I have a lot of papers
Sun Jul 16 11:22:38 BST 2006
Continuing with the moving theme, what to do about the couple of filing cabinet drawers worth of papers I have acquired over the last seven years in Oxford (plus the ones scattered around my room, desk, boxes from previous moves)? As for the books there is a technological solution.
All of the papers from the arXiv can be indexed in a single text file with lines of the form quant-ph/0601012 for example. For ease of entry use abbreviations eg qp0601012, ph0503145 then use a string replace in Vim to convert to the fullform. Then either automated download of the papers in much the same way as the isbn numbers for my books (but put in a delay of several seconds to avoid the rampaging robot trigger), or just the abstract page for each paper and grep out the title and authors to convert into a link. The arXiv isn't going anywhere and is open access, so I'll probably go with the second option and create a links page of papers in my blog.
Similar for the American Physical Society journals (Phys Rev Lett etc). Here each paper needs journal title, volume, issue and page in order to be downloaded from the website, for example PRAv62i5p25415 should be enough information to download the abstract (or entire paper) of Phys Rev A vol 62, issue 5, page 25415. For these journals I would be tempted to download copies of the entire papers, to guard against the future if I ever leave academia and hence lose access to the journals.
This should take care of the bulk of my papers, need to keep paper copies of the ones that are not online or not available for automated download. For example Nature papers are indexed by a number that is not a simple function of the volume and page number, probably to stop people from doing what I am trying to do.