Free Knowledge = Better Science
I am not battle-weary professor. Neither am sore after having my work rejected by multiple journals (not yet at least). I am a humble graduate student trying to prepare my first manuscript. In the process, I have discovered the pitfalls of multiple journals and their editorial policies; each with its own formatting guidelines, way of presenting ideas, kind of work expected, etc. While hunting for literature, I was faced with journals that my institution hasn't subscribed to; an article could cost upto USD 75! Thats when the ideas below occurred to me.
I believe that knowledge should be freely shared. Hence, I am against the practice of scientific journals charging exorbitant subscription fees to allow readers to access their online material. A better option would be to charge the authors themselves enough publication fees to sustain the cost of maintaining these knowledge databases. Most serious research projects are heavily funded, and there may be institutional memberships to get research published rather than access literature. Apart from allowing researchers with institutional affiliations to work from locations outside their organization, this would breed a lot of independent startups and amateur researchers who, without the need to buy expensive subscriptions, would be encouraged to contribute to scientific development, not unlike the 'hackers', who spurred great developments in computers over the last three decades. Furthermore, it would automatically put pressure on academics to publish, if at least to justify their institution's publication membership, which would be at a premium to subsidize under-funded organizations.