[UPDATE: Sente is currently undergoing major changes in their sync engine; however, they have yet to update their iPad software. Once the iOS version of Sente is updated I will write a new post about the changes. Till then, please be aware that this post is out of date.]
Last December I wrote a post, Reading Fast, Reading Slow, which covered the various tools I use in my digital workflow depending on the kind of reading I’m doing. Today I want to update that with an in-depth look at what I had referred to as “slow” reading, focusing especially on texts which I have available in PDF format. This workflow assumes you have an Apple desktop computer, an iPad and the following software: Sente for OS X, Sente for iOS, Goodreader for iOS, a Dropbox account and an Evernote account. This is not a review of any of these tools, although the strengths and limitations of Sente are discussed in terms of how they help or hinder this specific workflow. I don’t by any means consider this to be an ideal workflow, but after having experimented and researched numerous options based on the tools which are currently available, this is the one that works best for me.
As I’ve explained before, it would be best if one could search and add PDFs to Sente directly from the system’s default browser, as one can do with Zotero or Mendeley, but despite this limitation, I still find Sente to be the best software out there for organising one’s citations. Zotero, for instance, lacks the “status labels” feature of Sente which is so central to the workflow I describe below. Moreover, for this workflow to work, you just need to download the PDF itself from your browser, and Sente will take care of the rest. And the iPad apps currently available for Zotero and Mendeley are sorely lacking compared with what Sente offers. (Other options are Papers and Bookends, but I find Sente compares favourably to those as well.)
To start, I’m going to download a file from everyone’s favourite Open Access anthropology journal: HAU. The lead article in the new issue is “The birth of the word: Language, force, and Mapuche ritual authority” by Magnus Course. The first thing I need to do is download the PDF to my browser’s download’s folder. This gives me a file called “19-519-1-PB.pdf” which is really annoying because I have no idea what that means. I’m sure it means something to the editors of HAU, but sitting there on my desktop it is pure gobbledygook. No worries, however, because Sente will help me rename the file to something useful. All I need to do is drag and drop the file into my “reading” library in Sente.
This then brings down the following menu:
Now, if HAU had DOI names in the PDF file like many journals do, Sente would pick that up and automatically fill out all the citation data, but no worries. HAU information is properly indexed by Google Scholar so we can get the data by telling Sente to search Google Scholar by the article name. Just select the article title and tell it to search Google Scholar for the selected text:
You will then see the following search results, with a bullseye symbol next to each result. Just click the appropriate result and the metadata will be applied to the PDF which will then be renamed and filed according to your settings (more on that in a minute):
The file is now named “Course 2012 The birth of the word Language, force.pdf” and stored in a Dropbox folder called “reading.” (Be patient, I’ll get to the settings for this in a bit…) But before I do anything else, I want to assign the file with a “status” so I remember what I want to do with it. Apart from the ability to pull the metadata, rename the file, and save it to the appropriate folder, the ability to assign a “status” to each of my files is the main reason I decided to go back to using Sente. Let’s say this article is very important for my research, then I’d want to assign the status “Must Read”:
My other status choices are: Finished, Reading, Research (meaning I still need the PDF or I’m missing metadata), Someday, Skip (like “finished” but means I didn’t bother to read it). [You’ll see “lookup” in the picture. That’s what I used to call “Research” but I merged two different Sente databases and haven’t yet finished sorting all the imported references.]
The other thing I want to do is assign a “QuickTag” to the file, which allows me to know the general subject area. These are different from Keywords in that they automatically create QuickTag “smart folders” in the Sente menu. I might assign the QuickTags “language” and “ritual” as well. When I’m doing a research project or writing a paper I also create QuickTags for that paper or project which is very useful for quickly seeing what needs to be read next according to what you are working on at the moment.
The main annoyance with the workflow presented here is that I now need to sync two different databases with my iPad. It would be easier to simply keep my files stored inside the Sente database and just use Sente – but for several reasons that doesn’t work for me. For one thing, Sente won’t sync PDFs which are larger than 20MB and I have a lot of these. I have lots of articles and even books I’ve scanned and OCRd on my desktop which are larger than 20MB. Since Sente won’t handle these, I need to use Dropbox for them. Since, for this reason, I have to use Dropbox anyway—I might as well use Dropbox to sync everything else as well. Sente can be setup to store your attachments in an external folder and while Sente databases can’t be saved in Dropbox (the developer says this will corrupt the database), the attachments can be saved in Dropbox without any problem.
The second reason to set up Sente in this way is that I don’t use Sente to read and markup PDFs. Sente does have the ability to do this, but IMHO Sente is far inferior to GoodReader in this regard. Since I use GoodReader anyway, and GoodReader can sync with Dropbox, it is just simpler to have all my PDFs stored there and just use Sente to sort, rename, categorise and label my PDFs. Syncing two databases isn’t really that bad once you have everything setup, but the initial setup requires some explaining, hence this blog post.
Let’s first look at how to set things up in Sente. The attachments settings are set when you first create a new library, but they can be accessed anytime from the “Library Setup” tab on the sidebar:
Here are my settings:
I have Sente file and rename the reference and tell it to use the top-level of my Dropbox folder. That is because Sente will automatically create a folder with the name of the library “Reading” and put the files there. If you want you can further group the PDFs into subfolders. I’ve chosen to group them by “Publication Type” since I don’t want too many subfolders, but you can try other ways of organising your library if you like. I’ve also set up the files to be named with the author’s name, the year of publication and the article (or volume) title.
Unfortunately, Sente does not seem to have an option to clean up and re-file all your PDFs if you change these settings (or if you edit the metadata on a PDF). Other apps, like iTunes can do this and it makes things much easier. Sente really wants you to store your PDFs inside it’s bundle and not worry about how they are organised.
Before you can sync this library with your iPad, you also need to setup a synced library from within Sente. I find the need for this extra step quite annoying, but you only need to do it once. I won’t bother to explain how to do this here, referring you instead to Sente’s own documentation. You will want an “unrestricted” sync library, and you shouldn’t sync attachments on the iPad unless they are requested on demand. (I’m not sure, but I couldn’t figure out any way to turn off attachment sync altogether, so I still get error messages every time I add a file which is larger than 20 MB. But these errors can be ignored.)
The next thing you need to do is to set up GoodReader to sync your “Reader” folder in Dropbox. This is fairly easy, especially after you have set it up the first time, and the instructions for how to do so can be found in the GoodReader documentation on synchronisation.
Once this is all setup, you just have to launch the “Reading” database in Sente on the iPad and it will sync all your data no matter where you are. This is much better than some of the other options out there which require you to be on the same WiFi network as your desktop computer in order for sync to work. Sente’s system is truly cloud-based and (after initial setup) can be done anywhere. But it is not a good idea to have the same database open on both your desktop and the iPad. I’ve found it to lead to frequent crashes on the iPad if I change status labels while the desktop database is open at the same time. Hopefully this will be fixed in future updates. Another annoyance with Sente is that, unlike other apps, sync cannot work in the background. You have to keep the app open till sync is finished. However, since you aren’t downloading attachments, this happens fairly quickly (although I think it could be quicker).
Fortunately, GoodReader’s DropBox sync does work in the background. This is good because the files can be large and you don’t want to be sitting there watching them download. But you can’t use GoodReader while it is syncing, so if you recently added a large file or a lot of PDFs, you might want to pour yourself a cup of coffee while it is doing this. One nice thing about this setup is that, since the attachments in Sente are stored in the Dropbox folder, when you make annotations to your PDF files in GoodReader they will show up in Sente on your desktop as well. This happens without having to do any extra step!
One thing I really like about GoodReader is the ability to export just my highlights (“Email Summary” in the sharing menu). So if I go through the article and highlight several passages on my iPad, I can then export just those passages via email. Evernote gives you a private email address which you can use to export data to Evernote, so what I do is email my highlights from GoodReader to Evernote which I then use to search and organise my annotations.
Once I’m done with the article I will go back to Sente and set the status label to “finished” assign a star rating according to how useful the article is for my work (different from rating it according to how “good” it is) and make a short note about what I found useful in the article. This creates a reading journal of sorts, although not the social kind that Rex wanted.
This isn’t a perfect system, but I’ve been using it for a couple of weeks now and it works quite well for me. Previously I had just been storing downloaded PDF files in DropBox, but eventually it got overwhelming. I didn’t know what half the files were because the names were gobbledygook, I didn’t know what was important and what was just something I wanted to take a quick look at and I didn’t have any way of grouping PDFs according to project. Sente lets me not only see what is a priority for me to read next, but to see this according to which project I happen to be working on at the time. At the same time, this system allows me to sync large PDFs (which Sente can’t do) and to work directly in GoodReader without having to first import and download the files in Sente (which was much slower than GoodReader’s DropBox sync). This system also lets me look back at what I’ve read in Sente. When I write a paper for publication I create a Sente library just for that paper. This system also lets me look back at what I’ve read recently and easily pull in the relevant references for that paper. (Sente allows you to easily drag and drop citations between libraries.) In short, I believe that for digital nomads who want to read academic PDFs on their iPad, this is your best bet right now, although I’m always hopeful that something better will come along in a few months.
UPDATE: Specified the command in GoodReader which lets you share just your highlights: “Email Summary”
UPDATE: So it seems that you can get Sente to re-organize (rename) the attached files if you change the settings for attachment handling. Although there isn’t a re-organize command one could conceivably make a small change in the preferences and get it to update the file names that way.