The Scholar’s Space

Scholars are supposed to consult “the Literature” and contribute to it as part of their professional duties. Library types are supposed to preserve it, and make sure its treasures are available for future generations. In that subsection of  scholarly publishing that relates to medicine, however, there is growing reason to suspect that much of ”the Literature” is simply drug company flackery; very well conceived, slickly executed flackery, it is true, but flackery nonetheless.  And concern about preserving it is, to put it mildly, misplaced.  ”Ghost” authoring of journal articles has been known for a while, but studying it has been, for obvious  reasons, rather difficult.  “Opinion Leaders” are shy about admitting they allowed their names to be attached to articles they not only  didn’t write, but which were prepared by a Pharma company or its contractual agent. An article in PloS Medicine raises the ante a good bit. Why “ghost” just the article? Why not “ghost” the whole business…trial, data gathering, write-up, the works. A little judicious steering here and there, to make sure the right things get said and the wrong things left out, can do wonders for a product launch.  Read the article by Dr. Sismondo, and pay special attention to the section discussing the Sertraline trials. On the Scholar’s Space we spend a lot of time fretting about technology, assuming, operationally at least, that everything else in the hallowed research/publishing cycle is fine. What if we have it all backwards? What if the technology is the easiest, most tractable part, and it’s the rest of the process that needs worry and work?
Boo!

PS. In the context of all the hooha raised by the PRISM crowd about Peer Review and OA, notice in the PLoS article how neatly PR has been co-opted into the “marketing process”.   It’s no threat to the ghosters, and beating it is not only easy, but necessary for a successful “placement”.

There was a debate/discussion recently at Harvard, about the future of academic publishing. In this corner, in purple trunks, Dr. Howard Varmus,  Editor in Chief of the Public Library of Science (PLoS), and in that corner, his worthy opponent, Dr. Emilie Marcus, editor of the biggie bio journal Cell,  I wasn’t there, but I don’t think either party convinced the other. I’m relying on a report from the PLoS blog for some basic information in this post, and  PLoS is an OA publisher and Varmus a hot critic of established publishing practices in scientific journals. There, that should satisfy all the disclosure gods. You can  use the link to get the blogger’s take on what happened at the session, and draw your own conclusions. Some of the embedded links are interesting. The other part of the program was devoted to developments that may bring some of the theoretical advantages of network publication closer to  implementation. Discussants related experience with post-publication comments in Nature Precedings and PLoS One and so far, that traffic is quite modest.  Other capabilities, such as video demos of how experiments were done, may turn out to be more useful. Nobody knows which way the frog will jump. Maybe he’ll just sit there a while and think about it.
Harvard

Peter Suber, Open Access News, reports that an article on a popular subject (dietary supplements) was downloaded more than 16 thousand times in less than a month, and contrasts that with an article on an equally popular subject (acupuncture and pain relief), published in JAMA (Journal of the American Medical Association). These are only anecdotes, of course, but they help to highlight the potential of open access to reach its wider audience.

Stories like these also remind me of one of reasons cited by scientists (ScienceBlogs) for wanting to participate more fully in the online environment: there’s a lot of misinformation out there and instead of simply complaining about it and disparaging the value of instant, cheap, world-wide distribution, adding accurate, peer-reviewed scientific information into the mix is a much better option.

Peter Suber announced the publication of ARL’s bimonthly report, this double issue devoted to university publishing:ARL Bimonthly Report on university publishing. I have reviewed several of the articles already and heartily recommend taking a look at the publication. I especially found David Shulenburger’s article exploring the state of awareness of university administrators regarding their institutional policies to promote publication of research. It is quite an eye-opener. Especially the part about how neither IRs nor University Presses are even on the periphery of their vision. We have a lot of work to do, folks.

It seems that Google has achieved something close to world dominance, and that its various operations, plans and products are issuing in an unstoppable stream. But there are people who have their eyes on the Big G and the eyes are not full of admiration. In fact, they’re sizing Google up for a take-down. According to a very interesting story by Randall Stross in The New York Times  for June 24, 2007 which inspired this post,  various development teams   are concentrating on what they think to be the weak spot in Google’s armor…the search algorithm. That is also the strongest point, so a little bewilderment at the apparent paradox is perhaps in order. And the way into the heart of the Death Star Battlestation is to use…humans! Briefly, the idea is trump Google’s search returns by using editors to  refine the content of the database from which search hits are gathered and sent to the user. There are some advantages to this method: humans can detect and kill informercials  and other effluvia that have some topical information but are largely sales pitches, and the editors can construct useful subcategories into which the results can be sorted.  Editors can also create a basket of quality sources on different topics. This is  not really new. In the early days of  search enginery, some services were structured around  a directory of categories, created and managed by humans. It was a point of pride that no resource be added to the database without some human vetting. Keyword searching was the other, competing architecture, and in time, both methods  began to appear on web search services, with the human element going into decline. Now, some developers want to try that human thing again, and some products are emerging onto the market. One drawback, it seems, is that the editors will have to figure out some list of “important” or “frequent” search topics and terms in order for the editors to analyze and classify the results. So, there could be lots of material on the antics of various celebrities and not so much on the Flemish Wool Trade in the Ninth Century.  But, you have to start someplace. There is some big money to be made in search, even after Google, and quite a few groups are out to get hold of it. What comes to mind is the pictures of those first scrawny mammals, scurrying around the feet of T. rex and just waiting for their chance.  Some of these newcomers are trying to challenge Google on its core competency  by designing a better search and ranking algorithm.  It’s all early days yet. We shall see.

The new engines include: Hakia, Accoona, Powerset, Squidoo, Sproose, NosyJoe and Mahalo.
 
New Engines

I began to revise the Copyright Crash Course over the summer in preparation for moving it from the University of Texas System servers, where I worked in the Office of General Counsel for 15 years, to the University of Texas at Austin Libraries servers this fall. Along the way, I acquired a wonderful new design, thanks to the Libraries’ Matthew Villalobos, and a new emphasis, complements of my new job here at the Libraries, with its focus on scholarly communication.

The Crash Course includes 4 major topic areas: Own manage share, offering information about rights, responsibilities and opportunities in copyright ownership; Building on others’ creative expression, addressing the issues involved when we use others’ works in the process of creating our own works, either directly, or indirectly as when we use them in teaching; Copyright in the library, detailing the special rights belonging to our libraries under the Copyright Act; and University administrative interests in copyright, where I write about special concerns at administrative levels (policy, risk management, scholarly communication).

The site is as ready as it will likely ever be, though there are little tweaky things I will continue to work on through the end of fall semester. Nevertheless, it is definitely time to launch, so I invite you to come visit and let me know what you think.

The old site will remain up for a short time, after which the pages I have moved to the new address will be replaced by 404s that indicate the new site’s url.

Over at The Ubiquitous Librarian, Brian Mathews posts about a quick peek at IR statistics. Mathews asked the folks at the institutions with the ten largest DSpace collections for a list of their most-viewed items.

Nature News for Nov.7 has a brief story on the progress of the bill authorizing mandatory deposit of manuscripts reporting the results of research funded by the NIH.  The deposit measure is contained in a broader bill, to be voted on this week, that will increase the NIH budget to around $29 billion.  President Bush doesn’t like the increases, which he has termed irresponsible, and has promised a veto. But sponsors have attached the funding measure, and by inclusion the mandate, to another bill  relating to the Department of Veterans’ Affairs. This legislation is popular enough to gather the votes necessary to over-ride the expected veto, or so the supporters hope.

 I was waiting for something funny to appear….funny strange, not funny ha! ha!…and in the last paragraph of the Nature News story, there it was. The mandate would apply to fiscal 2008 ONLY. It would have to be renewed each year!  This is an absurdity, but is it one to be grateful for, or to reject?  The Solons are saying, yes, we buy the argument, but the secret motto of Congress, chiseled in invisible letters over the Capitol entrance is: “we don’t want to hurt or annoy anybody”. So, they give these guys something, and they give those guys something, not enough for either, but too much to say Congress didn’t help.  Talk about a poison pill. In the myth Sisyphus was condemned by the Gods, to roll a heavy rock up a hill, only to  have it slip away as he neared the top, so he would have to start again. OA advocates will have to do this every year.

Dorothea Salo makes a great argument for streamlining the submission process to upload things to institutional archives at Caveat Lector » Less cognitive load, faster deposit. I hope our own Texas Digital Library designers are on to this one. But she also identifies the licenses as a major area for improvement. That’s something I can help with, and I’ve made a mental note about it. We don’t need no stinkin’ licenses! At least not at the item level as concerns the relationship between the scholar and the library. She’s absolutely right. We do have to find a way to get Creative Commons license terms that match faculty preferences affixed to documents, however, but that seems like something that could be batched. But then I’m not a techie. Darn, I wish I were a techie.

She even goes so far as to suggest that maybe navigating submission processes should be offered as a service, at least in the backfile context. We are taking that approach to current submissions in a project with the School of Nursing here at UT Austin. We (Lexie, Roxanne and I) are working with the administration and several faculty members to create a process that would allow the School to submit faculty papers to PubMed Central quickly and efficiently. Faculty only publish one or two papers a year. A process you have to repeat that infrequently, especially one that can be a bit complicated at times, has to practically be “relearned” each time. On the other hand, if a departmental administrator deposits everyone’s papers as published (with required embargoes implemented for access rights), we create a more efficient process, one that faculty are more likely to take advantage of. Admittedly, there are faculty who will deposit their papers themselves and that’s fine. But there are also a lot of faculty who would be happy to hand the process off to someone else. In light of the possibility that the NIH policy suggesting submission to PubMed Central could become mandatory, if not this legislative session then sooner or later, exploring how to streamline submission is an important consideration now. Whether it’s PubMed Central or our institutional repository, there’s opportunity here to increase submissions and we ought to be taking advantage of it.

As for backfiles, she is right again. The need to batch-process these is critical. That’s the kind of thing that we can do and we can’t expect faculty to do in any effective or efficient way. Comments from our techies? Are we on this?