Computers in Libraries 2004

March 10-12, 2004
Washington, D.C.

Below are notes from some of the sessions I attended. Apart from incomprehensible notes from other sessions, I seem to have misplaced my notes from a very informative session sponsored by TLC on where standards are going.

Introduction and Keynote: Expectations for Our Digital Future

Clifford Lynch
After an upbeat welcome by the publisher of Information Today on the theme, "Be the Walrus," Clifford Lynch spoke of the "unexpected consequences of technological change." Among other items, he mentioned:

the irritation with cell phones, which resulted in blocking technology, and its ramifications for doctors and lawyers who need to be able to receive calls at any moment;
DSL and cable are not really fast--not fast enough for effective video and/or filmstreaming;
the wired classroom is being objected to by professors who find it distracts students; and wireless capabilities only makes this worse (and gets us back to blocking technology and its implications);
sensor networks--now you may be on web tv and not know it, no matter where you are. Today there are very popular "spy cafes" where everyone is watching someone else in the cafe (or elsewhere) with cameras located throughout the cafe and via the internet.

In addition, there are unintended implications, such as the use of email being undermined by spam, which may be an endgame in and of itself. Increasingly, organizations are using technology, not just law, to protect their "intellectual property", including its destruction on a programmed-in schedule. Institutional repositories are thereby becoming a thing of the past. Further, the law is being used to threaten anyone who gets hold of corporate records (including email) and releases it to the public. Such security concerns are expanding property rights to the loss of the information commons.

Library Web Site Meets About.com

David King
In his presentation on Kansas City Public Library's analysis of the top websites used by their clientele, David King described their realization that most users of their computers desired local information. Yet the library's website did not include much in the way of local links, so the users were using Yahoo and Google. This resulted in a change of design, using Cold Fusion and SQL programming to make local information more available through the library's website. The speaker pointed out that academic libraries do have a more defined customer base, so not all the findings of this public library would be applicable. Further, of 17 IT professionals, 3 do nothing but work on the library's website. (A lot of "oohs" and "aahs" at this comment on systems staffing.) Their new website is not yet in production.

Cool Tool Update for Library Web Sites

Darlene Fichter and Frank Cervone
This session was mainly a list and demo of various tools available on the web. Darlene Fichter described such items as SWiSH for creating Flash animations (from swishzone.com-$49.95), toolbars from Merriam Webster (IE only, www.merriam-webster.com/toolbars.html) and Javascript (www.lights.com/~scott/chmod.html). websiteoptimazation.com will evaluat the dowloading speed of a page. Frank Cervone mentioned opensource.isc.vt.edu/products/survey which creates online surveys in diverse formats, such as multiple choice, short answer, etc. Web based forums are available via www.sporum.org. insite.sourceforge.net is available for link verification and stats of your site. Also security software is available for free to non-commercial sites from www.gfi.com/lannetscan and www.nessus.org.

Browsing off the Beaten Path

Karen Venturya
Recognizing that commercial browsers (IE. Netscape, etc.) have serious problems, including lack of patron privacy (records websites visited, credit card #s, etc.) and lack of adaptability to users with different agendas, libraries should be using alternative browsers. Some of the options are Public Web Prowser (PWB), developed specifically for libraries. V. 1 is free, v. 2 is $100 per site/county per year. K-Meleon (based on Gecko/Mozilla) is a free, open source browser which is fast, light, customizable, standard-compliant (on Windows, not Macs) and will run with Mozilla, Firebird/Phoenix/ Firefox, Mac, Windows, or Linux. Libraries need these browsers that are able to lock homepages, freeze bookmarks or eliminate them, modify right-click menu, clear web cached, disable saving a user's history, control pop-up behavior (ads only, not legitimate new windows), set time limits, or even be set for kiosk mode.

Using Open-Source Software to Creat a Digital Library

Pam Osborne
Pam Osborne, of Mercy Corps. (www.mercycorps.org), described how she had used free software and built a digital library of grants, proposals, etc. which is used by this worldwide organization. She initially built a small collection while talking with all the users about what they needed. The needs varied from headquarters' need for information on what who was doing to the field offices' needs for grants available and examples of other office's grant proposals. One particular need by the field offices was for minimal download times, due to their use of laptops and unreliable or extremely expensive phone connections. Basically, Ms Osborne used open source software from www. greenstone.org (free to non-commercial organizations) and the creation of unique metadata tags for indexing, all available through the greenstone software.

Federated Searching and Open URL

Frank Cervone
Described federated searching as searching multiple databases with a universal search that produced all full=-text results for the user. These new generation federated search engines include a resolver (similar to SFX) that provides links to where the materials are available. He descriminated between metasearch (searching metadata) vs megasearch (search engines such as Dogpile and AllTheWeb). Federated searching allows the two, metasearching and megasearching, to be combined. The results may be reported in 1 of 4 levels:

Link to search interface (as Lexis does);
Seach & link to search list;
Search & return brief record (like most databases)
Search & return full record.

The communications protocols are in the process of change, making the level 4 results more common. Some of the current communication protocols are Z39.50, http, OAI-PMH, XML gateways, SRU/SRW, Direct APIs. These are to be distinguished from metadata protocols such as MARC, DC, MOPs, TEI, EAD.

In an aside, Mr Cervone mentioned that Northwestern University is setting up a separate undergraduate interface, although it is not yet in production.

Some of the problem areas in perfecting the federated search are:

Differences between search protocols;
Resource translation;
de-duping results;
relevance ranking vs date ranking.

NWU has SFX in production, in connection with EZProxy. Metalib was installed in March with a beta test scheduled for September.

Some concerns he expressed about federated searching were that the students were not concerned with quality results; but that is up to the librarians to teach the students the value. Relevancy rankings are more important than quality, although the presenter felt that tuning will eventually solve the problems with relevancy ranking. One major benefit of federated seaching is its support for interdisciplinary areas, which are growing in all universities.

Evolution of Search Engines: Update

Panel of Gregg Notess, Brad Hock, and Gary Price
This panel, hurriedly assembled when Chris Sherman could not attend, was moderated by Gary Price, who went through his colleague's Powerpoint, inviting comments from the other two panelists. Basically a seminar of "factoids." In his introduction, however, Gary Price pointed out that information overload is truly with us: last year 5 new exabytes were added to the internet. (1 exabyte = 1,000 terabytes.) This works out to 800 mb per person on earth, roughly the equivalent of 30 shelf feet of books per person! 92% of this data is stored o magnetic media, a very ephemeral format.

Yahoo combined with Overture then took over Inktomi;
Overture bought Alta Vista;
Overture bought Alltheweb;
Findwhat bought Espotting;
Yahoo bought Overture;
Google bought Blogger, Kaltix, Sprinks, ...;
Google launched contextual ads;
The big question: when will Google release its IPO to go public?
Yahoo dumped Google;
Yahoo built a new search engine;
Ask Jeeves bought Interactive Services (owners of Excite);
Yahoo now is as big or bigger than Google; its verticle search is starting to get really good, especially in comparison of the quality of results, not just the number, especially in light of the total size of its database;
Ask Jeeves now often gives the answer as well as links;
Pay to Play is growing;
Microsoft's is working through MSNBot to crawl the web;
Yahoo now has a Personal Searcher to search local urls;
Natural language is growing;
At Yahoo, the directories are out of favor;
Metadata is dead;
Metadata is not dead; Yeah, yeah--but that is what he said!

Gary Price continues to update his site at www.searchenginewatch.com/searchday

(Several items got left out because I could not take notes as fast as the panel could speak; some of the above may even be confused--check out Price's website before citing me.)

20040414

Conference Notes Index Page

Steve Morris' Home Page

FIU Library Page

      This page created and maintained by Steve Morris,                      Last Updated :
             Business Research Specialist,
             Reference & Instructional Services Dept.
             Green Library
             Florida International University

CIL2004.html

20040405