Computers in Libraries 2002
March 13-15, 2002
Washington, D.C.
E-Books and the Future of Libraries
Victor McCrary
This was mainly a discussion of where e-books have been and what new developments are coming. The players in the field have decreased dramatically in the last two years, and more established computer companies seem to be entering the market. One hindrance has been a lack of standards for e-books, but this is being worked on. Most companies today that are still in the market are looking to provide ebooks for PDAs as the vehicle of choice, not proprietary machines.
Search Engine Update,
Greg Notess
Greg's main theme seemed to be: They are all being bought out by each other, until only a few companies are left. Increasingly, search engines are expecting payment, in some cases for inclusion of sites, in other cases payment ups the location of a site within the list of results or otherwise emphasizes the site. In some cases, search engines must be paid to crawl a site beyond the top level page. In a test Notess ran, Google produced the most results on 23 of 25 searches; Alltheweb and Hotbot were clearly shrinking in their coverage of the web. Most search engines are "pictures of the past," reflecting what was on the web from 1 to 3 months ago. In a test, for instance, Google's average results were 25 days old, with the most recent on March 6 (1 week earlier); Fastsearch averaged 39 days, Hotbot 45 days, Teoma (a new engine)100 days, and Wisenut 233 days, with the most recent item July 31 (2001)!! Plus, some engines have multiple sites, with www.google.com; www2.google.com; www3.google.com; and google.com being an example. These sites often reflect different "editions" of an engine; just selecting "google" may send you to any of the 4 sites, however. In a test of uniqueness (and overlap), Notess did a search on a common topic, and found that Google had 12 unique items, Wisenut 6, Alltheweb 4, and Teoma 2. So far, only Google is indexing PDF, PS, .doc, .ppt, and similar items on the web. For examining what was on the web, there is Wayback Machine, www.archive.org
Google Busters,
Gary Price
Trying not to cover the same ground as Greg Notess, Gary first emphasized that he is not a Google-basher. Indeed, he emphasized that Google's schedule of crawling was impressive. The whole site is refreshed every 3-6 weeks, and some sites are crawled every 2-7days (including newsmagazines and newspapers.) His notes on other engines:
- NorthLight: $5 per month.
- Alltheweb (from Norway): news sources are refreshed every few minutes; provides an option to rewrite a query; will add PDF in next several months.
- Teoma and AskJeeves: Release of a new beta version soon that will be much larger.
- Lycos (powered now by Alltheweb): When used with IE, splits page, showing links on one side, results page on other.
- Wisenut (bought by LookSmart): has something called "sneakpeak."
Price also mentioned something called ResearchIndex, from NEC. The software is downloadable for free for non-profit organizations. This is a citation searcher.
Useable and Accessible Doesn't Equal Boring,
Frank Cervone
The main emphasis of this session is that graphics and other visually pleasing forms are important, even user-friendly, if done right. "Bobby approved" covers only one disability, not all. The ultimate point was, if you want your site to truly be user-friendly, you need multiple pages, adapted to different needs: sight, hearing, and physical impairments. A flexible website (from the users' perspectives) is the key. For instance, someone with physical mobility limitations will find clicking on graphics easier than linked words. (One simple means of dealing with both physical and visual impairments is using the ALT attribute in the IMG SRC graphic tag. Using the ALT attribute is sine que non with graphics on user-friendly pages.)
There are several different accessibility standards, such as Web Content Accessibility Guidelines (WCAG) and "section 508." (Bobby uses the older, WCAG, whereas the EU, US, Canada, and Australia have followed the section 508 guidelines in their laws.)
Some tips on web pages:
- CSS: Contrary to the 508 standards, this can make a page more accessible.
- Graphs & charts: use the longdesc= attribute to aid understanding for users of readers.
- Frames: Again, contrary to much wisdom, are useful when used with correct headers and/or a "noframes" tag to give a descriptive message or alternative (text) page.
- Color blindness: Allow user control of style sheets.
- Deafness: Add captions for audio.
- Blindness: Appropriate markup of tables.
- Age-related: Allow magnification, avoid pop-up windows.
- Learning disabled: Graphics help and make concepts easier to grasp.
Cervone also mentioned some basic tests, such as the Flesch Reading Ease and SMOG (Simple Measure of Gobbledegook). For testing you need to turn images and style sheets off. A couple of sites he gave were validator.w3.org and www.bbc.co.uk/education/betsie/ (a text-based site check.) (I hope the urls are correct; he promised to add them to the conference website, but there was nothing from him on it as of 5/2/02.)
Navigation Elements on Library Web Sites--What Works Best?,
Lesley Mayo
A research report with very basic results, of how students responded to library home pages when given a task of searching for a book title. After showing some disturbingly badly designed university and university library pages, as well as some the students in the survey found excellent, Lesley gave the summary: Navigational bars were preferred at the top, found second best at the side (preferably the right side); the bars were only used at the bottom when on very long pages.
Some other information discerned during the trials included that patrons first look at the center of a page then might look toward the peripheries (or might not.) (Not your traditional top-to-bottom, left-to-right style of reading books.) Very unpopular features of pages included drop-down menus. Site map search boxes were only used when mistaken for catalog entry boxes (because they were prominently displayed on the library's home page.)
The Research Wizard Experience,
Karen K. Hein and Marc W. Davis
Hein and Davis reported on the creation of their "research wizard," a one-query-searches-everything idea. Basically, their 10 reference librarians (15,000 students) create online pathfinders for students on demand and add them to their "research wizard." The sourcecode for the wizard (IInterfaces catalog) is free.
XML for Libraries,
Roy Tennant
XML is a relatively new way of marking up documents. It is useful for documents with long-term value when used with CSS. Much less value when used with pages that need frequent revisions of content. If you are interested in XML, see Tennant's website at:
http://escholarship.cdlib.org/rtennant/presentations/2002cil/xml/
Exhibits/vendors: I admit it: One of the main reasons I go to conferences is to see what is being offered by companies and to talk with the vendors. In that respect, unfortunately, this year's Computers in Libraries was sad. Only about half as many exhibitors were there as I remember being present 2 years ago. Several sales reps said that their companies had cut back since September on how many conferences and conventions they were attending.
Which is curious, because the paid attendance at CIL2002 was the highest in 4 or 5 years!
Computers in Libraries 2002 presentations (InfoToday site)
This page created and maintained by Steve Morris,
Last Updated :
Business & Legal Specialist,
Reference & Instructional Services Dept.
CIL2002.html
20020502