The most productive time on the Web is not spent using even powerful search engines (each of which behaves differently), but following recommendations from informed friends and other trusted sources.
Questions to ask yourself about search results:
- How authoritative is the person/organization who contructed the file?
- How frequently is the file updated? Do the links work, are they useful?
- Does the file provide bibliographies?
- Does it answer reference questions, like who? what? where? when? how many?
You can get some help in developing critical judgment about Web resource quality from the Internet Detective site.
Your transactions with the Web are not yet likely to be major transforming ones. They are more likely to become an accumulation of smaller advantages, each having to be convenient enough to be realized. Technological inconvenience is like an insurance deductible--if it is too high you get nothing back when you need it. Only a smoothly tuned computer setup will realize cumulative benefit (i.e. good bookmarking, printing, "live" URLs in email, a fast enough connection if from home, etc.).
What it means to publish will continue to change. Beyond the current print journals with parallel electronic online versions are electronic-only refereed journals, and beyond that we are seeing the beginning of "direct sales" of intellectual property (e.g. FatBrain.com).
Some of my past consulting jobs were just lookups in textbooks. People didn't take time to read or trust themselves to read correctly, even when it was fairly obvious where to look. Rhetorical question: What difference will the Web make?
The predictability of finding useful material is proportional to the degree of order in the "area" where you are looking and of your trust in the quality of the material. One of the newest topics in the latter category are "reputation managers".
If you have the advantage of advance planning, waiting for good stuff to come along is easier than having to find it right away. By "positioning" yourself suitably at the intersection of appropriate streams of information, the way authors do when collecting information for a book, "living in the neighborhood" of the topic, you become a more efficient "feeder". Let the information come to you.
Sociological and economic issues will have the greatest influence on the evolution of information retrieval.
Technologists usually give little thought to history, and to the challenges of publication in electronic form for archivists and historians. To study the evolution of a discipline, and even of technology itself, requires a durably accessible record. "Preserving Digital Information" is an important report.
Many more of my reflections can be found in EAS-INFO, an archive of intercepted Internet information and reflections on engineering and society mailed out via a distribution list (subscription info).
Dilbert, Darwin Awards, Doomed Engineers