Following up on the Salon article about the Jan Hendrik Schön fraud controversy, the APS has decided to retract serveral articles co-authored by Schön. Fortunately, they have decided to leave the articles online with clear statements of retraction.
December 23, 2002
December 20, 2002
I've been reading about Claude Shannon and information theory quite a bit lately so I thought I'd share some of things I've seen, both on the free Web and in journals & conferences:
- The best place to start is Robert G. Gallager's "Claude E. Shannon: A retrospective on his life, work and impact" from the IEEE transactions on information theory. 47(7): 2681-2695.
- From the 1994 IEEE Security Technology conference, "Claude Shannon's cryptography research during World War II and the mathematical theory of communication," by Everett M. Rogers.
- From the AIP's 21st International Workshop on Bayesian Inference and Maximum Entropy in Science and Engineering, "The information revolution in yet to come (A homage to Claude E. Shannon)," by Alberto Solana-Ortega. This is a kind of expansion and rethinking of Shannon's ideas and is therefore quite interesting.
- Another good Shannon bio, taken from his collected works. This one's by N.J.A. Sloane and A.D. Wyner, the editors of that book.
- Speaking of the collected works, here it is. If your math/engineering/computer science collection doesn't have this book, it should. I know we didn't so I just ordered it myself.
Posted by John Dupuis at 12/20/2002 02:28:00 PM
December 18, 2002
No one seems to know whether the Bogdanov Affair is a case of a Sokal-type hoax, a fraud or just a case of how theoretical physics is becoming an awful lot like Olympic figure skating -- very difficult to judge. John Baez's essay is a very good explanation of the case without the hype or finger-pointing.
Posted by John Dupuis at 12/18/2002 10:32:00 AM
December 17, 2002
Sauron the good guy? Hmmm. This essay by science fiction author David Brin is particularly interesting (and relevant to us) because it takes a critical look at the way we romanticize the past and demonize science and technology and how those ideas are reflected in Tolkein's books and the movies. BTW, I'm a big fan and can't wait to see The Two Towers.
Posted by John Dupuis at 12/17/2002 09:52:00 AM
December 16, 2002
Some not-very-surprising reports from the latest dlib. First of all, as shown in King & Montgomery's "After Migration to an Electronic Journal Collection:Impact on Faculty and Doctoral Students," electronic formats are quicker and easier for faculty to use. Similarly, as shown in Rudner & Gellmann's "Who is reading on-line education journals? Why? And what are they reading?" if you build it online, they will definately come and read. An interesting question: why would any scholar in her right mind publish in a print-only journal? Given the option, wouldn't she naturally prefer a free, online journal?
Posted by John Dupuis at 12/16/2002 11:49:00 AM
December 12, 2002
O'Reilly is my favourite publisher of compter books, bar none. And I think that has a lot to do with how sensible and intelligent Tim O'Reilly is. Take a look at this discussion of the whole Internet piracy/file sharing controversy and you'll see what I mean. Too bad he doesn't run some of the record companies, too. From Lisnews.
Posted by John Dupuis at 12/12/2002 10:20:00 AM
December 11, 2002
Indexing and abstracting services, BEWARE! In fact, as the thrust of this article from MSNBC indicates, everyone who's job is involved with sorting, classifying and providing access to information needs to know how the landscape is shifting beneath our feet. We think libraries and librarians are more important than ever in the brave new googlized world, but does anyone else? (BTW, I just checked Google to see if I was the first to coin the term googlized, googleization or googlization, and I'm not. Sigh.) From Lisnews.
Posted by John Dupuis at 12/11/2002 11:19:00 AM
December 10, 2002
Absolutely terrific article in the new Issues in Science & Technology Librarianship by Ibironke Lawal called "Scholarly Communication: The Use and Non-Use of E-Print Archives for the Dissemination of Scientific Information." It uses a very well done survey to explore the scholarly communication patterns of various disciplines with respect to e-prints. As expected, physics & astronomy come out on top and just as interestingly, chemistry comes out on bottom. This article brings together in a concrete way a lot of the threads running throught the science librarian community these days about free online scholarship, the Web and the role of institutional repositories versus discipline-based resources. Highly recommended.
Posted by John Dupuis at 12/10/2002 12:47:00 PM
December 2, 2002
After a bit of a silence due to a rather busy term, a bit of a reminder about why we all do this kind of thing. And I don't mean just for tenure, although since I'm tenure stream I can't pretend to say it doesn't influence my desire to maintain this blog. I know I do this also to be able to share my ideas and concerns with other academic (science) librarians who are in a similar position. And if doing this helps my tenure chances, well, I have to say that that's great too. The blog that most influenced me to start this up was the Scholarly Electronic Publishing Weblog, which I think is by far the best current awareness site on the web for librarians interested in the future of scholarly publishing.
And while I'm pointing to Ex Libris and talking about the future of scholarly publishing, take a look at this article from a couple of years ago. It really highlights for me the challenge of reference service these days. We're straddling two worlds, one print and one electronic. Our students want to be 100% electronic but we know that so much of what is useful for them is still in print and it can be a bit of a struggle to get them to look at print resources. But, how to get to that 100% electronic world, and who will pay for it. It should be very interesting to see where we end up in even 10 years from now.
Posted by John Dupuis at 12/02/2002 02:12:00 PM
October 29, 2002
Another installment of "popular science watch." First up is an article from Scientific American on the developer of information theory, Claude Shannon. It is a very understandable outline of what information theory is all about as well as some interesting insight in Shannon himself. There are also links to more about information theory, including an online version of Shannon's original paper.
Next up is a posting from the CBC about the physics of stone skipping, including links to the research. Who knew?
Posted by John Dupuis at 10/29/2002 11:28:00 AM
October 24, 2002
An article pointed out to me by the chair of the Math & Stats department here at York: "Mathematicians and the Mathematics Library: A Librarian's Perspective." It's by Sara Rutter and it's in the Notices of the AMS, volume 49, number 9, pages 1078-1081. It is based on Ranganathan's Five Laws and relates each of them to being a math librarian. The most appropriate is that the library is a growing organism, changing and evolving to meet the needs of its users -- something very important in this time of change in scholarly publishing.
This article reminded me of a few other older ones from Notices of the AMS. "Mathematics Research Libraries at the End of the Twentieth Century" from volume 44, number 11 and "Twenty Centuries of Mathematics: Digitizing and Disseminating the Past Mathematical Literature" from volume 49, number 7.
Posted by John Dupuis at 10/24/2002 03:29:00 PM
October 23, 2002
Going back to my October 4th posting, the full document is online. The title is "Dimensions and Use of the Scholarly Information Environment," which is quite a mouthful. A telling quote from the conclusion: "Library directors and college and university administrators face an increasingly complex institutional and informational environment. The population they serve is far from homogeneous in its level of sophistication, information needs and infrastructure requirements." No kidding. I think this is quite an important document because is starts to give us an idea of how scholars are adapting to the web and more importantly, how the scholars of the future -- today's undergrads -- are being shaped by it. The ideas and expectations that they develope in their formative years will drive the scholarly information landscape for decades to come. Or should I say the landscape that replaces the one we are familiar with from the past and which is clearly in transition.
Posted by John Dupuis at 10/23/2002 03:14:00 PM
October 18, 2002
One thing you have to say about Google -- it's really made combatting plariarism the hottest thing on university campuses these days. York has recently subscribed to turnitin.com and I know that there are other services out there. It also seems that every libray is coming up with good ant-plariarism sites, like these: 1, 2, 3. Google, ironically, also has a very good directory entry. This whole topic started percolating around in my mind when I read this article by Bill Savage called: "Get Caught Plagiarizing: Thinking About Cheating? Think Again, Moron." It takes an interesting tack: why do students plagiarize: ignorance, fear, arrogrance and all the rest. It's funny and vicious -- what could be better? Article from lisnews.
Posted by John Dupuis at 10/18/2002 11:29:00 AM
October 15, 2002
d-lib is an ejournal I check out on a regular basis. This month's issue has two articles of note for us here today. First of all, "Comparing Library and User Related Costs of Print and Electronic Journal Collections: A First Step Towards a Comprehensive Analysis" more or less confirms what seems to me to be fairly obvious: that a library's total cost of managing journal subscriptions is less for ejournals than for print journals. The study also indicates that faculty and students make heavy use of ejournals, especially in terms of reading a wider range of journals than they would have before. Like I said, this seems obvious, but it's nice when reality works out like it should. I can certainly say that my experience with my own research-related work tells me that I much prefer using ejournals to print journals, almost to the point where I won't use a journal if I don't have access to the electronic version. At York, most of the library science print journals are at the humanities/social sciences library, about a five minute walk from my branch. The idea that I should have to get up and walk five minutes, find the location in the stacks of the journal I want, find a free photocopier and actually stand around photocopying the thing flashes me back to my old MLIS days at McGill. It seems so studenty. I have to say, I understand why faculty prefer ejournals because I do too.
The other interesting article is "Open Citation Linking: The Way Forward," about which I have no accompanying rant but is still well worth the attention of anyone interesting in the whole Open Archives movement.
Posted by John Dupuis at 10/15/2002 03:30:00 PM
October 10, 2002
October 9, 2002
An interesting new free service from the fine folks at ISI is ISIHighlyCited.com. What it does is identify the most highly cited researchers in several different scientific fields, such as physics & chemistry. They intend to have 250 scientists profiled eventually, but most fields seem to have about 100 in there already. Given that it's not complete, it seems a bit premature to complain about the service, but I can't help myself. In the last couple of days, 9 scientists have been awarded the Nobel prizes in medicine, physics and chemistry. How many of these obviously important members of their respective scientific communities appear on ISI's cite? That's right. Not one.
Let me quote from their info page: "…an expert gateway to the most highly influential scientists and scholars worldwide" and "...gives research professionals working in a variety of occupations an invaluable tool to identify individuals, departments and laboratories that have made fundamental contributions to the advancement of science and technology in recent decades."
I would maintain that citation frequency has very little to do with the actual influence and quality of the work and that we should use ISI's new service very carefully. In terms of tenure and promotions decisions, what is it worth? I'm not sure. Personally, I take most of what is there with a small grain of salt. What would you rather have? Your name on an ISI web page or a Nobel prize? End of rant.
Posted by John Dupuis at 10/09/2002 11:52:00 AM
October 8, 2002
Some recent popular science articles that I've found very interesting. The first is from The Atlantic Monthly, titled "Homeland Insecurity." It's about how the security technologies we think we can rely on to keep us safe and secure aren't as foolproof as we'd like, both because of how we use them and the limitations of the technology. It's a very interesting discussion of cryptography and biometrics. The second is from Salon, and it's about the recent controversy within the physics community about Jan Hendrik Schön's apparent fraudulent research reports. This is the best summary I've seen of the issue.
Posted by John Dupuis at 10/08/2002 10:22:00 AM
October 7, 2002
Here's a link to a self-interview by Michael Bishop. I hesitate to say "favourite science fiction writer" and you'll just have to read the interview to find out why. To my mind it is one of the best expressions of the angst science fictional types experience about the lack of "respect" their genre gets in the wider literary world. From Locusmag.
Posted by John Dupuis at 10/07/2002 02:50:00 PM
October 4, 2002
I think we all kinda know that people turn to the web as their first choice when they do research -- I know I often do. The interesting thing about the research highlighted in this article from The Chronicle for Higher Education is that it shows that researchers still trust print resources more. Outsell is the company that actually did the research. The full results, as mentioned, will appear at the CLIR website within a month. The CLIR has lots of interesting reports, such as The Digital Library: A Biography. From sepw.
Posted by John Dupuis at 10/04/2002 11:27:00 AM
October 3, 2002
We science librarian types were probably math nerds in elementary and high school. I always find it interesting how readily people will admit to being incompetent in math when most of the time they wouldn't brag so mightily about being incompetent in anything else. Almost a badge of honour, in a way. This article from Salon is a bit of a chicken soup for the numerate soul.
Posted by John Dupuis at 10/03/2002 02:06:00 PM