Handling PDF Documents
This page describes my strategy for handling PDF documents. The page is fairly Mac OS X specific. In addition, I use Sente to manage my collection of PDF documents, and much of what is discussed here is of interest only to users of that fine Mac-only application, but other sections of this page will be of more general interest. I hope that some or all of it will be useful to you, or that it may help you to refine your own strategy.
The following topics are covered on this page:
- Searching for manuscripts in PubMed using Sente.
- Downloading PDF documents
- Renaming newly downloaded PDF documents
- Moving PDF documents to appropriate folders
- Dropping PDF documents onto records in Sente
- Moving records to a master collection in Sente
- Retrieving PDF documents in Sente
- Summary
Searching
Of course, for many the simplest way to search PubMed is to use the web interface provided by NCBI. However, for some time now, I have tended to use first EndNote and now Sente as my primary searching tool, for two simple reasons: (1) each program simplifies construction of complex Boolean searches and helps you narrow searches to specific fields (e.g. Authors, publication date ranges, etc.); and (2) when using EndNote and Sente, I automatically have a record for each reference that I can keep in my master database.
Oh, there is one further advantage to Sente. It is lightening fast, even when your search yields hundreds of records. Of course, you still have to scan them, but Sente can even help you construct filters to narrow the scope of what is displayed.
A full discussion of how to use Sente to search PubMed can be found in its manual, though you can probably get by quite nicely without reading it. There are only two things to note here:
- Sente is automatically set up to take you to PubMed "LinkOut" pages, which contain links to the article of interest, stored on the journals website. This is a simple two step process.
- Starting in the last several years, most articles have unique DOI codes (Digital Object Identifiers), and for recent articles, PubMed records include the DOI. One of the members of the Sente User Forum posted a very useful suggestion for modifying Sente's preferences to allow it to directly link to an article in a one step process. I strongly suggest that you implement this suggestion.
Finally I have set up Sente to automatically collect eToc-like information, and I suggest that you read my page eToc Replacement Therapy with Sente for more details. My process of downloading and managing PDF documents more often than not starts here.
Downloading PDF documents
Even my mother no longer needs to be told how to download a PDF document. I have only one suggestion here. If you're not already doing so, for God's sake make yourself a dedicated download folder and set your browser preferences to automatically store newly downloaded documents there. Don't let them litter your desktop.
Oh, there are two more things. I happen to dislike browser plugins that display PDF documents in a browser window. I can understand the attraction for some people, but it doesn't work for me. As a consequence, the approach that I describe might be a little different than what you encounter if you display PDF docs in your browser window.
I have my computer set up to automatically open newly downloaded PDF documents in Adobe Acrobat (though Acrobat Reader will work just fine). The starting point for the next step is an open PDF document in Acrobat.
Renaming newly downloaded PDF documents
I've never understood why this is the case, but the names for PDF documents assigned by the journals are decidedly unhelpful, and as a consequence, they simply have to be renamed. I have yet to encounter the perfect naming convention (which is perhaps why the journals don't even attempt to bother), but my favored solution can be found on my page PDF File Naming Conventions.
There are a couple of additional useful tricks that speed up my workflow that are worthy of comment.
Two options for how to rename a PDF file.
First, you could invoke the "Save As" command from within Acrobat, and use the "Save As" dialog box to navigate to the folder on your disk where you'd like to store your file. I used to do this all the time, but my computer had an intermittent bug whereby Acrobat would often crash when performing a "Save As". Since I could never figure this out, I developed a second alternative, with which I am now quite comfortable.
I navigate to the PDF file in the Finder, and change it's name there. But how do you get to the file most quickly? Of course, you could enter the Finder, navigate to your downloads folder (which you've put high up in the Sidebar on your Finder windows in 10.4, right?). But, there is a quicker and more powerful way. In most OS 10.4 applications, if you hold down the command key and mouse click (and hold) on the document name in the title bar of the document window, the directory hierarchy corresponding to that file will be displayed. If you drag down to the first folder in the list (the one that contains the file of interest), you will automatically be transferred to the Finder window corresponding to that folder, and the file of interest will already be highlighted and ready for you to change it's name. Try it. You'll like it.

Most people—but not everyone—will know how to change a filename
in the Finder. The way to do it is to click on the file name ONCE, wait a
few seconds, and click again. Don't click the second time
too fast, or you will open the document (which, of course, is already open).
You are now ready to type away.
But what do you type? If you're following my naming conventions, you'll need access to the last name of the first author, the journal name, volume, year, and page number. You could try to remember these, or jot them down on a notepad or a PostIt note with a pen or pencil, but this is a pain. There is an easier alternative. All it requires is that you arrange to have two windows on the screen at the same time; the Finder window displaying the contents of the folder where your PDF file resides, and the Sente window. The windows should be arranged so that you can see the both the PDF file, and the Sente record, with the requisite information displayed for naming the file, as in the figure below.
In this case, I have renamed a file in the "•Internet Downloads" directory as "Kublin.2006.JID.193.1.pdf" based upon the first record in the Sente window. You'll develop your own strategies for arranging the windows, but it's not hard and doesn't take much time.
Moving PDF documents to appropriate folders
The next thing I do is move the new PDF file (in this case "Kublin.2006.JID.193.1.pdf") to an appropriate folder. There is no one right or optimal way to do this, and the system I use is discussed on my page PDF File Naming Conventions. All I do is open a new Finder window, navigate to the subfolder where I want to store the new PDF file, and drag it there. In this case, I'm going to store this file in the folder ~/Documents/PDF Files/ PDF by Subject/H.subject.PDF/HIV pathogenesis. I could have stored it somewhere else. Don't agonize over where to store it, because you're going to create a link to it in Sente that will make it easy to find.

Dropping PDF documents onto records in Sente
The next thing that I do is click-hold on the PDF file in the Finder window and drag it onto the Sente record, creating a link to the file location in the Sente record. This adds a blue dot to the Sente record, as seen in the figure below.
Again, it's possible that you might have to move some windows around so that the relevant windows are visible, but it doesn't take long. They don't need to be arranged exactly as in the figure above -- I just did that to keep the screen shot to a reasonable size.
Moving PDF documents to appropriate folders
The Sente record corresponding to the document I have just downloaded is in a collection labeled "003 eTocs". In my Sente system, this is actually a hierarchical collection of eToc-like searches. Since I delete records from these searches all the time, I want to move the record to a permanent master collection, in this case "002 JDA Master Collection". All I have to do is drag the record from one collection to another. It's very simple, and now it's always there and very easy to find in Sente, simply by selecting the master collection and typing a few keywords into the search bar.
Retrieving PDF documents in Sente
If you want to look at the PDF document in the future, simply search for it as above, and double click on the blue bullet.
Summary
This page takes a lot longer to read than it will take you to implement this strategy once you become comfortable with it. Until someone develops a better automated system for organizing PDF documents, I find that this strategy is rapid and gives me the ability to rapidly retrieve articles whenever I want.