What's going on here?
This presentation is about relational or text-based database searching on the Internet.  Flat files (phone directories) and directories (like Yahoo) will not be covered.

Choose the right tool for the job.
This is the toughest part --- pick the right engine for your research task.  The best tool for the job isn't always the biggest or easiest to use.  It isn't always the one everyone else - especially the general public - is using.  Windweaver tells you which engine to use based on your search goals at
http://www.windweaver.com/searchguide.htm#WHICH

Keep the toolbox handy.
As PC Magazine notes, one tool won't do the job anymore.  Develop a collection of engines you understand and use them.

Map your course before you leave home.
Experienced researchers talk about the complex searches executed through fee-based electronic services.  Borrow some of this good research behavior by making a plan for your Internet search before you begin.

Go ahead.  Kick the tires.
It's tough to keep up on all the engine news, so run your own tests periodically.  Or, even better, subscribe to Search Engine Watch, the premier tire-kicker, at
http://www.searchenginewatch.com

Take 'er in for a diagnostic:  Engine Surveys
PC Magazine: Your Complete Guide to Searching the Net (12/02/97) is the best yet at
http://www.zdnet.com/pcmag/features/websearch/edchoice.htm

Terry Mertz collects engine reviews, tips and more at
http://www.library.carleton.edu/staff/terry/websearch/

The surveys usually reflect the general public's engine interests on the Internet.  How do the results of these surveys fit your more sophisticated professional needs?  Of the features rated, what really matters to you in your work?

A few search tips . . .

• If the engine has an advanced search feature, use it.  And read the help files!
• Some Boolean operators do not work unless you are in the advanced search area of the engine.  Quote marks are not Boolean operators.
• Use correct spacing.  Enter alternate spellings by using the (precedence) feature.
• Directories are indexed by humans.  Yes, search companies hire real researchers!  Decide whether having humans make your choices for you appeals to you more than machines making the choices.
• Is the information you want in the title of a page or deeper into the document?  Search engines index Web sites in different ways. One Web site marketer advisor, Submit It, reports that Excite does not use tags (titles) but instead generates keywords summaries are automatically generated by looking for common words within a page.  HotBot and Infoseek index by keywords and meta-tags.  Lycos creates titles and descriptions from the text and cannot create an abstract if a page opens with an image map.  Learn more about how Web site creators can 'push' their sites to indexers and engines at

http://www.submit-it.com/subopt.htm

For more help, visit Timothy Day's Strategies for Searching Electronic Databases at
http://www.libraries.uc.edu/libinfo/ocas/strategi.html

Or visit Mike Holderness' The Internet for Journalists site for search tips and explanations at
http://www.poptel.org.uk/nuj/mike/lecture.htm#searc

http://www.zdnet.com/pcmag/features/websearch/ms.htm

Visit the Engine Collectors:
Some people have done amazing jobs of collecting search engines.  Here are a few:

Internets.com - http://www.internets.com
Search "1000's of databases" here.  Internets mission is to be the "largest, fastest and most recognized global link to online resources including archives, libraries, research databases, catalogs and statistical data.  Access MedNets here.

Web-Search - http://www.web-search.com
"Why choose between search engines when you can have them all?" Web-Search asks.  Despite the lofty come-on, Web-Search's approach might interest researchers.  It's indexing the search engines on individual and corporate sites.

Brain Forest - http://www.brainforest.com/search.htm
No descriptions here but there are more engines than we can ever count.  The purpose of some can be inferred.  Others will take a slow day to explore.

Windweaver - http://www.windweaver.com/searchtools.htm
Windweaver has a modest collection of engines but outlines a few directories and libraries, too.

Beaucoup - http://www.beaucoup.com
Beaucoup links to thousands of engines, directories and indices.  Go to the Big List for a collection of international engines.

And your specialty is . . . ?
Looking for regional, international or specialty engines?  Go to Amanda Harizan's  collection at
http://web.ntu.edu.sg/library/specialframe.htm

Visit IFINDIT's collection of regional search engines at
http://www.ifindit.com/regional_seaches.html

The Exotics:
Here are a couple of search sites that put a different twist on searching --

Human Search - http://www.humansearch.com
With an index now of questions it has answered, Human Search is even more intriguing.  Touted as the "human-based search engine," Human Search answers questions posed by visitors.

Ask an Expert - http://www.askanexpert.com/askanexpert
Ask an Expert gives you a search engine and then categories as the first places to look for answers.  If that fails, you email their stable of experts.  For example, you may ask the Cuba Expert, Banking Expert or even the Fundraising Expert.

What's next?
At this writing, the advent of portals are the rage.  Lots of deals being closed, plenty of alliances being made.

The claim is that the focus is shifting from indexing the entire Web to providing the kind of content which attracts visitors.  This is bad news for researchers because of course we want it all indexed, right?  Their content choice won't be our content choice.

The latest?  Stickiness - how long visitors stay at a site.  It's a concept that is affecting the price of ads.

Oh, and rumor has it that Microsoft is going to come out with something bigger and better . . . ..

                              "Warning - optimism ahead."

Jesse Berst's ZDNet Anchor Desk interviewed Dr. Joseph Bordogna, Acting Deputy Director of the National Science Foundation on May 21, 1998.  NSF got us where we are today, you might say.  Dr. Bordogna talked about the future of the Internet as a time when "knowledge will be available to anyone, anywhere, anytime." He uses the term "knowledge" instead of information to "extend the concept beyond mere data to relevant information in a useful form." (Researchers may wish to stand and cheer now).

Headers, Spoilers and Dual Exhausts
Downloadable software products which will improve searching -- they're here.  Many of you are already using the Electronic Monk (http://www.electricmonk.com) or Alexa (http://www.alexa.com). Keep your eyes open for more tools like these.

Time for the Zen of researcher-type Web searching.
So, for the time being, we researchers relax and ride along with the current state of Web searching.  We accept that this is not the world of fee-based searching. We are happy to be living in such an exciting time electronically.  Hmmm . . .

But we keep our good research eyes on the issues facing the evolution of Web searching.  What influence will the marketers have on searching?  We know they want to bump their sites up in the indexing by pointing robots to their sites for quicker indexing.  You know about 'stickiness,' now a measurement for advertising revenues and perhaps driving the development of portals, channels and other features.

How functional portals will be for prospect researchers is still to be determined.  Personalizing a portal to meet the needs of professional researchers isn't likely to be in the lineup at this early phase.  But as this concept develops, we may begin to approach the kind of search capability researchers seek.
 

                                         - Washington Irving (from Stephen Stanoudis' Quotations)

Boolean Operators:  a search language which allows you to indicate how your search terms relate to one another.  Relational concepts include nesting, truncation, proximity searching, inclusion and exclusion and others.

Database:  a computer file or system of data organized in records or fields for fast retrieval and ease of updating. The two types of databases researchers encounter on the Internet are flat files (phone books) and text or relational (two or more databases linked) databases.

Directory:  a hierarchical database organized by subject.  Directories are not machine-indexed but are divided into categories instead.

Dynamism (Dynamically, Dynamic): a term used to describe a machine replicating a human-like relational leap.

Library:  a directory or directories of online databases or files.

Portal:  an entrance to the Web;  a starting-off point for Internet users.

Robots (or spiders):  agents which crawl the Web searching for new or updated Web pages.  These agents record the entire text of every page within a Web site they visit.  The agent then visits all the external links noted at the site.

Search Engine:  a program that retrieves links to documents and other information on the World Wide Web based on a query posed by a user.

Stickiness: a measurement of market appeal, this term refers to the site characteristic of keeping visitors for a length of time.  It forms a point of gauging advertising fees.

Supersite: an integration of Internet, Intranet, Extranet and E-commerce sites to support business and communication networks. (For an example, go to http://warnerchappell.com/)

Syntax Adaptation: an application of dynamism where meta-searchers convert a single search string to the specifications of two or more search engines so that the engines may effectively complete the search.

Web site rating services: companies whose business it is to measure Web sites on a variety of scales, compare them and report.  Rating services are having an affect on advertising revenues.