Academics: What's your filing system?

December 14, 2006 7:18 PM

Academics: What's your filing system for academic papers? As I finish up my undergrad degree and enter grad school, I'm reading a lot of papers. Currently, they're just in a tall stack on my desk and in a folder in my Mac. I'm interested in hearing filing solutions -- both hard copy and electronic -- for academic papers.
posted by pealco to grab bag (28 comments total)

I use endnote (also available for macs) to organize my papers. Most of the papers I read are available as PDFs so I just embed them into each endnote record (the only way I would find them again). I still print papers, stack 'em on my desk, read and eventually lose somewhere.
posted by special-k at 7:26 PM on December 14, 2006

I've completely moved from paper to PDFs. I no longer have any papers on paper.

Generally, I work like this:

(a) Find a paper in a database.
(b) Export it's citation to Endnote from the web.
(c) Download and save the PDF, as follows:

[Record Number] - [Author] [Year].PDF

ie.

45 - Grime 1979.PDF

Where "Record Number" is it's Endnote ID. I save papers in a "To Read" folder, then move them to a "Done" folder once I've read them and written notes.

(d) If it's a hardcopy paper, I go find it, and scan it all in to a PDF file, and save it as above. Luckily we have a quick photocopier-scanner that does direct to PDF.
(e) I use the "Notes" record in the Endnote record to...well, make notes, and record other papers to find.
posted by Jimbob at 7:35 PM on December 14, 2006

(The only failure in this system is the inability to scribble all over papers, make notes in the margin etc. but I don't find it that much of a drawback, especially since searchable notes in Endnote are more useful than scribbles on a piece of paper in a filing cabinet.)
posted by Jimbob at 7:37 PM on December 14, 2006

Currently, they're just in a tall stack on my desk

What field are you in? You've already discoverd the fractal filing system that most of my math professors used.
posted by IvyMike at 7:37 PM on December 14, 2006 [1 favorite]

Citeulike. I believe there are other systems, but I find having an online system very handy as well as having one that prioritises my ever-growing reading list.
posted by edd at 7:38 PM on December 14, 2006

I'm a materials science major, and whenever I read papers and find them remotely interesting, I file them away in a binder according to topic. I've got binders dedicated to crystallography, thermodynamics, thin-films, and so on. Of course there is overlap, but as long as you're filing them away and preferably using tabs in some ways to organize your binders, that's a good way to go.

I'd have to say that, while digital versions are convenient in some ways, nothing really beats a hard copy of a document. It seems old-fashioned, but the contrast ratio on paper still beats even the best displays. And writing notes is just so much easier on paper.
posted by Aanidaani at 7:40 PM on December 14, 2006

wow Jimbob, that's hardcore! btw, I'm an ecologist too (across the pond) and I do a similar thing except that I directly embed the file into each record (so endnote copies and internally links to it). I use the highlighter function in Acrobat to mark important passages and I use the comment box to make notes.
For non-PDF papers, I generally get someone in the library to scan & email the file to me and still do the same.
posted by special-k at 7:41 PM on December 14, 2006

Just refound Connotea - one of the other online systems, and one run by Nature.
posted by edd at 7:55 PM on December 14, 2006

Take a look at DevonThink Pro; it lets you organize, file, and search through PDFs and notes.

For hard copies, ye olde filing cabinet is still a great technology. I find it helps to put each document in its own manila file folder labeled with author and title, then file alphabetically in hanging folders. Easy to find what you need.
posted by agent99 at 7:57 PM on December 14, 2006

special-k I had no idea you could embed PDFs in Endnote! I'm still on version 7, though, so many I need to upgrade...
posted by Jimbob at 8:00 PM on December 14, 2006

Alphabetically by author or title or theme or all of the above?
posted by fionab at 8:00 PM on December 14, 2006

I file all my hard copies of papers in a hanging file, using folders that are coded by topic. That way, I can find something without wracking my brain for the author's name or the paper's title.

I file things according to how I expect to use them later. For instance, a paper that introduces a new research technique I want to try later will go in the methodology folder, even if it's actually about, say, the macroeconomic impact of the Earned Income Tax Credit. If I also think I may go looking for that paper when I'm writing about the EITC, I'll stick a scrap of paper with the title of the paper and the folder it's in in the EITC folder.
posted by chickletworks at 8:06 PM on December 14, 2006

I must say I really like jimbob's answer. But I'll throw in my answer just to contrast it with a hard copy solution. I recently finished a graduate degree in mathematics where I wrote several papers and a moderate sized expository thesis.

The key for me was using index cards which was a tip from my supervisor. First, when ever I tracked down a useful reference I photocopied it or printed it out so I had a physical copy.

If after reading/scanning I decided the paper was useful I wrote down all the basic bibliographic information on an index card. Just make sure you have enough information on the card that you can track down a fresh copy of the paper/text if necessary.

Then I tagged the relevant sections of the paper with those stick on bookmarks that 3m sells. You can just as easily high lighter, I happen to like the stickies. Then I stuck the paper in a drawer with out any organization. I had less than 100 papers that I was dealing with, so tracking down a particular paper only took a few minutes. Obviously, organizing these papers in the drawer may have some value as well.

Then the interesting part is I had a 2'x2' space on my desk devoted to these cards for my thesis. Cards would cluster together based on depending on each other and where the results were used in the thesis. I could go into details on what it means when one card was under another, etc. but any system that you come up with will be much more natural to you. What was really useful was having these index cards that mapped to papers that mapped ideas. This meant I could physically move these ideas around in my virtual paper. (Thank god I'm not a category theorist other wise I might have tried to search for deeper meaning in all of these mappings).

For papers that I wrote having this layout wasn't necessary. Instead I just kept small stacks of index cards for each paper which was usually sufficient since the emphasis was on new results.

Since my supervisor has been doing this quite a bit longer than I, he has taken this a little further. He has several old library card catelogues that he keeps these index cards in. Then on the back of the index cards he notes where he's used the paper indexed on the card (I imagine he also stores the reverse mapping somewhere as well, though if nothing else this is in the bibliography of the paper that he wrote). He also keeps perfectly formed bibliographic information on each of these cards but I think that is more a reflection of a difference in character than a difference in experience.
posted by snoble at 8:14 PM on December 14, 2006 [1 favorite]

I use Endnote on my Mac, but I've been reluctant to upgrade for so long that it's terribly out of date. Can't speak for newer versions, obviously, but I always felt like I was fighting the program. I didn't like it's Cite While You Write integration with Word (and have since moved away from Word anyway). One thing that's nice is that you can specify citation format, and simply "copy formatted citation" and paste it into your document.

Right now, I'm testrunning Yep on my Mac. It's basically iTunes for PDFs. Has tagging of PDFs and grouping. So, I tag each PDF with various relevant words, and group them by subject. Overlap of subjects is not an issue because each PDF can be put into multiple groups. It's by no means an Endnote killer, but Yep's a pretty nice PDF organizer.

Hardcopies of old papers I enter into Endnote by hand and simply file away in folders, arranged alphabetically by first author's last name. I have yet to work out a good solution to PDFs that I print out and read and make notes on. They go into the filing cabinet, but I'm not too diligent about noting which PDFs are printed and which aren't. I just don't have the time to scan all those old papers into PDF form, either.
posted by cyclopticgaze at 8:44 PM on December 14, 2006

I prefer hardcopy (for scribbling etc.) and I keep them in hanging files sorted by topic. I also keep an electronic bibliography in which I add brief notes about how useful or interesting a paper is, and any particular sections I think I might like to use later. That's in Excel - I tried Endnote but the system I have suits my purposes and I like to keep it simple.
posted by nomis at 8:55 PM on December 14, 2006

I suggest creating a paper database in Excel. Set it up so that the first column is just a number, starting from 1 and going on. The next column could be author(s), then title, then relevant keywords (I'm in a materials testing group, so it would be things like "microstructure" or "tensile test" or "SEM"), then date and citation. It can be as detailed as you like. That way, each paper is assigned a number, and anytime you're looking for a particular paper, just do a search in the database for whatever you are looking for.

The easy part is you can then just take all the paper copies you have, write their number in the corner of each one, and get a big 3-ring binder to hold them in order. If you have electronic copies, just make sure you include the paper # somewhere in the file name. Also, unless you keep electronic and paper copies of all your papers, include a column that says which format the paper is in.

It only takes maybe 30 seconds to enter the information, but it can save you hours of searching your office. It's worked pretty well for me so far.
posted by SBMike at 9:12 PM on December 14, 2006

Graduate student in Earth Sciences

Hardcopies: I file them by last name in a large drawer. Each author gets an individual folder. I used to do it by subject but my subjects cross over so much that I ended up spending too much time trying to figure out which subject a certain paper fell under. I use post it notes on the outside to summarize the paper.

pdf's: They are in files sorted by journal name. Each file is named by the author and then the year (same format as a citation in a paper).

If I haven't read a paper it goes into a file named "to be read". Once I've finished reading the paper it goes into a new file called "to be filed". Then after a few of those papers have accumulated it gets entered into endnote and filed as above.
posted by kechi at 9:12 PM on December 14, 2006

I agree with SB Mike
posted by Rumple at 9:15 PM on December 14, 2006

Tall stack on top of a filing cabinet. Oranizing them one of these days...
posted by ontic at 9:47 PM on December 14, 2006

All of the suggestions made here so far are good ones, and one piece of advice I would add is to think about how you think when inventing/choosing a filing system. What are the ways you intuitively organize and remember things? The last thing you want to do in grad school is create more cognitive stress by having to work through a filing system that you can't remember. So, FWIW, I tended to just leave all my hard copy articles in accordion file folders labeled with the class I first read the paper for. Then, if I use the paper for another project, it goes into that project's file. That just tends to be the easiest way for me to remember things. All my electronic articles are stored in one folder on my hard drive (called "articles," very inspired), with the file name saved as Author(year) and sometimes a few key words from the title.
posted by DiscourseMarker at 9:52 PM on December 14, 2006 [1 favorite]

Bibdesk, bibdesk, bibdesk. You can put in all of the normal data, toss in an abstract, annotate, rate, note if you've read it, group in categories, add keywords, make smart folders, and it even keeps everything in a pleasant BibTeX format for ease of exporting. Oh yeah, and if you give it the pdfs to work with, it can organize them automatically too. Also, free. I'm a physics grad student and I love it. You could probably use the keywords or create your own metadata type to assist with filing hard copies, too.
posted by Schismatic at 9:58 PM on December 14, 2006

I personally like physical copies on paper the best, for ease of annotation and durability and because it's easy to store not just my own finds but also stuff that other people give to me. I use lots of colored file folders, labeled by their course, project, or general research topic, and usually with the approximate dates as well. And then there are catch-all folders for each year's overflow (material from talks I attended, random articles I read out of sheer curiosity, etc.) that didn't fit neatly anywhere else; these overflow folders are where I look first when I can't remember the location of an article, and in such cases it usually turns out to be there (since, if it was associated with a specific course or project, I would usually remember that without too much help).

Every once in a while I become convinced that scanning all this junk and organizing it digitally will be a better plan for the future, and then I remember that it would take money and weeks of scanning and file cataloging just to get back to the state of organization that I already have.
posted by RogerB at 11:18 PM on December 14, 2006

[Biochemistry here.] I prefer hardcopies myself [I seem to work better when I can bring the papers anywhere and scribble on the margins and/or draw things as needed.] PDFs have never worked very well for me. I find them annoying to read, and I can't seem to design a good filing system for them, since so many papers fall between categories. Important papers that I haven't read yet, I tend to print out and carry around with my everywhere [stuffed in the pages of my all-purpose notebook] so that I can read them whenever I have a chance. Less important ones remain on my computer until I have time to scan them and see if they're worth printing.

I file the hard copies into binders by subject - or, rather, I file them by project/class, and by sub-topics within that project/class. Things that fall between subtopics and things that are notable for some reason [source of an experimental protocol, etc.] get little post-it tabby things on the side with notes about whatever makes the paper notable. Within a subcategory, papers are more or less sorted by author. It's old-fashioned, at this point, but if you like writing notes on actual paper, you might find that a similar system will work fine for you.
posted by ubersturm at 12:07 AM on December 15, 2006

Maybe the Mac has it built in (Spotlight?), but I'd also recommend using Yahoo Desktop Search or one of the alternatives (X1, Copernic, Google) to index them. I found after accumulating 100s of papers that it was easiest just to put them in folders corresponding to general categories, and search for them as needs be.

Also, I haven't tried it, but this could be useful:

Zotero - The Next-Generation Research Tool

It seems like it manages the references, like Connotea and CiteULike, and also indexes the PDFs, all in one.
posted by Boobus Tuber at 4:45 AM on December 15, 2006

Another vote for Yep. Here's what I like about it:

1. The default browse view previews the PDFs, and you can zoom them from small to very big. Bonus: if you mouse over a PDF, you get a pop-up magnification of what your mouse is over. This is insanely useful.

2. Full-text searching of PDFs.

3. Tagging. You can apply any number of tags to a PDF and then show arbitrary sets & subsets of tags.

Some smart person above said that the important thing is for the system to work with how you think. The way I remember papers is usually that I remember one author or an important phrase and for those cases full-text search combined with previewing is very useful. Or, alternatively, I have groups of papers that are relevant to different projects, and that's where the tagging comes in handy. So Yep is great for me, but it's all about how you work.
posted by myeviltwin at 7:02 AM on December 15, 2006

My filing sucks. When I started grad school, I had copies of few enough papers that filing by topic worked. Now there are too many, with too many interconnections and belonging to multiple topics per paper, for that to work elegantly, but I still haven't upgraded. I'm constantly reprinting papers that I know I have a copy of somewhere, because I can't find them buried in the "1999 project about mental content" folder.

It's all physical papers for me; I need to be able to write on them as I read. I keep them in hanging folders in a filing cabinet, and in huge semi-organized stacks in every space I'm allowed to fill.

Eventually I'd like to move to a system where I have them filed alphabetically by author (or maybe two parallel systems, broken down into For My Research and For Other Things, or something along those lines). There would be a sheet of paper in each author's folder listing all the papers I have copies of, so that when I take one out, I remember that I've already got a copy of it. There would be sheets of paper in one central place listing the ongoing bibliographies of different projects I'm working on, so that I can put papers back without worrying that I will forget them.

Whenever I'm working on a topic actively, I have one or two manila folders full of whatever papers I'm working with. The folder is labelled with the project name and is usually "anchored" to one location, or at most two.

I haven't moved to Endnote or similar, though I probably should.
posted by LobsterMitten at 7:20 AM on December 15, 2006

Another ecologist here. I use basically the same system as kechi above, using EndNote to keep track of everything (I get new versions of EndNote free from my university).

When I started grad school the whole idea of PDFs of articles was brand new and most everything was still done on paper. I'm hoping in the next few years to go paperless and several of these answers have given me good ideas.
posted by hydropsyche at 8:06 AM on December 16, 2006

Very late to the party English lit, um, person here. My system is a filing cabinet for hard copies, DEVONthink for PDFs (which I OCR if necessary using Acrobat) and Bookends. I used EndNote for years and gave up because the upgrades were too expensive.

I tried to create a "perfect" filing system for years, and then realized that a system where I could find the paper I wanted in a minute or two was more than sufficient. And in two minutes, I can search in a lot of places. It seems like a long (frustrating) time to be searching, but two minutes of real-time is nothing.

DEVONthink is *amazing* for what it does. I absolutely love the "see also" feature, and honestly, feel like I can hardly live without it. I have made some cool discoveries with it -- relationships and secondary meanings I wasn't previously considering -- and I don't think Spotlight or Yep can even compare. The interface for DEVONthink, honestly, sucks. I would love to have the looks of Yojimbo or Yep and the AI of DEVONthink, alas.
posted by terceiro at 6:30 PM on December 18, 2006