April 25, 2005
NPR story about testing for technology competence
NPR's Morning Edition show today featured a story about a new standardized test created by the makers of the SAT to be used to test for computer literacy. Several students were interviewed; one stated that "everyone in my generation is expected to know how to use computers and make Power Point presentations, but not all of have those skills."
The story was a bit muddled from a librarian's standpoint, because the issues of computer literacy and information literacy were compared interchangeably, which they are not, in my mind. Computer literacy relates to specific technical skills of using computer technology and software; information literacy refers to the ability to make judgments about the sources and quality of information resources. One can have outstanding computer literacy skills while being information illiterate. One student interviewed for the NPR story described his primary source of information as Google, and that his professors "didn't care" about the quality of his information, as along as he cited it correctly. The NPR reporter went on to say that the faculty of this student's institution did, in fact, care a great deal about the information, but was at a loss to know how to bring these skills to the students.
KSL librarians are trying to work closely with faculty (particularly those involved with SAGES) to bring these information literacy skills to undergraduate students. Click here for a current project sponsored by KSL relating to information literacy. (Maybe win an iPod Shuffle too!)
April 22, 2005
An Interesting Article About the Problems of Digitizing
I am currently pushing on my colleagues at the library a very interesting article by Richard Preston that appeared in the April 11, 2005, issue of The New Yorker about the digitization of "The Hunt of the Unicorn" tapestries from about 1500 that hang in The Cloisters, the branch of the Metropolitan Museum of Art in New York that displays medieval art.
In the process of restoring the tapestries, the museum decided to digitize them, both front and back, so that there would be a permanent digital record preserved in case a catastrophe caused the destruction of the artifacts themselves. The museum hired very reputable consultants and photographers to make very high resolution images of the tapestries. The images were made in segments, with the idea that they could be "stitched together" into a seamless single image using Adobe Photoshop software. But when the museum staff tried to do so, they found that the images were far too large (filling more than two hundred data CDs) and complex to manage.
The museum then turned to two mathematician brothers, Gregory and David Chudnovsky. The brothers are number theorists and built their own supercomputer out of mail order parts. Their previous claim to fame had been to using their homemade supercomputer to calculate the humber pi to beyond two billion decimal places. The brothers thought that assembling the images would be a piece of cake.
But when they made their first attempt, it failed, and the Chudnovskys had no idea why. Upon further investigation about the process used to digitize the tapestries, it was discovered that they had changed shape very subtly while lying on the conservation lab floor being photographed. The Chudnovskys realized that they were working with an image of a three-dimensional structure. This required the brothers that they would need to recalculate every pixel of every image in order to make the image of the tapestries and to correct for subtle differences in color that had occurred during digitization. It was a series of computations, taking three months with their supercomputer, comparable to that of DNA sequencing. After the computations were completed, the final assembly of the first image took twenty-four hours of supercomputer time.
At the end of the article, Richard Preston, the author, describes revisiting the restored and re-hung tapestries at The Cloisters. Despite the phenomenal technological feats that had been used to create the digital image, the real things were still vastly superior, "full of velvety pools and shimmering surfaces, alive with color and detail. ... In comparison, the digital images, good and accurate as they were, had seemed flat. They had not captured the translucent landscape of the Unicorn tapestries."
April 13, 2005
CNI Briefing Session: OhioView and Remotely Sensed Imagery
John Millard, Digital Services Librarian from Miami University of Ohio, gave a project briefing on the collection of satellite imagery that has been created as part of the OhioView project. There is an explosion of interest in this satellite imagery due to growing awareness of the utility of the data in agriculture, cartography, education, forestry, geology, and urban planning.
OhioView is collaborating directly with research faculty at multiple institutions to build and manage a shared collection. The facilities of OhioLINK and the NASA Glenn Research Center in Cleveland are used for storage and display of the vast amounts of data. This consortium has been active for about ten years. Mr. Millard explored the next steps, particularly as related to funding the ongoing project, and aspects of the project related to the aging of the LANDSAT satellites that collect the data.
CNI briefing session: Shibboleth and InCommon
Steven Carmody, "Mr. Shibboleth Of All", gave a project briefing on recent activities with Shibboleth, an inter-institutional approach to authentication and resource sharing that is beginning to replace IP address authentication for sharing electronic library resources, especially those licensed from commercial publishers. (I note that Case has a nascent Shibboleth pilot project.) Mr. Carmody pointed out that Shibboleth is now poised for wide-scale adoption. NSF will adopt Shibboleth for its Fastlane service on July 30 as part of the US Government's e-authentication initiative. A number of library vendors with whom we do business have already implemented Shibboleth. These include EBSCO, Elsevier, OCLC, JSTOR, and ProQuest. OhioLINK is beginning a pilot project, of which Case is a part.
What this means to the end-user is that in most cases a login (your Case ID and password) will be required to use library resources to their fullest level. Some lower level of service might be available through IP address authentication. (These policy issues have not yet been determined for Case.) There are several issues that are raised for libraries, the most significant of which is library walk-in visitors. Mr. Carmody said, "We have a way to handle that;" however, he did not elaborate.
The bottom line is that vendors of online electronic resources for libraries are becoming much more aggressive about enforcing their licensing agreements and the days of "everybody can use it for free" may be drawing to a close.
CNI briefing session: Orphan Works
One of the most interesting CNI sessions was given by Marybeth Peters, Register of Copyrights at the U.S. Copyright Office. The Copyright Office is examining the issues raised by "orphan works," i.e., copyrighted works whose owners are difficult or impossible to locate. This comes into play when someone is seeking copyright permission to reproduce, digitize, or use a copyrighted work in some other way. Concerns have been raised that the uncertainly about ownership may needless discourage subsequent creators from using such works in new works. Libraries certainly want to make use of such works for digitization; other examples (even more difficult than textual materials) include photographs, documentary films, early musical recordings. Photographic images are particularly problematic, since often there is no attribution of ownership or creation on the item itself, so there is no way to track down the copyright owner.
Ms. Peters stated that all parties concerned--publishers, the copyright office, the Congress, scholars, libraries--acknowledge that there is a serious problem. (It is probably one of the few times that such a diverse group has been in agreement about anything related to copyright.) The Copyright Office is in the midst of a study regarding orphan works, and comments from the public are solicited. After the time of comment is completed the Copyright Office will compile recommendations about potential legislative, regulatory or other solutions.
Two CNI project briefings
On Monday, April 4, at the CNI meeting after the opening plenary session I attended two small group "project briefings" where libraries and other organizations give reports on specific activities.
The first was about a project Stanford University Library is doing with the software developer Grokker. Stanford has been collaborating with Grokker to develop the software's federated searching capabilities and the presentation of search results, which purports to return search results in a way that is more useful and much easier to navigate than the usual list format. Grokker groups results topically and presents them in an interactive visual map. Stanford has done a fair amount of marketing of the Grokker project on their campus. It is clearly still in the realm of research project, although it was possible to see the potential, especially for students who think more visually than verbally.
The second session of the afternoon that I attended (out of a choice of 7 or 8 possibilities) was devoted to a project that the library of the University of Minnesota has started to provide personal, class-related and departmental blogs for the students, faculty and staff of the university. Anyone with a university network ID and password can create a library-hosted blog. They are using the Movable Type platform, as is Case. They have not yet, however, upgraded to the version 3 of Movable Type that Case uses.
Several interesting facts:
- ~65% of blogs are abandoned within the first month.
- ~50% of blogs have one post or fewer (ie, people sign up, but never post anything.
- FERPA regulations have come into play, because several faculty members have required students to participate in class blogs, but FERPA prohibits the requirement that students divulge their private information. U of Minnesota has gotten around this by creating group blogs where all of the authors are anonymous.
- The University of Minnesota libraries plan in the long run to archive and preserve the content of the blogs as part of the "cultural memory" of the institution, in much the same way that the student newspaper is a formalized cultural memory. (The question was raised as to whom the blog data belongs to. University of Minnesota is considering a "click through" license that will require bloggers to grant the library the right to preserve the postings. Bloggers can now delete any or all of their blogs at any time, and this will continue to be the case.
- They plan to upgrade the search engine to something other than the built-in Movable Type search capability, since it does not work well for large numbers of posts.
Coalition for Networked Information, Washington, DC, April 4-5, 2005
Last week University Librarian Joanne Eustis and I attended the semi-annual meeting of the Coalition for Networked Information (CNI) in Washington, DC. These meetings always feature a wide array of updates on what libraries are doing in the way of networked information and digital library projects. They also always feature sessions on emerging technologies and standards as well as some thought provoking sessions that may not have immediate applicability to libraries.
The opening plenary session of last week's meeting featured representatives from the so-called "Google Five" -- the libraries with whom Google has negotiated agreements to digitize all or large parts of their libraries. These libraries are Stanford, Harvard, University of Michigan, New York Public Library and Oxford University's Bodleian Library. A press release about Oxford's arrangement can be found here. Information about Stanford's project is here.
The libraries are all under strict non-disclosure agreements with Google, so there was not a wealth of information divulged. The amount of material to be digitized at the libraries seems to range from "pretty much everything" at Michigan to "a tasty selection in the tens of thousands of volumes" at New York Public. It was clear that Google had done a "divide and conquer" move with these libraries, in that they each did not know who else was involved in negotiations until they read it in the news release.
It appears that Google will give the libraries preservation-level copies of the digital images that are created as part of the project. This will run into many terabytes of information, for which the libraries do not yet have a clear plan as to how they will be used. Harvard and NYPL seemed the most cautious.
For all of the massive marketing splash that Google made with this announcement, there seems to be no doubt that for the foreseeable future, Google's plan to include some unknown part of this full-text material in their database will not radically affect libraries or the need for libraries to continue digitization of their own materials. When someone says, "When is Google going to have all the books in the work?" you can honestly answer, "No time soon."
Congratulations to Nathan, Jenny and Abigail
The Kelvin Smith Library congratulates the library's Technology Officer, Nathan Lambert, and his wife, Jenny Lambert, on the birth of their daughter, Abigail Grace Lambert, on Wednesday, April 6, 2005, at 6:23 AM. Abigail was was 17 inches and weighed 5 lbs 14 ounces.
All family members are doing well. Nathan is taking some time off work to be with his family. We expect him back in a week or so.