Baton Rouge --- With so much data now available through electronic libraries, research will begin moving away from modeling and observation and will focus more on analyzing data for conclusions.
This trend was the subject of an LSU Center for Computation & Technology and LSU Department of Computer Science IT Eminent Lecture Series on Tuesday, June 5, featuring Michael Lesk, a professor of library and information science at Rutgers University.
Lesk showed examples of how a lot of data is now available electronically, including data from years ago that has been scanned for digital observation. Also, numerous sensor applications are available today that make it easier to digitally capture and store scientific data.
Lesk predicted that with this technology, data lookup will replace experimentation as the new norm for research. For example, Lesk said, most meteorologists and climatologists predict long-range weather patterns by using equations. However, they now could find years with similar weather conditions, observe the data for that year and predict the weather for a particular day based on previous data.
These advances also could lead to greater social changes, such as making medical records available electronically. This way, when a patient comes into the emergency room with certain symptoms, a doctor could not only access that patient's prior medical history, but also could look up patients with similar medical histories and symptoms and see what their health outcomes were, then use that information to treat the patient.
Lesk acknowledged these advances are not certain and present some new problems researchers will need to tackle. Scientists will have to develop a set of data-sharing ethics, determine what is the best method for storing the data and decide where to store it. While most scientists agree cyberinfrastructure is the best means of archiving the data, there still is the problem that Web sites can go defunct and technology could change, rendering the data unreadable.
Lesk suggested one solution might be for software developers to begin examining data storage solutions and valuing that as much as data creation, citing a NASA example as a reason storage solutions are so needed. NASA has lost the only high-resolution video footage of Neil Armstrong and the first moon landing. If data storage solutions were in place, this footage could have been stored electronically and would not be lost.
In conclusion, Lesk noted electronic libraries are the wave of future scientific research, but technical needs exist, such as better data mining software, more data encoding and visualization applications and expertise from a research group to create an effective, usable system that allows others to access and observe the data. Also, Lesk said the research community needs a good, public example of how observing old data can be useful in solving modern research problems to fully embrace the idea.
For more information on CCT lectures, please contact CCT Manager of Public Relations Kristen Sunde at 225-578-3469.
Publish Date:
06-06-2007
