An Image-Based Document Reader with Editing Functions for Education and Research on Digital Humanities.

This paper focuses on an electronic document reader
we have developed, a new method of lecture using
this reader, and some possible examples of its utility for
research on digital humanities.
Introduction
Recently, more and more documents have been stored
in digital archives. Electronic documents have many advantages,
e.g., high performance when searching text,
instantaneous document transmission, and negating the
need for archive space of the corresponding physical
documents. Because of this, we can read thousands of
digitized books anywhere with our notebook PCs.
Despite these benefits, however, it is often said that most
people find it difficult to remember the contents of digital
documents when reading them on a computer display,
compared with reading the physical paper document. We
can propose several reasons for this. Firstly, any physical
movements which indirectly assist our recollection, such
as eye-movement or page turning, are not experienced
when reading digital documents. Secondly, making
notes or memos or marking on the electronic document
is usually not possible, and even if it is, we cannot do it
as easily as on paper.
Some reports say that taking memos or highlighting
parts of documents improves our understanding and remembering.
Since such activity requires proactive decisions
on selecting contexts, these actions activate our
brain much more than silent reading. Of course, these
markings and memos help us recollect our thoughts during
a later reading of the document.
To overcome these shortcomings, we have developed a
document reader that enables us to freely mark digitized
documents. The reader we have developed has several
marking functions, such as, highlighting areas of text, inserting
bookmark stamps, and adding memos to the text.
One of the most unique features of our reader is to realize
these mark-up functions with usability almost equal
to that of paper media. This point is important because
such an easy-to-use text reader has not prevailed yet.
Our reader stores mark-up data in a file separate to the
original document. As the mark-up data is written in a
simple text-numeric format, it can be easily used by a lot
of software applications, such as spreadsheets, statistics
analysis or image processing packages.
In this paper, we will start with an explanation of our
reader’s functions and then introduce two applications
using it: a new method of lecturing using the reader, and
examples of using mark-up data for the study of digital
humanities.
Functions Of Our Reader
We now introduce editing functions equipped in our
document-reader. An example text containing mark-up
highlights, stamps, and memos written on the text is
shown in Figure 1.
Figure 1: A sample document with markings
Highlights:
We can highlight text just by click-and-dragging the
mouse along the text. Any other operations like opening
menus or clicking icons are not required. Dragging
from left-to-right highlights the text in red, while, dragging
from right-to-left highlights in blue. You may think
that highlighting in blue is a little bit difficult due to the
right-to-left movement, but in exchange for this inconvenience,
you do not have to choose a separate color to
use. That is, blue is naturally used for short key words or
important contexts.
Stamps:
If you push the space key, a stamp appears on the text.
The stamp has two faces: “?” and “!”. The intended purpose
of the stamp mark “?” is to mark hard-to-understand
concepts, or points that need to be followed up later. The
“!” mark is used for “interesting” or essential points of
concepts.
Stamps also act as bookmarks. If you push the “<” or
“>” key, the reader jumps to the previous or next stamp,
respectively. If there is no following stamp in the current
document, the reader jumps to the next bookmark in the following document.
Text memos:
You can add notes or memos anywhere in the document.
Push the “Tab” key to enter text-input mode. Characters
you type on your keyboard will appear at the position
of the mouse cursor. To exit input mode, push the “Tab”
key again. Unlike the other editing applications, our
reader does not show any text-input-box which spoils
the look and feel of writing on the paper.
User’s mark-up data file format:
A user’s mark-up data is stored in an associated file. An
example of a user’s mark-up data is shown in Figure 2.
You can see that it is quite a simple form of text-numeric
data. The number in the top line indicates the position
of the cursor when closing the document on the previous
reading. Each of the following lines describes the
marker type and the coordinates of the surrounding rectangle.
In the case of a text-memo, the user’s text is added
as a character string after the coordinate values. These
data are used not only for displaying mark-ups, but for
other various purposes, such as taking mark-up statistics,
extracting highlighted-regions from the document, and
comparing comments written by different reviewers. We
will now show some examples below.
Figure 2: Mark-up data of Figure 1
Use During Lectures
Thanks to the improvement of our computer literacy and
network technology, computer-assisted education, (so
called e-learning), has become more and more popular.
Our university has several computer rooms, but they are
mainly used for learning computer software or programming.
So, we have been developing a new style of lecturing
using our reader to make the most use of computer
rooms for the classes of humanities. The arrangement of
a typical class is as follows.
First, the students read the lecture notes for the next class
with our reader, highlighting or putting stamps on phrases
they cannot understand clearly.
Next, the teacher receives student mark-up data via the
network prior to the lecture. The lecturer checks the student
reports with a summarized view (shown in Figure
3), and then, he examines which areas of the lecture students
are having a hard time understanding. As a result,
the teacher can modify his lecture so that it is more suitable
and understandable for the students.
There are several other benefits for the students, too;
1) They can understand lessons more deeply by paying
attention to areas they could not clearly understand. 2)
Their sense of participation is promoted because their
mark-up results directly affect the lecture content or explanation.

These major benefits are achieved by relatively simple
editing actions by the students. According to a class
evaluation taken at the end of the semester, most of our
students approved of this style of lecture. By investigating
the mark-up data gathered from them, we can extract
difficult-to-understand keywords and sections of
the lecture. In the near future, it may be possible to find
relations between their marking style and their learning
results.
Figure 3: Summarized views of student marks.
Using For The Study Of Digital Humanities
A particular feature that paper media have is that we can
easily write on them, however, the reality is that, it is
usually prohibited to write on research materials. By using
our reader, we are completely free from such restrictions.
This is one of the benefits of our writable reader
used as a researching tool.
In addition, by displaying marks or comments written
by more than one researcher on the same document, we
can clearly recognize which points should be focused on,
and what the differences are between their ideas. This
function is also useful for referee reading or proofreading
performed by several reviewers.
Finally, since each piece of mark-up data indicates its position
in the material, it is easy to extract marked regions
from the original material. After gathering these regions,
they can be shown as listings or KWIC (keyword in context)
forms. We can also use the marked region as a template
for searching over the document. In this way, this
reader can be used as a front-end to researching tools.
Conclusions
In this paper, we have confirmed the utility values of our
digital document reader for understanding computerized
documents and its usefulness in education. As mentioned above, our reader can be used in several ways for the
research of digital humanities. Especially, by handling
materials such as image documents, our reader is suitable
for handwritten documents and those with a large
number of figures.
You can download our text reader from our website. We
would like you to try it and give us feedback. Please feel
free to stop by our poster session, as we are going to
demonstrate it and introduce new methods for research
on digital humanities.
URL: http://www.img.is.ritsumei.ac.jp/~h-seki/beeReader/