iPapers, an OS X app for managing PDFs of papers [via Michael McCracken]. Big problem: I learned a long time ago that it's really impossible to name PDFs by PubMed ID number, which this program relies on. It needs to parse AuthorYear and offer a choice of search results (or use some kind of content hash) to be practically useful.
Comments
All fields are optional, email address will not be shown; no HTML, URLs are automatically hyperlinked.
Have you used it? What's so impossible about naming PDF reprints by PMID? Unless your specialty area is not covered by PubMed, I can't really find problems with it. Mapping PMID to any other naming scheme can be automated completely using EUtils from http://eutils.ncbi.nlm.nih.gov/entrez/query/static/eutils_help.html
Yes, I have used it: it seems like quite a nice program, if a little buggy.
The problem with the file naming is that, once you have a PDF, it's really difficult to find the corresponding PMID. At the moment I have a folder of about 250 PDFs named as authoryear (which is the only naming scheme that works well when organising them through the Finder). The only way I could start to use iPapers is if it could import those PDFs and automatically run a PubMed search based on the author, year and some title keywords extracted from the PDF, presenting a list of possible results from which the correct result would be manually selected (similar to what Gotcha Covered Pro does for album covers).
This is what I did a couple of years ago when I was experimenting with naming PDFs by PMID (with the aim of identifying them on P2P networks), and decided it was just too much trouble and time-consuming for anyone to realistically use.
THAT certainly is the most painful part. I wish journal sites list corresponding PMIDs next to PDF download links on their pages.
However, iPapers solves this for new downloads via PubMed search by assigning PMID automatically for new downloads. iPapers certainly makes good use of WebKit (guts of Safari), doing rendering of various journal download pages.
Downloading via iPapers works well for journals with direct article-level linking from PubMed. iPapers is still a bit buggy, but I am now downloading papers using it. The only problem is that PubMed data lags a few weeks for some journals, so often the latest articles may not yet have a PMID assigned.
I have over 2000 articles named by PMID, and can get to them via Finder as well with author/year scheme (for all authors) using:
http://www7.bpe.es.osaka-u.ac.jp/pubmedpdf/
These two combined work so nicely for me. But I agree, if you already have a large collection based on a different scheme, it is difficult to change over. I had to change over, because I have tried various naming schemes, but could not keep consistency in putting files into folders. With iPapers and PMID-based collection, the problem has been solved for me. I am really bad about organizing, but I don't have to, any more.
"iPapers solves this for new downloads via PubMed search by assigning PMID automatically for new downloads."
This works fine if you're willing to open up iPapers and run a new search for every paper you want to download, but I do all my searching and browsing in the web browser - PDFs open in a PDF plugin and get saved directly to disk from there. The incovenience of having to open an extra program just to be able to get the PMID is just as bad as having to go back to a new search in the browser. HubMed makes this easier by keeping a history of the fulltext links you've followed, but still, it's an extra few steps that personally I find outweigh the benefits.
I agree, it doesn't work well when browsing new journal issues and deciding to save (or not). But I decided that it is worth going an extra few steps by having 2 Safari windows open, one to PubMed to copy PMID, and another to a journal site. It's worth it for me. Besides, I have a hard time deciding under which author name I should save the file. Well-known lab head or the first author who may be relatively unknown? If you pick the former, you will regret it when the little-known become established. A given article should appear under ALL authors in Finder, which is easily possible if stored using PMID and then processed a little using a symlink script.
Where iPapers shines is for getting all papers written by a particlar author, for example. It's so easy to do a PubMed search for an author, and then start clicking on the articles found.
I'm late into contributing to this entry -- been stuck in a rut at work lately. But I have a question that may add more to the naming convention confusion. I've tried to understand the benefits of using DOI -- Digital Object Identifiers -- to no avail. Can someone attempt to explain it to me? I know this is a topic that is not very related to what has been discussed here, but it has some similarities, no? I.e. you're trying place files (digital objects) in the ether somewhere that you know you will always be able to access -- similarly, you want to place PDF papers in your computer somewhere that you know you will always be able to access immediately. If you know of where I can find a good explanation of DOIs, let me know. Thanks.
The following seems to explain reasonably well.
PMID: 14624257
http://dx.doi.org/10.1371/journal.pbio.0000057
http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?cmd=Retrieve&db=pubmed&dopt=Abstract&list_uids=14624257
My question and interest is whether MY computer system or MY electronic library is able to intercept redirection to objects to see if I already have them before automatically redirected to objects owners who want to charge (again). If everything is open access, then all that is not needed, but that won't happen any time soon.