1

I have nearly 1750 papers in PDF on one folder, all pertaining to medical research. What's the best way to organize them? I have thought about creating one folder for every subject/sub-category, but I want to know if there's anything more efficient that many academics use.

  • i do often use the same strategy... – Roboticist Apr 3 '16 at 14:58
  • 1
    How do you usually access them? What kind of searches do you want to be efficient? – Federico Poloni Apr 3 '16 at 15:47
  • Used to do filenames with author first, but find it now incredibly helpful to put the year first. I found it really helpful to think of papers in terms of classics or recent ones. Sorting them by subject into folders is something I am also doint, but this is sometimes tricky as subjects overlap. Actually, I am thinking about sorting straight by journal, but have not come around to do it. – vanao veneri Apr 3 '16 at 17:41
0

I highly recommend a software to manage this like Endnote, Zotero, or Mendeley. The last two are free if your library is small enough (depends on the size of the PDFs). That will make things much easier by cleaning up the metadata and helping you use search tools to find the right papers.

In each of these software, you can also sort the papers around. The key is that within these software, you can put a paper into multiple folders/categories/groups/whatever else they call it. I use this liberally. Note that the software will keep the reference as one reference with multiple links, and so if you do things like annotate or add notes to the file they will be visible at all locations where you can find the paper.

  • Before doing anything make sure the metadata is all correct. You can always last resort with using the search bar if it is.
  • First put each paper into a field. So I would make a folder like breast cancer, and then have a subcategory like metastasis if I have a bunch of papers specifically on that subtopic. Finding odd papers reduces to remembering what subtopic it's from, and finding that subtopic folder within the right topic.
  • If a paper works in multiple different places, I put it in each place.
  • Then add the papers to project folders. I make a master folder for each project, and subfolders for different parts of the project, put the appropriate references. The software can now automatically generate a file with all of the references for the paper which helps with citing.

Using this method, you can always dive down into a folder/subfolder that makes sense, either by project or by field, then sort by title/author/year/etc (and easily switch between these), or use the search in the subfolder, and always quickly find what you need.

But... you will need to take a day or two to set this up. That's a long and tedious task, but definitely worth it. Then I usually just do a minor fix up for new additions every few months to keep it tidy.

| improve this answer | |
0

I sort papers by author, since the main times I need to look at them is if either it is cited somewhere and I want to check what precisely the paper says (in which case I am handed the authors by the citation), or I need to look at a specific paper because I recall there being something useful in it (in which case I almost always recall at least one author).

The more practical side of my system consists of a set of menus in my top bar, one for each of certain ranges of starting letters. Each menu then has an entry for each author belonging to that range, and each such entry has a list of papers by that author (so a paper with many authors will be listed once for each author). This way, any paper that I have already indexed like this is just three clicks away, and even when I only recall the author and not the precise title, it often helps to see a list of the papers by that author (especially as the list will be limited to papers I have added myself, rather than a full list of all the papers that author has ever written).

A similar system could of course be set up by subject if that would be more helpful, possibly with major subjects separated into subtopics (again listing papers several places if it belongs to several subjects).

| improve this answer | |

Not the answer you're looking for? Browse other questions tagged or ask your own question.