How should I title my scientific pdf documents?

Here’s a question for those of you who use DT to store scientific papers/articles.

I’m working on my bach. thesis at the moment, and so far I’ve been using the article title + author + year as a naming convention for my titles. However, I’m starting to notice that:
a) the title is hardly ever visible anyways, because it’s too long.
b) I tend to have author(s)+date in my head most of the time anyways, rather than the title.

However, for later reference, it might be more useful to actually use the article title as the title of my DT entry. If I click on ‘See Also’, I would get a list of (partially visible) titles, rather than just a list of names. On the other hand, I would probably have to click on the documents anyways to expose the full title. And if I understand correctly, the title has no effect whatsoever on the AI.

So my question is: how do you guys do it? Do you just use ‘author(s)+year’ to title your DT entries? Do you use the full article titles? Or maybe a custom title? And if so, why?

Personally, I favor a title that’s descriptive of the content, conveys the subject of the paper.

Although citation styles favor attribution beginning with the principal author,it’s not uncommon for multiple authors to share equal claim to the content. And it’s not uncommon for a junior author to have been the actual contributor of novel ideas. Sometimes the senior author is mentoring the career of a newcomer, lending his name to increase the potential of a grant or publication. Unless I’m familiar with the field and with the people doing the work, the order in which authors are listed may be uninformative.

There are landmark papers in which the second, third or fourth author in the citation was the important one. :slight_smile:

This is a good question.

I generally go with the full title of the article itself, but I will truncate or adjust if it is a title which does not fully capture the article’s subject matter or content. I do this to help DTPO’s search feature, actually, and I find that it works well. I find that DTPO’s search feature (either Phrase or Title) is all I really need to pinpoint an article in my database. In fact, I’ve been able to locate an article based purely on a concept that I recalled reading. It took a bit of trial and error with the search function (and I can’t wait for the power of search from DevonAGENT to come to DTPO) but I got there in the end.

On the topic, I was faced years ago with a choice: do I collect and store pdfs of journal articles by topic or by subject/topic or by the journal name. I rationalised that, more often than not, the title of journal will tell me more or less about the wider topic (but obviously not always the specific topic), thus I opted to have every pdf reside in a folder bearing the name of journal from which it came. What helped this decision was, at the time, my use of Copernic Desktop Search, which would find anything on a Windows hard disk. These days, DTPO fills this gap nicely (and more elegantly, I would add), so in many ways the actual filing of articles isn’t always necessary.

I keep journal article pdfs in one massive ‘journal articles’ folder on my hard drive, split obviously into numerous sub-folders with the names of journals. The top level ‘journal articles’ folder is then synchronized with DTPO. In other words, I do not import pdfs of journal articles into DTPO, but rather use it as a storage tool and one with which I can search.

I’d be interested in hearing how others manage this.



I’m eager to know a little more about your PDF filing system, as I’m struggling to organize mine. I like the idea of having a central spot for my PDFs, as they enter from all angles, via Papers, Safari downloads, etc. I assume your subfolders are smart folders, yes? (That’d keep the Massive Folder intact, and avoid duplication.) And if you’re constantly funneling new PDFs into the Massive Folder, how did you set up DevonThink to automatically add the latest additions? Or do you add them one by one yourself? And do you annotate and take notes on your PDFs, via Skim, or in DT, or some other method? All insights warmly welcomed …

Hi there,

I could have be more clear - apologies :slight_smile:

Basically, DTPO is not technically involved in the filing and organisation of my PDFs but it is directly involved in searching them. I make use of the ‘synchronization’ feature. Let me explain.

Completely outside of DTPO, all of my work is contained under one single MAIN folder, within which I have multiple other folders, such as WORK, TEACHING, ADMIN, CONFERENCES, PUBLICATIONS, LECTURES (RECORDINGS), RESEARCH and ADVISING. Again, this is on my hard drive, not in DTPO. Within the RESEARCH folder, I have a folder called ‘JOURNAL ARTICLES’ and naturally quite few others (e.g., broad topics of central research interests). That JOURNAL ARTICLES folder holds all my journal articles, organised by journal title. It’s just under 800MB in size and has 1750 items, so it isn’t that small. When I download a pdf, it gets put into the appropriate journal title folder within the JOURNAL ARTICLES folder on my hard drive, not in DTPO.

But here’s where DTPO comes in.

Within DTPO I have a folder off the root called ‘MAIN’ which I use to synchronise with the MAIN folder I mentioned above. I do this by highlighting that MAIN folder in DTPO and then using the keyboard shortcut OPTION+COMMAND+S (also in the File menu) to synchronize the contents. As I understand it, this doesn’t actually import the content of the MAIN folder on my hard drive (I wouldn’t want it to), but rather indexes it. Someone do correct me if I’m wrong on this. Anyway, I do this perhaps once a day, usually in the morning, or even during the day if I have added a bunch of new journal articles or other content to the MAIN folder on my hard drive.

This way, everything that is in my MAIN folder on my hard drive is actually accessible via DTPO but not held by DTPO. And of course this means that it is fully searchable, including the contents of those 1750 journal articles. As I am writing, I can search within that JOURNAL ARTICLES folder because, as you know, you can specify where you want DTPO to search.

It’s a system that works really well for me, but would love to hear examples from others.



I’m afraid I don’t have much in the way of a contribution (yet!) but wanted to say that I’ve learned a lot by reading through this series of replies. I recently stumbled upon DTPO and have now fallen in love with it as I try, somewhat desperately, to organize my articles, snippets, trains of thoughts, and formal writing projects under a single roof to maintain my sanity as my dissertation project gets under way.

I’ve already got a giant folder of articles that is actually organized by having the lead author, year, and title as the file name. This just happened to work for me as I found I was finding relevant articles in journals that I never would have associated with the topic. Hrmpf. When I started playing with DTPO it occurred to me that the index versus import function was going to be a little problematic if I wanted to edit my files. Most of them are .pdf and I like the highlighting features and such available through Acrobat, although I’ve started using Skim recently.

I’ve actually gone for a rather odd approach. While I’m wandering around collecting article, I have my giant articles folder which holds everything that I that was interesting at the time that I bothered to download/capture/copy it. I keep this folder indexed to a separate DTPO database and both the database and the indexed folder sits on my portable external drive (so that I can always travel with it, but can also unburden my laptop’s hard drive of gigs of articles).

However, for articles that I need to annotate (a smaller subset of the articles folder), I actually import these items into my DTPO dissertation database, and then I make sure to edit the PDFs by opening up the relevant DTPO package file and getting at it that way. The reason for this was twofold. The first is that I needed a way of quickly seeing what I had already annotated versus what I had not. The second is that I wanted my dissertation database to be contained so that it was easy to backup and easy to move from one computer to another. For instance, between my laptop and the desktop in the lab where I actually do my work. I should say that even if I don’t annotate an article, but need to either mark that I had read it and found it unhelpful, or had read it but just had some separate notes for it, those also get imported into the dissertation database.

So for me, setting up one database with my core work materials as a self-enclosed entity while having an indexed larger corpus of works to choose from looks like it will pan out for me. I’m still in the early stages, so 500 megs into this little experiment I may change my mind! But for now, it is working. And it means that if I re-open an article that I annotated, say, back when I was an undergrad and realize that all of my notes are totally worthless now, I just import in a fresh copy of the file from my giant articles folder. :slight_smile:

I’m very interested in how others are dealing with these issues.

Author+Keyword works well for me, but since the set-name-as hotkey works like a charm (again), I just use the title or a keyword from it. I have stopped to care that much, since entering the perfect title takes much more time than clicking the name in the search results. I had to find out that there is an inverse correlation between a perfect database structure and my output, measured in articles, at least for me. So I concentrate on adding interesting content, sometimes linking it; thinking about titles is for rainy days.

Comment: to annotate a PDF you’ve imported into your database, simply select it, Control-click and choose the contextual menu option Open With. Choose the application to use for annotation (Acrobat, Preview, Skim) and the PDF will open under that application. Save the changes. In your database, select the PDF again and invoke File > Synchronize. That’s it!

Bill, thanks for that! That is much easier and would have taken me a while longer before I found it!

My filing of pdfs is very simple. A title loks like:
Einstein 06 J Phys Chem-keyword1 keyword2 etc.
i.e. first author year journal and key words
It has worked great for several years with thousands of refs and is shared by my research group an many others.
Good luck!

My filing of pdfs is very simple. A title looks like:
Einstein 06 J Phys Chem-keyword1 keyword2 etc.
i.e. first author year journal and key words
It has worked great for several years with thousands of refs and is shared by my research group an many others.
Good luck!