Interactive Lessons for Tablet-based Teaching
A Proposal for an Open Data Format
Heiko Weible
1
and Edgar Seemann
2
1
Visual Generation, Berlin, Germany
2
Furtwangen University, Furtwangen im Schwarzwald, Germany
Keywords:
Tablets, Schools, Teaching, Interactive, Android, iPad, Multimedia, Math, Foreign Languages.
Abstract:
Tablet devices like the Apple iPad are ideal for interactive learning content. Unfortunately, at the moment it is
still quite difficult for authors to create interactive content. There are very few standards in the area of tablet
computing, in particular, if authors do not want to restrict themselves to a single platform e.g. iOS or Android.
In this paper we propose an open data format, which makes it easy for authors to create interactive content.
We achieve this by separating content description from the visual appearance and behaviour. Thus, allowing
to create interactive content without the need for complex tools or even programming skills. The data format
may be implemented for any of the leading platforms. Thus, giving content creators the control over the usage
and the distribution of their content.
1 INTRODUCTION
Computing devices are becoming more and more
popular in education. While previous efforts like the
One Laptop per Child project by Negroponte et al.
(Negroponte, 2006) have seen some success, there are
now two trends which are about to change our educa-
tion systems and our way of doing education.
These trends are, on the one hand, the field of on-
line learning and on the other hand tablet-based teach-
ing with devices like the Apple iPad or comparable
Android tablet devices. Both fields are very new, but
exciting progress has been made in the past year.
In the field of online learning, we have seen the
first courses with tenths or thousands of participants,
something which was unimaginable a decade ago.
The Stanford university with their professors Thrun
and Norvig have been pioneering in this domain with
their artificial intelligence class (Thrun and Norvig,
2011). Since then we have seen the appearance of on-
line universities like Udacity (Thrun et al., 2012) and
Coursera (Ng and Koller, 2012).
Tablet devices have only been available for less
than two years. Yet we see a large number of deploy-
ments at schools in North America and Europe, for
example (Farias, 2011; Klinga, 2011). Touch-based
interaction is very well suited for educational content
and both teachers and students have been quite enthu-
siastic about the new possibilities.
Figure 1: Main screen of the developed tablet software for
high school students.
While these developments and adoption of new
technology is certainly welcome, standards have yet
to be established. In the field of online learning e.g.
each individual player (e.g. Coursera, Udacity) has
its own teaching style and platform. For tablet-based
teaching, there are standards for electronic Books
(e.g. ePub, iBooks), but as far as truly interactive
content is concerned there are no established meth-
ods, learning styles or data formats available.
Based on our experience from tablet deployments
at various schools in Germany, we propose both a new
learning style for tablet-based teaching and an exten-
sible open data format, which may be freely imple-
64
Weible H. and Seemann E..
Interactive Lessons for Tablet-based Teaching - A Proposal for an Open Data Format.
DOI: 10.5220/0004350200640069
In Proceedings of the 5th International Conference on Computer Supported Education (CSEDU-2013), pages 64-69
ISBN: 978-989-8565-53-2
Copyright
c
2013 SCITEPRESS (Science and Technology Publications, Lda.)
mented by educators.
Our proposition of a new learning style is dis-
cussed in detail in a separate paper (Weible and See-
mann, 2013). The contribution of this paper is there-
fore the proposal of a novel data format tailored to-
wards the requirements and properties of tablet-based
and online teaching. Our goal is to facilitate the distri-
bution of interactive learning content to all platforms
(e.g. iPad, Android, Windows).
In the following we will first discuss related
projects, then continue to describe requirements for
tablet-based learning content. We continue with the
specification of our open data format and conclude
with an outlook on future developments.
2 RELATED PROJECTS
The fields of online and tablet-based learning are
both rather new. Historically online learning started
with the distribution of conventional books and lec-
ture notes over the internet. Demonstations e.g. via
Java-Applets followed, but remained limited in scope.
In particular, there existed very few truly multimedia
or interactive content which covered a larger topic.
With the appearance of online video services like
e.g. Youtube online learning shifted to providing
video lectures. In the beginning, these lectures of-
ten were simple recordings of traditional university
lectures. With the success of Salman Khan’s video
tutorials on Youtube (Khan, 2006), the community re-
alized, that a different learning style is necessary and
that videos should be much shorter.
Only recently, we see interactive courses, which
combine short video clips with online questions and
exercises. In particular, the online universities Cours-
era and Udacity (Ng and Koller, 2012; Thrun et al.,
2012) use this new online learning style. Even though
these platforms build on web technologies, there is no
common data format underlying these courses. It is
also not possible to work with the interactive mate-
rial when no internet connection is available. How-
ever, we start to see cooperations on data format and
platform usage. Most importantly edX has recently
been founded as a joint project by MIT, Harvard, UC
Berkeley and the University of Texas.
Google has also recognized the need for a com-
mon platform and recently released a course builder
to create online courses (Google, 2012).
For tablet-based teaching there are even less com-
mon standards available. Firstly, there is no estab-
lished learning style and publishers seem to feel a dis-
tinct uncertainty how tablet-optimized content should
look like. Moreover, the competing platforms (iOS,
Android) require both educators and publishers to
make a choice which to support. The past two years
have been too short for researchers to tackle the prob-
lem and there is unfortunately very little research
available (see e.g. (Isabwe et al., 2012)). We see,
however, two developments. On the one hand there
are data formats for electronic books (e.g. ePub,
iBooks). On the other hand, interactive educational
apps appear, which cover a certain educational topic.
While iBooks is a specified standard which may
be used by publishers and educators, it has various re-
strictions. Firstly, it is a proprietary standard, which
cannot be adapted or extended to support additional
features. Secondly, distribution is exclusive to Ap-
ple’s Appstore for iOS devices. Publishers in some
countries have therefore been reluctant to Apple’s
iBooks standard. Interactive educational apps, on the
other hand, share no common learning style or data.
They are also difficult to develop and differ very much
in terms of user experience. A common standard data
format along with a reference implementation would
be certainly welcome by many educators.
3 REQUIREMENTS FOR
TABLET-BASED TEACHING
Tablet devices provide a set of benefits compared to
both desktop computers and laptops. First, they are
typically smaller and easier to handle. This is not
only true for the actual hardware, but also the included
software.
For the learning experience touch screens are cru-
cial since they allow more natural and quicker interac-
tion with the device. A survey conducted after the first
months of our tablet deployments clearly showed this.
In fact, we were surprised by how few students missed
a physical keyboard or mouse (Weible and Seemann,
2013).
Based on the experiences from online learning and
the positive effect of touch interaction, we believe that
learning content for tablets should exhibit the follow-
ing characteristics.
Content should be divided into small manageable
parts. It should be presented in a way, which allows
the student to see, hear and interact with the content.
The interaction is accomplished via touch-based ex-
ercises or questions. An example, could be a cloze
procedure, where possible responses can be dragged
into position with finger touches.
Content Creation. In some sense, developing such
interactive content is quite similar to game develop-
ment. In order to allow educators to create content
InteractiveLessonsforTablet-basedTeaching-AProposalforanOpenDataFormat
65
without special knowledge like programming skills,
we need a data format, which abstracts this complex-
ity. That is, the educators define the content only. The
behaviour may be created in a reference implemen-
tation by a programmer. Thus, educators can reuse
these reference implementations for different styles
of exercises and questions (for more detailed infor-
mation see section 4).
A common open data format would allow edu-
cators to share their content and remix other exist-
ing contents. Additionally, programmers could pro-
vide implementations for various operating systems
or adapt the visual appearance and behaviour of the
content to improve the learning experience.
Offline Content. Another hurdle for the deploy-
ment of learning content in schools is the lack of in-
ternet connection. Even in developed countries inter-
net access in every classroom is not available. Thus,
learning content needs to be available offline, which
eliminates the use of many great online resources.
Again, if the content, which is available online would
be in a standardized format, it could be easily trans-
fered to tablet devices (this applies to free content e.g.
released and creative commons licenses or paid con-
tent).
Self-learning. Tablet-based teaching should focus
on self-learning. That is, students should be able to
work through the content on their own. As it is the
case for online learning (e.g. Udacity, Coursera) this
will drastically change who can participate in educa-
tion and what students are able to learn. In our opin-
ion, this is one of the greatest strength of tablet-based
teaching to empower the students to better learn for
themselves.
4 PROPOSED DATA FORMAT
As pointed out in the previous section the data format
should describe the actual content. The visual appear-
ance of this content is part of a reference implemen-
tation, which is independent of the underlying data
format. That is, similar to a graphics format, which
may be displayed with various image viewers.
The design of our data format follows the subse-
quent design goals.
Interactive Content should be Easy to Produce.
For educators it is hard to produce multimedia con-
tent. They have to record audio, video and sub-
sequently process this raw material. This often re-
quires complex software (e.g. video editing software),
which combines video, audio, images and text. The
involved software is not only difficult to learn, but the
content generation is a time-consuming process.
It is getting even more difficult, when the con-
tent should be interactive. In online scenarios, this is
mostly accomplished through programming via e.g.
HTML5/JavaScript, Java Applets or Adobe Flash.
In order to allow the generation of multimedia and
interactive content with minimal effort, our data for-
mat divides content in small chunks or frames, which
can be described through simple text. These frames
may contain videos, images, audio and even interac-
tive exercises.
All content within such a frame is played at once.
That is, on opening the frame e.g., the text is dis-
played and read to the student and an additional vi-
sualization in the form of an image or video is dis-
played.
An example would be:
(01) Not far from River Rhine, not far
from the Swiss town Schaffhausen is a
small cave called "Kesslerloch".
<Kesslerloch.jpg top>
Here the number (01) specifies the frame number.
It is followed by a text, which will be displayed on the
screen. An audio track which possibly reads this in-
formation to the student does not have to be specified
explicitly. Since the audio belongs to a specific frame,
we can associate its file name with the frame number.
In the above case 01.mp3. In many cases, e.g. for-
eign language education, the content author and the
speaker are not the same person. The above naming
convention allows both content authors and possible
speakers not to worry about file names. Associated
media files are specified in brackets. Here a JPG file
is used as an illustration at the top of the frame.
Note, that the splitting in more manageable frames
also removes the need of alignment of media files and
text, which typically requires a lot of effort in video
editing.
Obviously, such a simple format restricts the pos-
sibilities a teacher has, when creating content. This is
intentional. On the one hand, we want the content to
be clean, i.e. there should not be an overflow of infor-
mation or clutter. On the other hand, there is a trade-
off between simplicity and speed of creation and flex-
ibility. As we will see in the subsequent paragraphs,
however, the data format contains a set of predefined
templates, which can be used. Thus, allowing a wider
variety of ways to present content and to include in-
teractive elements.
Note, in particular that the content may even be
CSEDU2013-5thInternationalConferenceonComputerSupportedEducation
66
created without the need for a development environ-
ment with a standard text editor.
Data Format should be Human Readable. All
content should be described in human readable form.
This has various advantages. Firstly, it allows a
speaker to read through the data files and record the
respective audio tracks without special tools. A for-
eign language teacher can e.g. create a lesson, save
it in the above data format and have a native speaker
dub the content. Note, that content, which needs to
be skipped by the speaker is always in parantheses
(e.g. the frame numbers or the media files). XML
files, which are often used for data storage are mostly
human readable as well, but they are visually much
harder to process by the human eye. This is especially
true for speakers, which are typically not familiar with
XML, at all.
The XML and other file formats are optimized to
make it easy for a computer to process the data. In
our file format, we want to make it easy for humans
(either authors or speakers) to create and process the
content. In some cases, this means, that a parser for
the format is slightly more complicated to implement,
but that is a one time effort. whereas content genera-
tion is a continuous process.
For our deployments at high schools in Germany
(Weible and Seemann, 2013) we have already created
a parser and reference implementations of a learning
environment based on this data format for all major
platforms (iOS, Android, Windows).
4.1 Interactive Elements
Our data format describes a set of interactive ele-
ments, which may be used by content creators. Again,
the actual visual implementation is part of a reference
implementation, not of the data format itself.
Shuffled Letters. The first type of interactive exer-
cise is called shuffled letters. It is intended to bet-
ter recall newly learned words and to remember their
spelling. The main application is in the area of for-
eign languages, but it is also applicable to learn terms
in the fields business or engineering. An example for
the description of such an exercise is depicted below:
(02) In our Solar System, there are many
...
-> Correct! In our Solar System there
are many
*
planets
*
.
From the above data, the computer automatically
generates an anagram of the solution, which is in this
case the word “planets”. The student has to then to
use touch interaction to reorder the different charac-
ters.
The corresponding frame as implemented in our
own reference implementation looks as follows:
Remember, the design goal has been to make it
easy to create content. In this case, an author just has
to provide a sentence, which needs to be completed
(denoted by the ellipsis). The success message starts
with an text arrow. This part is, obviously, only dis-
played once the student has found the correct solution.
The corresponding audio tracks need to be stored in
files named 02.mp3 for the first part and 02b.mp3 for
the success message. If an audio track for the success
message exists, the message will not be displayed, but
only read to the student. The correct solution is spec-
ified in the success message by surrounding it with
stars. The same technique is commonly used in e-
mails to highlight a word. As can be seen, it is ex-
tremely easy and fast to create such an interactive ex-
ercise with the above scheme.
Shuffled Words. In a similar way, we can create
other exercises like shuffled words. The main appli-
cation here is to learn the grammar of a foreign lan-
guage by bringing the words of a sentence into the
correct order. For this, we specify the words or parts
of the sentence in square brackets and separate them
by commas. Alternatives from which a student has to
choose (e.g. does versus do) are defined by a slash.
An example is:
(03) [how, long, does/do, the journey,
take/takes]
-> Correct! The answer is
*
how long
does the journey take
*
Note again that the author does not have to shuffle
the words manually, as the computer can shuffle them
automatically at run-time. This procedure is also less
error-prone, since it makes the exercise more readable
for the author. When shuffeling words manually, au-
thors sometimes tend to forget to specify one of the
words in the sentence.
InteractiveLessonsforTablet-basedTeaching-AProposalforanOpenDataFormat
67
Multiple Choice Multiple choice exercises are the
most commonly used type of question. In our data
format the questions may be specified in the following
manner:
(04) How many planets are in our solar
system?
- 5 planets
- 6 planets
- 7 planets
- 8 planets
-> Our solar system contains
*
8
planets
*
. You’re right!
-> There are actually
*
8 planets
*
in
our solar system. Mercury, Venus, Earth,
Mars, Jupiter, Saturn, Uranus and
Neptune.
Note, in particular, that there are two lines start-
ing with text arrows. Next to the success message, a
result message is added when the student has picked
the incorrect answer. Following our naming scheme
for the audio tracks, a speaker would store the ques-
tion text in a file 04.mp3, and the two result messages
in the files 04b.mp3 and 04c.mp3.
In our reference implementation, the above ques-
tion is displayed as follows:
Numeric Computations. In math education as well
as in other technical disciplines, most exercises re-
quire the user to enter the result of a numeric com-
putation. This application is also specified in our data
format.
An example would be:
(05) Marc is giving two pieces of
chocolate to his brother and keeps three
pieces for himself.
3 pieces of 5 pieces = ?%
-> Correct! He is keeping
*
60
*
%.
-> Try again, 50% are half of the pieces
and four pieces constitute 80%.
Again, it is extremely easy for an author to cre-
ate such an exercise. A question mark with a leading
white space character is used to indicate, that the stu-
dent should fill the answer at the corresponding po-
sition. For tablet devices a custom virtual keyboard
is displayed to allow the student to conveniently enter
the result.
Further Elements. In a similar fashion further in-
teractive elements can be specified e.g. fill in the
blanks style exercises. We have currently integrated
twelve styles of exercises and plan to expand this in
future versions.
The full specification and feature set of the first
version of our data format will be finalized and re-
leased to the public on http://scoolbook.org.
5 CONCLUSIONS
Tablet-based teaching will become pervasive in the
next few years. There is demand for content, which
is tailored towards this new style of teaching and self-
learing. Currently, there is no standard for this new
type of content. In this paper, we have proposed a
simple, open data format, which may be adopted by
authors to create interactive content.
The format is extremely easy to use and imple-
ment. In various tablet deployments at schools in Ger-
many, we are already working with a reference im-
plementation for this data format. Our feedback from
teachers working to create new content has been very
positive. In fact, interviews with authors has shown,
that especially non-technical authors very much pre-
fer this format over XML-based formats.
We encourage the community to contribute further
ideas for this data format, create content based on it
and adopt it in other deployments.
REFERENCES
Farias, M. (2011). Pilot-Projekt Tablet PCs im Unterricht
der Landesabteilung Informatik. http://urly.it/1i5w.
Google (2012). Google Course Builder for Online Educa-
tion. https://code.google.com/p/course-builder/.
Isabwe, G., Reichert, F., and Nyberg, S. (2012). To-
wards integrating technology supported peer-to-peer
assessments into mathematics education - experiences
with ipad mobile tablet technology. In 4th Interna-
tional conference on computer supported education
(CSEDU).
Khan, S. (2006). Khan Academy. http://
www.khanacademy.org.
Klinga, S. (2011). Tablet deployment in schools in the city of
Mannheim. http://www.mediaculture-online.de/blog/
?p=8821.
CSEDU2013-5thInternationalConferenceonComputerSupportedEducation
68
Negroponte, N. (2006). One laptop per child. http://
one.laptop.org.
Ng, A. and Koller, D. (2012). Coursera Take the
World’s Best Courses, Online, For Free. https://
www.coursera.org.
Thrun, S. and Norvig, P. (2011). Artificial Intelligence
Class. https://www.ai-class.com.
Thrun, S., Sokolsky, M., and Stavens, D. (2012). Udacity
— Learn. Think. Do. https://www.udacity.com.
Weible, H. and Seemann, E. (2013). Bringing tablets to
schools lessons learned from high school deploy-
ments in germany. In 5th International Conference on
Computer Supported Education (CSEDU).
InteractiveLessonsforTablet-basedTeaching-AProposalforanOpenDataFormat
69