers “a haiku introduction to open access,” a “mercifully small sampling” of 15 haiku. I’ll quote the first, second, and last:
If you publish it,/and readers can’t afford it,/does it make a sound?
They don’t pay authors,/editors or referees./Then they want the rights.
The current system/evolved over centuries./So did dinosaurs.
It appears that LOCKSS is making progress; Project Muse is involved, half a dozen OA journals are cooperating, and both HighWire Press and Berkeley Electronic Press are experimenting. For more information, see lockss.stanford.edu/projectstatus.htm
Carol Tenopir offered “Open access alternatives” in the
PLoS Medicine “goes live” October 19. “There is no doubt in our minds that open access is the future of medical publishing,” says the press release [emphasis added]; multiple models have little place in the PLoS worldview.
SPARC Open Access Newsletter 77 (
Peter Suber’s lead essay, “Praising progress, preserving
precision,” wants to maintain strict definitions for OA while welcoming
initiatives that widen access without meeting those definitions. “The
best-known part of the BBB [
So is Cites & Insights OA? Apparently not (even if it was scholarly)—because I don’t automatically agree to let users republish this material in priced publications. It appears that, in Creative Commons terms, true OA only allows the “By” license, not the “Noncommercial” license. But Suber goes on to say that BBB does not require removing barriers to commercial re-use, even though I can’t see anything in either of the statements that would allow such a barrier. More to the point, this essay is concerned with the “false sharpening” of the OA definition. He doesn’t think that derivative works and commercial re-use should be required parts of OA, even though he personally prefers both. (I’ve reread the cited definitions four times now, and I still can’t see how an OA publication can pass the definitions and prevent commercial re-use…and that’s a shame.)
Suber goes on to praise initiatives that, by his standards, aren’t really OA—but do improve access. That is, I think, appropriate. I’m not surprised that most abusers of OA definitions are commercial publishers, including the truly bizarre case of Thomson Derwent offering a fixed fee for use and calling that “open access licensing.” Maybe we need a clearinghouse for “enhanced access” initiatives.
Suber wants to educate newcomers and maintain clean
definitions. I think that’s great. I also think the BBB statements are
difficult to read cleanly, at least based on my inability to read them the way
Suber reads them. Clarity would be useful, and that clarity might be achievable
by referring to a Creative Commons license. BioMed Central makes it easy (if
perhaps tightening OA too much): You have to agree to a “By” license to publish
in their journals. If there was common agreement that “By-NC” was the minimum
standard—thus allowing copying, noncommercial redistribution, and all the
rest—I think there might be more clarity.
“The effect of open access and downloads (‘hits’) on citation impact: a bibliography of studies,” OpCit project. opcit.eprints.org/ oacitation-biblio.html
This ongoing chronological bibliography may be worth
bookmarking and checking every few months. I downloaded the
Antelman, Kristin, “Do open-access articles have a greater research impact?” College & Research Libraries 65:5 (September 2004): 372-82.
C&RL isn’t (yet) open access, but I believe Antelman posted the PDFs of this article to an accessible repository as soon as that lack was pointed out. The short answer is Yes—“across a variety of disciplines, open-access articles have a greater research impact than articles that are not freely available.” For the longer answer, read the well-prepared, well-written article.
Davis, Phil, Terry Ehling, Oliver Habicht, Sarah How, John M.
Saylor, and Kizer Walker, “Report of the CUL Task Force on Open Access
Cornell produces a lot of scholarly articles: More than 3,600 a year, according to the dynamite appendix to this study. Cornell University Library spends a lot on scholarly journals: $4 million, or half of Cornell’s entire serials/database expenses. Of that $4 million, 43% goes to Elsevier—and 16% of Cornell-authored articles appear in Elsevier journals. Remove Elsevier, Kluwer, Wiley, and Springer, and Cornell’s journal expenditures go down to $1 million—although 70% of Cornell-authored articles appear in all the rest of the journals.
The CUL task force sought to bring some facts and clarity to the table, meeting weekly for the first half of 2004 to discuss issues, coordinating research, and compile the report. While the resulting report is only one data point, it’s an unusually thoughtful and detailed one—and it’s one from an institution that could logically stand to pay more in a universal OA publishing system, at least if the true cost per article turned out to be $1,500 (or any sum above $1,100). (Here’s a direct statement: “It is unlikely that CUL will save money under any producer-payment scenario.”)
The report looks at a range of possible scenarios, takes a clear-eyed look at costs and benefits, includes an extensive bibliography, and is well worth reading. Strong OA advocates will not be happy with the results—but maybe they should pay attention, since this is as carefully considered a case as I’ve seen.
Specific recommendations include fostering and supporting OA initiatives “that respond to or resonate with real needs of specific scholarly communities,” applying carefully-stated and sensible selection criteria in considering OA projects, and continuing an environmental scan on the state of OA—and raising awareness among scholars. Here’s the paragraph that precedes the specific recommendations:
While the traditional subscription model has certainly been abused by some publishing interests, our Task Force is convinced that subscription can still serve as an equitable model for disseminating scholarship under some circumstances, particularly when administered by scholarly societies, university presses, and academic libraries. We have concluded that the Open Access and subscription models can coexist and are in fact likely to do so for the foreseeable future. The pragmatic approach our Task Force is recommending for CUL should be understood as a continuation of the course the Library has taken up to now vis-à-vis Open Access publishing: a flexible, experimental approach that commits to support specific, viable applications tailored to particular needs, pursued as a key component of a diversified strategy of scholarly communications reform.
Donations to support Cites & Insights are accepted at http://citesandinsights.info.
All original material in this work is licensed under the
Creative Commons Attribution-NonCommercial License. To view a copy of this
license, visit http://creativecommons.org/licenses/by-nc/1.0 or send a letter
to Creative Commons,