For the academic side of things I would recommend Bookends X. This is the most mac-savvy database program for managing your database of books and articles. And if most of your PDF's are articles, etc. you can add these as attachments with the attach file button on each entry. That's the way I would go with your academic stuff.

http://www.sonnysoftware.com/

For general digital reference I think you're just going to have to accept that you're going to need to do more data entry on the front end, but not as much as you think. The Finder-like programs like DevonThink, Yojimbo, Journler, etc. provide for ways of linking data that is already in the Finder. Why not just keep things organized in the Finder better? Some batch renaming application or applescript would probably suffice for this.

I designed the Ready-Set-Do! GTD implementation for the mac, so I use that for keeping all of my digital stuff organized. I also happen to be an academic myself currently finishing up my dissertation. Good luck with your further doctoral studies.

Would relying on the PDF's metadata solve this problem? You could name the file whatever you like and assign it a title, author, keywords, and so on. Then, in your directory/folder view, you can sort however you want. Desktop search software will be able to read the metadata also. Having a program manage this info for you seems like too much overhead, IMO.

If you can get away with text search and keywords, I'd at least consider a web-based CMS. I keep my lab's help documentation, my lecture slides and supplementary material, etc. in a Plone site. With some tweaking, you can get it to do full text indexing of PDFs, Word documents, Excel sheets, Powerpoint presentations, RTF files, etc.

I'm in the middle of changing addresses for the server right now, so the following link might point to the old server with broken text indexing. If so, give it a day or two, and it'll work. But if you hit http://www.cae.tntech.edu/ and put something into the search box at the top right, you'll see what I'm talking about. e.g., if you enter 'tolerance', you'll get a list of two PDFs from an old class, and you'll see that the word doesn't show up anywhere in the title, summary, or anywhere other than the actual text in the PDF.

You can also define a keyword taxonomy for tagging any files you upload. And it'll let you upload files over WebDAV. So the methodology would be to upload your PDFs to a Plone server that you can access from wherever you need to, tag them if you want, and let the server do all the heavy lifting on searches.

I've read some really positive reviews online about Papers.

I'm sort of keen on getting some sort of open source application, as I don't really want to put a lot of time putting thing into some sort of database framework and then have it be obsolete and hard to extract in a few years when some better solution comes around, or some such. As I already have DevonThink and rather like it, I'll probably continue to use it in at least some capacity, (as a back up or index most likely) but I'm not sure that replacing it with a similar sort of organization would fix any of my issues.

I sort of tow the line between social sciences and humanities work (a little bit stronger bent toward the former, but still) with the end result being that I find myself a bit out of place with a lot of the tools that exist for one group or the other.

I just (re)discovered Bibdesk, which I think seems like a pretty darn good kind of program for what I'm looking to do. I think I can add a filing number to the program and then make my magic with the sequentially numbered files, unless someone has a naming convention that I haven't thought of.... Any good naming conventions that people are using would be a great help.

Thanks everyone for your suggestions. I'm interested in hearing more about how you all manage your collections (with stuff like naming conventions, or filing conventions) with or without programs like DevonThink or Eagle Filer and so forth. For instance, do you file archives away by journal, or by publication date, or by author, or what, and then how do you name the files. I'm leaning at the moment towards BibDesk, but I think I still have a lot to work out with regards to actual usage ands organization.

Thanks again.

Cheers,
sam

Don't understand any advantage to rename everything. Surely most docs have some kind of meaningful name currently.

Have you looked at Papers?

well. some things have very meaningful names and some don't al all. The ones with meaningful names are pretty inconsistently named some have all the authors in the file names, some have only the first author, some have authors last names some have authors last names and initials... and so forth.

I'm thinking that I might rename the files when/if I enter them into bibdesk, such that the file name will be the same as the cite-key that I'd be using in BibTeX/LaTeX....

You could email the documents to Gmail and put as many descriptors as you might need in the email subject line and body of the email.

Sample subject line: Date of article | Author of article | Book or journal name | Primary Subject (or section of your thesis |

Then, once you’ve emailed the .pdf to yourself you can tag the email. Tags could include: Archive, Reference, Active Project, Thesis, etc. (This would also serve as an offsite backup of your materials.) I think the jury is still out on whether to trust free online tools such as this with valuable material; it obviously shouldn’t be the only place the material is stored.