Searching for the needle of information in the haystack of the computing web pages

摘要

Perhaps because we were the third web site in the world,perhaps becaure our laboratory operates under the principle of anarchy,perhaps because coordinating computer professionals is like herding cats -- much of our documentation became a haphazard collection of web pages maingained by many different people on many different web servers.The answer was obvious:add a search engine.After gathering a list of servers,cleaning up the irrelevent files,protecting some directories (lide logs) from being indexed,adding meta tags,and making sure that the search index was run on a regular basis; we found that searches still resulted in pages and pages of irrelevant links.We obviously needed more control.the "grand" search was fine for the experienced web searcher who was looding for some very specific information,but we were not helping the average user.Yes,we had a table-of-contents type page where users could wend their way down to the topic of their choice,but everyone has their own logic for grouping topics and one person's "obvious" is another person's "obsure" (and every department head has an opinion about what should appear on the main page).