chroot said:
I'll also teach you another secret: most universities have paid subscriptions to all the major journals. If you walk into the library and sit down at one of their computers
Around here, yes the university has paid online subscriptions to all journals, but you can't sit down at a computer without a university student ID.
Nonetheless, universities are no longer maintaining paper subscriptions, and tend even to be disposing of what they already have (space is only increasing in value). (I imagine Canberra's National Library of Australia *might* have all significant scientific journals.. but for people living anywhere else in the country?)
This means that if you want to read some "obscure" (ie. specific) journal article, you need to persuade some university student to download it for you.
This is worrisome. People frequently should be able to research scientific literature, which normally involves first reading one paper, then finding all the papers it referenced, then reading some of the papers that those (in turn) referenced. Unless the publishers individual article download rates are all exceedingly cheap (which they aren't), it just isn't affordable for a private individual to go through the literature like that. And because relevant articles invariably span several different journals, an (expensive) online subscription with anyone publisher is also insufficient. Even if the papers are out of copyright (part of the problem must be that copyright law is now dictated to serve powerful corporations rather than purely encouraging new creativity), I'm not aware of any "project gutenberg for scientific journals" (anyone? [Edit: oh, is PROLA free? What about other journals?]).
Somehow, as a side-effect of the means that were historically necessary in order to share information, we've ended up in a situation where (though the majority of scientific research is publicly funded) the public cannot access the resulting scientific literature.