Review of iAnnotateTuesday, May 4th, 2010 by Gene Golovchinsky
Recently I tried an experiment with one of our iPads to read and review a journal submission using the iAnnotate application. For the purposes of review, I had to read and comment on the draft, and then write up the review. I approached the problem much like I would had I been reading on paper, which included highlighting important, controversial or confusing passages, writing comments and reactions in the margins, and flipping around the document. These were all activities we had supported in XLibris, and I was curious how the iPad would stack up.
The short of it is: loading documents: OK; readability: Great; inking: poor; highlighting: poor; text annotations: OK; within-document navigation: so-so; between-document navigation: OK. Overall: good for reading, not good for active reading.
Getting documents onto the iPad for the iAnnotate software to use is not trivial. For reasons unclear to me, it is not possible to browse to a web page with a link to a PDF file and then transfer that PDF file to the device for reading and annotation. In fact, there seems to be no way to save a document onto the device from the browser. Instead, one has to install a server on a networked computer to which the iPad has access. Once installed, the server is used to make certain directories available to iAnnotate app, from which it sucks down PDF files.
The process is a bit tedious but not too bad. There are some synchronization issues between the iAnnotate app and the server. At one point, the iAnnotate application was not seeing the server until I started VPN and restarted the application. I don’t know why connecting to VPN was required, or why the server wasn’t seen without restarting the application. iAnnotate did not see the server when I was at work (and thus at a different IP address). Back at home, I wasn’t able to reconnect to the server without quitting and restarting it.
One other annoying feature was that after the documents are loaded through the “get documents” menu, you cannot open them directly; instead you must tap the “+” button at the top left of the screen to add a new tab, and then navigate to the newly-uploaded documents again.
The first thing I noted that the document is quite readable on the iPad. Granted, it was a typical double-spaced journal article manuscript, but none-the-less, it was very readable even when zoomed out to show the full margins. A two-column conference formatted paper is also legible, although obviously not as nice.
The full-screen mode made it easy to read the document without the widgets getting in the way.
The iAnnotate application supports two kinds of ink: thin free-form pen-like strokes, and thick translucent highlighter marks that correspond to text. I tried to use both in the course of my reading, with so-so results. First, it should be said that the iPad is inherently not designed for using a stylus. Yes, one can use a stylus with it (I used a Pogo Sketch stylus), but because the display is designed for touching with a relatively fat finger, the stylus tip is also relatively fat (compared to tips for tablet PC styluses, for example). The tip is so large that it obscures the point of contact with the screen, making it difficult to write.
In addition to this occlusion problem, the CPU (or perhaps the iAnnotate application) is just not fast enough to track the stylus when it is used to write. One can make broad circles and margin bars which look fine, but any kind of fine manipulation has sufficient lack to make effective writing impossible. Another problem with writing on the screen was that resting the hand on the screen often confused the digitizer, causing spurious marks to appear on the screen.
Highlighting works better in some ways because it is constrained to the bounding boxes of the text; thus when the boxes are available, they get a nice yellow color to them which makes the text stand out nicely. The downside is that if the text bounding boxes are not available for some reason, the highlighter doesn’t work! The document I was reading, for example, for some reason did not have valid bounding boxes for regular text, although italic text was properly characterized. The result was that I could highlight only italicized text! Furthermore, I could not highlight any text that was in figures which were generated as images rather than containing regular text. Overall, therefore, highlighting proved to be something of a dud.
Text annotations were easy to create and straightforward to type on the soft keyboard. They could be displayed in a floating dialog box over the text, or reduced to a small icon. Unfortunately, the floating dialog obscured the document, and the icon required tapping to show the text. I would prefer an option where the text, if there isn’t too much of it, is shown overlaid on the document using a translucent post-it or something like that. The idea is that the text and the document should be readable at a glance without requiring any interaction.
Scrolling within the document was fine from page to page, although I would prefer for motion to be constrained to the vertical dimension when scrolling between pages. It was a bit disorienting to have the page move from side to side at the same time.
Navigating via annotations was not as good. Annotations were available through a sub-menu of the main menu, but the list of annotations showed the text of textual annotations (which was good) but only the word “annotation” for ink. Thus it was impossible to tell which menu item linked to which passage (although the page number was occasionally useful). I really missed the notebook view in XLibris for reviewing what I had read and marked up.
Each document was kept in a separate tab, making it easy to switch between documents, although no more than six tabs could be opened at any time. I don’t know if this is a major problem, however. It certainly didn’t come up in my reading.
The iPad is a nice device for reading, but the iAnnotate application is not ready for prime time with respect to annotation for the purposes of making sense of and navigating through documents. It’s not clear whether the limitations are inherent in the design of the hardware, or whether some clever optimizations for creating ink will mitigate some of the shortcomings described above. In any case, reliance on a stylus when the device wasn’t truly designed for it will limit overall ease of use.