The reader's idea of electronic distribution of technical journals (Adam J. Coleman) can go one step further. Assuming that everything is distributed electronically, it would be trivial (technically) to introduce a system to show a full record of the number of orders for individual "papers" as well as feedback (useful/not useful) by the purchaser. For example, as an occasional freeware writer, I judge my success by the download count (in Apple eWorld).
The inspiration comes from a professor (at McMaster!) who published more than a dozen papers in the first year he started. It made me realise that the so-called "knowledge explosion" is in fact more like "publication explosion" thanks to desktop publishing. When someone says "knowledge doubles every 18 months", it may be true in fields such as genetics, but not in others where there are lots of rephrased copies of the same thing. Just look at the sheer number of posts in USENET newsgroups asking and answering simple questions.
Using electronic distribution is an opportunity to change the "Publish or Perish" paradigm to more logical "Create New Knowledge or Perish". The human resources review process can even be automated by a decision support system with an intelligent agent to scan the download count (quantity) and grading by the readers (quality). Of course, the weighting system will have to be carefully designed. Eventually, there will be encyclopedia of true knowledge base so that every progress of civilisation would be properly attributed to the people who really contribute knowledge rather than those who publish most and speak loudest.
