I have three relevant sections in my files. I have a "project" section, in which each file name is a topic, an idea that may or may not be part of a research paper at some point. May include handwritten "links" to particular papers, etc. I have a "Papers" section, which has papers organized by author's last name. "Papers" includes books, etc; it was a useful name for the section. And then there's "everything else" organized alphabetically.

I have my regular A-Z filing system for everything not related to school. Plus a drawer for school/research projects. Not really alphabetical, but by topic, e.g., Draft notes for article X, Notes from Workshop Y, etc. Finally, a banker's box full of articles that I've printed from pdf or copied from the library. These are A-Z, with a unique key given to each paper that I think I'm going to cite. For papers I don't think I'll cite, I just use the first author's last name.

This is a bit tricky. I have a standard file system with some articles in there, but I think I'm going to keep my articles in a separate system after all. It just bugs my mind for my articles to be jumbled up with other stuff.

One of the things that I did do with my system is that I decided to file everything A-Z, but I also have an index. One of my concerns about filing things was that I'd forget what was in there. It takes some extra work but having an index removes that fear and thus removes one of the barriers I had towards filing.

I'm sure there is a way to incorporate endnote into this. Perhaps endnote could act as the separate index for your articles. At some point, though, I'd like to just scan everything and have them as PDFs. I tend to keep my notes separate from the physical articles, so there's no need for me to have hard copies.

Hey Caseykoons!

Another GTD academic here... I have to admit I've thought a lot about this, and haven't really implemented it very well yet. [I'm not completely implemented, unfortunately, but I'm getting there slowly.]

I think what I'd be likely to do... now that I think of it... is most likely file my academic journals in a separate filing system, and file them by first-author's last name. [If you have normal academic journals like in science... if they're different, i.e., not very self-contained, you may have to try something different.]

For me, I write in LaTeX and use bibTex for my references... BibDesk on the mac works with bibtex very nicely, and allows me to have various tags related to the work and notes. So, when it comes time to find that weird article about X that I remember reading... I can search the abstract, or my notes in my computer about it. Now, most of my articles I have in pdf form auto-organized by BibDesk by author's first name... so my physical file would do well to be organized in a mirror-fashion I think. Hmm.. you've movtivated me! This is going into my kGTD list to do! :)

The problem I've had is how to deal with subjects... I think what I might do is print out an index (most likely created from my Bibdesk program), which includes my most commonly looked for subjects, and place that at the very front of my drawer... this way if my computer is down, and my coworker wants to see everything on "Y" that I have... I can throw him my subject organized bibliography, and say "have fun!" ;) [That is, if I'm being nice. I should instead say: "dig a bit, I did" that way he'd learn. ;) he he he... Mwaa haa haa!!! _sorry_ got carried away a bit....]

I figure: the point of the "One filing system" GTD favors is because it's so easy to forget where you put things. (For instance-- the time I decided to seprate out all my "finantial" files (bank statements, credit card info, etc) into a seperate filing box. But I found that there were very fuzzy boundaries to what constituded "finantial" info, so it would indeed have caused problems later with recall: "what did I file this as?"

However, any sub-set of files that is truely hard-edged will not suffer from having it's own file system, if appropriate. I keep academic papers in a seperate file system alphabetized by first author (I use Biblioscape as my index to the system, if I need to find something based on some other criteria). I'm not going to forget whether what I'm looking for was an academic paper or not, therefore I'll know what system to look in.

Similarly, a writer is not going to forget if the thing they are looking for is something they wrote or not. A self-employed person looking for a client/prospective file is not going to forget that they are looking for a client/prospective. And so on. :)

I'm new to GTD. But I've found myself consumed with thoughts of organization in the past few days. I'm about to start a PhD program (in the arts), and I know that organization is crucial. My organization in the past has gotten me by, but I'd like to work out a good filing system that I can stick with for notes and such.

That's why I was excited to see this thread and was happy to heard that I was not alone in my struggles to keep notes organized.

So I agree with everyone that says that you should keep an a-z file for non-academic stuff. I also like the idea of sorting notes on papers and books by the last name of the author (perhaps with a separate file for works in progress), and I too think that an index is crucial. But how to index? That sems to be an important question.

Some people have suggested Biblioscape or Endnote for such an index. I don't have experience with these programs. But I thought today of an indexing system with gmail, and I'd be interested to hear what people think.

How about for every book or article that you want to file away (perhaps done at the end of every day), sending an email to yourself with a "notes" tag (e.g., "+notes") and the last name of the author, the title, and--importantly--keywords that concoct all in the subject. Then in the body, you could put the full citation. In the case of notes that you've typed out (I can't seem to stick to typing out notes only or writing all out longhand, but can anybody? Seriously--does anybody manage to stick to one or the other?), I'd also put the notes into the body of the email. In the case of notes that I've written out, I'd just type out "See paper file". Then you could simply search subjects with the keywords when you want to find notes on certain subjects, and if you want to find particular notes, you can use more specific terms, such as words from the title.

The virtue of this system, so far as I can see, is: 1) easy online access to your files, so portability 2) I'm not crazy about the idea of being tied down to a program that might become defunct at some time or another. An email-based system could outlive gmail (it could work with any email, thoug you'd lose the "+notes" function). Gmail simply provides a good search function. 3) at least for me, I use gmail all the time. So easy access.

The down side seems to be that it takes work. But I'm not sure that it would take that much more work than other filing systems. All of these systems seem to take dedication.

If you use endnote for citations, you'd probably have to enter things both into your email and into endnote (or biblioscape), and so the fewer steps would be a reason to use those programs exclusively. But might this system I propose work similarly well? Stupid idea? Would you all recommend those programs for indexing?

Any thoughts on note-taking?

Dan

My primary allegiance to Endnote has to do with the import and electronic remote connection functions. I almost never enter references into Endnote manually. When I do searches on WorldCat or ATLA or some other database, I use those databases' ability to save records in a format for importing to Endnote. If I am working with a book I pulled off the shelf at the library, I search LOC or some other major library such as Yale by ISBN to get the record info. I like using university libraries so that I also get the LOC call number with it.

The only time I enter records is if I am working with an edited book of essays that I pulled off the shelf in the library. There's no really easy way to search for book chapters as records, so I have to enter those manually. The other time is when there is a book that is so old that it doesn't have an ISBN. I can usually search by author and title and get it, but sometimes I can't and have to enter it manually.

Personally, I don't like the idea of mixing my notes up with other stuff in gmail. I suppose once they were tagged they wouldn't really be mixed up with everything, but I also don't like the idea of being locked into gmail.

If I were going to go the route of keeping notes online I would probably use Backpack. One of the cool things about their service is that you can email things to pages you have on there.

As it is I either type notes or take them by hand depending on my mood, but the end goal is to import those notes into their respective entries in Endnote.

Well, I'm also a PhD wannabe-GTDer... I keep most articles in PDF or PS form on my computers. Most files (not all, which is a problem I need to fix) are named and are filed under a hierarchical path that represents the topic area. I'm toying with the idea of making the topic areas more general since my particular research area covers a number of subtopics.

Right now, I'm planning to purge most of my hard copy papers as I have them in electronic version somewhere. The exceptions being those I had to photocopy, theses, or interlibrary loan papers. I like hardcopies though, because they're easier to read on deadtrees.

I don't use any special software to manage the references. Just a giant bib file which I may split up according to topic area. My main reasoning for this is that I don't want to get tied to any software format. Any suggestions on software that only works with .bib files and doesn't use a proprietary format?