Reason: I think every scientist should have the ability to reference something simply with a URL.
Motivation: Since access to many journals is by electronic means, my hard drive began to fill up with poorly named pdf files. I needed a way to keep track of them and organize them. A natural way to do this is with a database.
Real databases are network based and so using them to share resources is an obvious and natural idea. In addition once they're in a database they are searchable. We treat Information as objects, and thus we need tools for organizing, searching, and serving these objects.
Everyone I know who reads and writes papers has lots of references to keep track of and spends lots of time curating their own libraries. Why not share them? I waste an inordinate amount of time trying to find papers that I can't get my hands on, and once I have them would often like to share them with other people. I dread when I need a paper published in Nature as I know it will be difficult for me to get a copy of the paper. Nature has refused to deal reasonably with the UC system to provide on line access for the library, and if the issue I'm looking for is not in it's place - I'm out of luck.
Any documents one has in one's library can be indexed, stored, reference, and served in this way. Patent applications, research reports, published papers, abstracts, etc. Is it legal to do this? I beleive it falls within the limits of fair use. If I have a pdf copy of something a link to it appears. It's my library and that's what I made it for - so I could have access to my resources from anywhere. Sharing them is an extremely useful fringe benefit. People can use their own discretion if they choose to click on a pdf link. No one's ever told me, "No you may not photocopy my photocopy of that paper."
I'd like to make a freeware version of My Lib that people can use and improve. In the future it may have some Napster like ability to actively find resources that someone chooses to share. Currently it is simply a MySQL database accessed via scripts written in php and perl. I plan to put up a page describing how to create MyLib for those who have UNIX accounts with MySQL access. I won't have time for a more serious development effort until after I graduate - which should be soon. Chris Seidel Summer 2000