Slashdot Mirror
A Bid for Public Access to Fed-Sponsored Research
An anonymous reader submits "Your taxes support lots and lots and lots of research that gets published in journals that you can't access without paying absurd fees to the journal publishers. So, for example, if you'd like to read the latest research on SIDS (Sudden Infant Death Syndrome) because your pregnant wife had two sibs die of it, you can't, even though you paid for it. Well, somebody's trying to fix this — there's a pending bill (Federal Research Public Access Act of 2006, S.2695) to require public access to Federally-funded research: This would let anybody access the work for free within six months of its acceptance for publication by a peer-reviewed journal."
There's a problem with this: scientists can't promise success. I'll be afraid of accepting a grant if it's based on me promising to deliver results. The current system is that my next grant application will be reviewed based on what I did with the previous one. But it's crazy to expect basic research to work like clockwork. Moreover, it's difficult to judge things in hindsight. In mathematics in most cases people don't solve the problem they set out to solve, and in any case they do it in a completely different way from what they said they will. It's best to think of a theoretical science grant and giving someone money and saying "just do something good with it". What you really want to know is how many papers were written using the grant and if they were good, but not if the specific things the reseracher thought they might do three years ago were actually done.
Experimental science is very different. Grants are much bigger and are earmarked for specific projects. But still, say the project failed -- what the scientist wanted to do cannot be done. Then this is, in itself, a scientific result. Now it may be that in hindsight it may have been better to fund a different approach that might have worked, but the real question is whether this could have been figured out in advance.