Interoperability Standards for MicroLearning

Abstract

For institutional teaching, learning objects need to be linked into a Learning Management System (LMS). The SCORM and AICC standards have provided the standard for creating platform-independent learning objects, but newer standards like Common Cartridge (CC) and xAPI have new capabilities. The concept of MicroLearning in particular has specific requirements, for example, mobile learning needs to be supported, and learning content needs to be being prepare in many small learning units. These new standards support these specific requirements of MircoLearning much better than the previous standards. This paper provides an overview and a comparison of the various standards and the implications of their application in the MicroLearning context.

Full text

MicroLearning Conference 7.0, Stift Goettweig (Austria), 26.-27.September 2013
Interoperability Standards for
MicroLearning
Re in ho ld Behri nger, Le ed s Me tr op ol it an University
Abstract:
For institutional teaching, learning objects need to be linked into a Learning Management System
(LMS). The SCORM and AICC standards have provided the standard for creating platform-
independent learning objects, but newer standards like Common Cartridge (CC) and xAPI have new
capabilities. The concept of MicroLearning in particular has specific requirements, for example,
mobile learning needs to be supported, and learning content needs to be being prepare in many
small learning units. These new standards support these specific requirements of MircoLearning
much better than the previous standards. This paper provides an overview and a comparison of the
various standards and the implications of their application in the MicroLearning context.
Introduction
The term MicroLearning was coined in 2003 by the Research Studios Austria (KnowledgePulse,
2012), describing “learning in small steps”. While this concept of learning has been around since
antiquity and is being used in a variety of learning situations, computing technology now allows a
new adaptation of step-wise learning, by using computer systems for automated triggers
(“Lerntaktgeber”) of small learning units. When using mobile devices for learning, this learning in
“small portions” becomes a necessity because of the technical limitations of mobile information
access.
In general, learning with mobile devices has received significant attention in the past decade, as
developers of learning systems and content attempted to use the wide-spread availability of mobile
phones worldwide for teaching and learning purposes. While initially the focus was on the question
how to best use the inherently limited technical phone capabilities, with the advent of smartphones
the focus has shifted towards making use of the specific technical capabilities that are provided by
those smartphones.
In order to be able to provide content for a wide range of devices and learning platforms, it is
necessary to define standards which allow a large variety of such devices and platforms to access
learning content and to interact with it. Such standards have been defined and are nowadays widely
being employed. MicroLearning does have particular requirements for these standards.
This paper will provide an overview on the technical capabilities of Learning Systems and technical
standards that are relevant for the concept of MicroLearning. Hereby the focus will be on technical
aspects rather than on pedagogical ones. This in no way implies that those pedagogical aspects are
trivial – on the contrary: the limitations of the learning device have a strong impact on learning and
the ways learning progress can be achieved. In order to successfully do this, the pedagogical aspects
need to take into account what is technically possible. Naturally, the technical developers need also
to take into account the pedagogical aspects of MicroLearning, so that suitable technologies are
being developed for the benefit of the learner and not only per se for technical advancement.

MicroLearning Conference 7.0, Stift Goettweig (Austria), 26.-27.September 2013
Background
MicroLearning, Mobile Learning, and Ubiquitous Learning
Mobile phones with their inherent limitations are well suited for distributing micro-learning content,
allowing, for example, to use 140-character Short Message System (SMS) messages to deliver
learning content and gather responses / feedback from learners. This appeared to be very well
usable in particular by learners in the 3rd World, for example Africa, where mobile phone networks
are more stable and accessible for online connectivity than wired cable networks (Traxler, 2012).
Even simple inexpensive mobile phones did allow this interaction via SMS, and research in Mobile
Learning (m-Learning) began to investigate the development of approaches that could work with
such limited data exchange and communication. With the advent of smartphones, in particular with
the iPhone in 2007, a new dimension of this mobile interaction was possible. Going beyond the 140
character limit of SMS and Wireless Access Protocol (WAP), which had been developed for classical
mobile phones with their limited technical capabilities, the new smartphones were more powerful in
terms of multimedia (fast processor, high resolution screens), interaction (touch screen, speech,
camera, multi-sensing) and connectivity (4G and beyond). This did allow completely new forms of
interaction with mobile devices and did appear to bring desktop-like functionality into the mobile
domain.
These new capabilities also brought new and different challenges upon the use of mobile devices for
learning, and such smartphones required newly designed learning systems and paradigms
specifically for exploiting these technical capabilities. In particular, the capability for location-based
applications, social networking, and user-generated content provided new opportunities which
needed to be fed into learning paradigms, hereby exceeding the “traditional” mobile learning.
Considering Gartner’s hype cycle for educational use of mobile technology (“technology trigger, peak
of inflated expectations, trough of disillusionment, slope of enlightenment, plateau of productivity”),
Laru and Järvelä (2013) have postulated that the era of Mobile Learning (m-learning) is already over
now: it had represented the beginning and the peak of the hype cycle of learning with mobile
technology (around 2008). Instead, they write, the era of Ubiquitous Learning (u-learning) has
recently begun. This u-Learning not only takes into account the mobility and portability of mobile
phones, but also the fact that learning information is available everywhere (where a network exists)
and on a variety of devices, and that users can interact with this information seamlessly, with mobile
devices and computers “embedded in the background of daily life” (Laru and Järvelä, 2013). This
means, learning content needs to be prepared in such a way that it is optimally presented on the
device which the user is employing. The notion that Mobile Learning has been replaced in the
worldwide learning research community by the term Ubiquitous Learning, is, however, not backed
by evidence from Google Trends: the search statistics for “ubiquitous learning” (normalised to the
peak interest) shows a steady decline (Figure 2) after its brief peak in 2007, whereas for “mobile
learning” the trend appears to be at least stable since 2011 (Figure 1). This means that this new term
has not (yet) been universally accepted by the community, even though the actual meaning of
research in m-Learning has indeed shifted to more complex issues, enabled by the technical
capabilities of mobile devices.
Orion Partners (2011) asked if “… Mobile Learning [is] fated to follow the familiar technology hype
cycle” and came to the conclusion that this concept is here to stay, even though its prime time may
still have has to come.

MicroLearning Conference 7.0, Stift Goettweig (Austria), 26.-27.September 2013
The technical advancement towards smartphones also moves to the 3rd World: even in Africa where
currently most phone users are still using the “traditional” low-bandwidth interactions through SMS,
it is expected that by 2017 the majority of phones will be smartphones (Evans, 2012).
Figure 1. Google Trends statistics over time of the search term "mobile learning" between 2005 and 2013.
Figure 2. Google Trends for the search term "ubiquitous learning" between 2005 and 2013.
What is MicroLearning?
Among the consortia which are working on MicroLearning, the overall consensus is that this term in
general refers to “learning in small quantities”. The attribute “micro” can refer to time and content,
and the research questions around this topic are covering investigations how to make optimal use of
the limitations, but also of the specific beneficial features of mobile devices. The limitations are also
prevalent in newer high-capability mobile devices, because these devices still limit the amount of
information that can be conveyed to the learner: the screen size of a smart phone is limited, even
though it is colour, has high resolution, and can be of a reasonable size in tablet / pad computers.
Also the wireless network bandwidth is limited, especially in areas of high usage, because it is shared
with all other concurrent users. And the interaction which the user can provide, precludes fast
typing because thumb keyboards are not suitable for entering large amounts of texts. Therefore, the
fundamental scientific and pedagogical questions of the research in MicroLearning are still relevant
also for the newest generations of smartphones and other mobile devices. Investigations on the
theoretical pedagogical level of MicroLearning have focused on frameworks (Arroyo, 2006) and
taxonomy (Baumgartner, 2013).

MicroLearning Conference 7.0, Stift Goettweig (Austria), 26.-27.September 2013
Interoperability
A fundamental technical issue of learning systems in general is how learning content can be
authored and disseminated so that it is compatible to various user devices and platforms. It is
possible to develop closed proprietary systems which work on particular devices and software/OS
platforms, but this will only reach a limited learning audience, and will be eventually obsolete as
technology moves on rapidly. As a consequence, learning systems and standards have evolved which
use the web and HTML as a standard to disseminate knowledge and facilitate learning. Web
browsers are available on any hardware and operating system, and the standards of HTML and web
programming (e.g. JavaScript) have made it possible for learning to be done on basically all
computing systems and provide a sufficient level of syntactic interoperability. Based on this, the
following standard architecture for learning systems has evolved: A Learning Management System
(LMS) provides learner management and gives access to a set of learning objects, which can be
accessed by user clients through standard web browsers. In general, no additional native software is
needed.
This has provided the interoperability from the viewpoint of the learner, who can now access the
learning content from basically any web-enabled device and can engage with the learning system
through learning and assessment. This approach, however, does not yet address how learning
objects can be moved from one LMS to another. While these learning objects are all web-hosted and
use web-based HTML, the learning objects can be proprietary to the individual LMS. In order to
provide a solution for porting learning objects from one LMS to another, standards like AICC and
SCORM (Sharable Content Object Reference Model) have been developed. They are based on
manifests which describe the functionality of the learning object and hereby provide semantic
interoperability. This includes a standard for defining how these learning objects interact with the
LMS, to get data from the LMS about the learner and how to send feedback from the learner into
the LMS, e.g. assessment answers.
General Technical Interoperability Standards
SCORM and AICC
In 1993, the Aviation Industry Computer-Based Training Committee (AICC) created a standard for
describing the structure and the content of a course. This was not yet based on XML, but worked
based on the HTTP AICC Communication Protocol (HACP) for communication between the course
content and the LMS, where an HTML form is used to send data to the LMS, which in turn sends
information back as a text string (Symons, 2011). Based on this, the Sharable Content Object
Reference Model (SCORM) was developed by Advanced Distributed Learning (ADL, 2013), using XML
files and manifests for describing the course content. SCORM is in principle a collection of standards
and specifications for web-based learning and defines standards for interoperability between
learning management systems and learning content, so that Sharable Content Objects (SCOs) which
contain the learning content, can be exchanged among LMSs. Hereby both the SCO and the LMS
must be SCORM-conformant. Several versions of the SCORM specifications are in use, of which
SCORM 1.2 and SCORM 2004 are the most popular.
The benefits of using such a standard for the interaction between LMS and the Sharable Content
Object (SCO) which contains the learning content, are in allowing these SCOs being used by any
SCORM-conformant LMS and by making these SCOs reusable in different overall contexts. Without
this standardisation, learning content would have to be prepared specifically for each LMS and
would not be portable between different LMSs. SCORM also specifies the packaging of SCOs into a
ZIP file with the Package Interchange Format. A SCORM-conformant LMS provides a run-time API,

MicroLearning Conference 7.0, Stift Goettweig (Austria), 26.-27.September 2013
based on the JavaScript Application Programming Interface (JS API) which the SCO can use for
running standardised routines. This dependency on JavaScript can lead to some compatibility
problems between various web browsers which have different versions of JavaScript embedded.
Fortunately, most modern smartphone web browsers support the very latest JavaScript version, and
compatibility problems only occur in legacy web browsers. One requirement in the SCORM
specification is that the LMS and the learning content need to be hosted on the same server.
Most content authoring systems support both the creation of AICC and SCORM objects. For content
developers, writing content for SCORM-conformant SCOs is easier and faster than for AICC-
conformant SCOs, because the latter often requires that the data coming from the LMS need to be
broken apart by user-created functions (Symons, 2011). For secure connection, AICC supports using
HTTPS, whereas SCORM does not provide secure communication and is quite easy to cheat on
(Addison, 2009), for the reason that it uses plain JavaScript. This is a serious drawback in the
applicability of SCORM for certification purposes, for example in tests, examinations and
assessment.
Common Cartridge (CC) and QTI
To overcome some of the shortcomings of SCORM, the IMS Global Learning Consortium developed
the Common Cartridge (CC) standard (IMS Global, 2013). CC focuses on organisation of distributed
learning content and works with the paradigm of internet-supported learning. Similar to SCORM, CC
supports learning content to be developed independently from an LMS, allowing the content to be
used across a wide range of LMSs and to be shared and re-used without vendor / platform lock-in. In
addition and going beyond SCORM, CC also supports modular web-distributed publishing models
and blended learning, where an instructor can be in the loop. This means that the learning content
does not have to be stored on the same server as the LMS, but can be distributed “in the cloud” or
otherwise across the internet. CC also adds facilities for collaboration and mash-ups as well as virtual
content through URL references.
The specific CC specifications cover the following items:
1. Format for the exchange of content between LMS and course content object(s).
2. Authorisation standard for components.
3. Standard for Metadata based on the Dublin Core.
4. Standard for tests and assessments.
5. Data launch and exchange with external applications
6. Populating online discussion forums.
Both SCORM 2004 and CC are based on IMS Content Packaging; therefore SCORM SCOs can be
converted automatically to CC. CC is currently (October 2013) in version 1.3, which was released in
July 2013. CC supports the following standards: Metadata ISO 15836:2003 (Dublin Core), IMS
Content Packaging v1.2, IMS Question and Test Interoperability (QTI) v1.2.1, and SCORM 1.2/2004.
Related to CC are other digital learning standards, developed by IMS Global: Learning Tools
Interoperability (LTI) allows to integrate rich learning applications with LMS or other environments.
Learning Information Services (LIS) specifies the management of information exchange, describing
people, groups, memberships, courses and learning outcomes.
SCORM is addressing the portability of self-paced computer-based training, whereas CC focuses
more on including the instructor (blended learning). According to IMS Global, CC is also easier to
implement than SCORM (IMS Global, 2012). But it will take a while until CC is supported by as many
platforms as SCORM is currently.

MicroLearning Conference 7.0, Stift Goettweig (Austria), 26.-27.September 2013
For defining questions and tests, the IMS Question and Test Interoperability (QTI) specification has
been defined. It describes file formats for tests, questions, and reporting results. Furthermore, it
defines a runtime model for processing questions and tests. It currently is in version 2.1; earlier
versions have been already widely adopted according to IMS (IMS Global QTI, 2012). QTI is
supported natively by CC since CC v1.0.
Experience API (xAPI)
In order to overcome further limitations of SCORM, another new API has been developed: The Tin
Can API was realised by Rustici Software (Tin Can API, 2013) and was renamed to Experience API
(xAPI) in April 2013. It is similar to SCORM, but extends its functionality to reporting multiple scores
instead of single scores, improved security, platform transitions, and tracking of numerous types of
learning objects and resources (mobile learning, simulations, virtual worlds, serious games, real-
world activities, experiential learning, social learning, offline learning, and collaborative learning). No
web browser and no LMS are required, and the interaction from the learner with the learning object
can be recorded as a grammar with noun, verb, and object. This allows a more free feedback by the
learner in response to tests and questions.
xAPI can be added to an LMS by adding a Learning Record Store (LRS). xAPI learning activity is then
conducted through this LRS, but can also be guided through the LMS.
Authoring Systems
These interoperability standards are difficult to implement “manually” in a learning object or
resource, because the quite complex standards can only be observed and followed by experts in
software development. In principle, developers of learning content could write HTML and JavaScript
code and implement these standards, but that would be very tedious. Content developers instead
need authoring systems which automatically with a click of a button export the learning object into
one of these interoperability standards. In Table 1 a summary is given about popular authoring
systems and the interoperability standard that they are observing. The information about this has
been collected from the individual authoring tool websites in October 2013 and will unfortunately
soon be outdated.
Table 1. Support for interoperability standards by learning content authoring systems, as of October 2013.
HTML5
AICC
SCORM
1.2
SCORM
2004
xAPI
(TinCan)
CC
mobile
multi-
device
Articulate
Storyline






AContent
Free, Open
Source

 1.0
Adobe
Captivate






CourseLab 2.4


eXeLearning
(no new
development
since 2010)
Free,
OpenSource


Lectora
Publisher




 1.0

Trident 2.0


MicroLearning Conference 7.0, Stift Goettweig (Austria), 26.-27.September 2013
Currently, most content authoring systems support the SCORM interoperability standard. It is,
however, to be expected that newer versions of content authoring systems will eventually all
provide support for CC and xAPI.
Learning Management Systems
If learning objects are developed with interoperability in mind, the hosting LMS must naturally
support these interoperability standards. Table 2 shows how popular LMSs support these standards
(this information was collected from the LMS websites and from information provided by IMS Global
on their web site). The listed LMSs may support more interoperability standards than indicated this
table, and this listing of LMSs does not claim to be complete. xAPI is being integrated into many
LMSs, but currently in most cases is not yet officially supported.
Table 2. The support of interoperability standards by popular Learning Management Systems, as of August 2013. Additional
LMSs are listed in the IMS table http://www.imsglobal.org/cc/statuschart.cfm
AICC
SCORM
1.2
SCORM
2004
xAPI
CC
LTI
Agilix Brainhoney
(no new
development
since Dec 2011)



ALEKS


ANGEL



Apollo

Atomic

Atutor


Blackboard





CAMS

CompassLearning



Course 360



CourseMill




Equella

Ganesha



Informetica



(soon )
ILIAS



iSpring Online



Jenzabar eRazer
1.4


Jenzabar JICS
7.5.3

Learning
Environment
(Desire2Learn)


Litmos



Mahara 1.4.1


Moodle 2.0



OLAT


Saba



Sakai





MicroLearning Conference 7.0, Stift Goettweig (Austria), 26.-27.September 2013
UCompass
Educator


Requirements of MicroLearning
MicroLearning poses several unique requirements for the LMS and also for the authoring systems.
One of these requirements is that the learning content to be prepared for this approach must be
broken down into small units, and the LMS must support a large number of such content bits. Also,
this requires the capability to deal with many user interactions, and possibly to support a non-linear
sequencing of the content. In addition, the distribution channels must support mobile devices. The
older interoperability standards AICC and SCORM only partially support these requirements. SCORM
does allow JavaScript and HTML to be tweaked so that there is automatic adaptation of learning
objects to display size and technical interaction capabilities of the devices. This violates, however,
the paradigm of separating content from function, and therefore is not a good practise. Also it
requires the content developer to deal with technical functionality, which is not desirable.
The newer standards CC and xAPI provide more means for inherently supporting the requirements
of MicroLearning. One of the features of these standards is that hosting of content can be
distributed and does not have to be on one single server. Also, the security is improved as these new
standards have improved authentication methods to ensure secure content transmission.
Furthermore, collaboration and sharing is made possible with these new standards, and this is one
of the characteristics of using mobile devices, to be able to contribute and interact to a larger
degree. In xAPI learners can now provide feedback to questions and tests in a free form, following a
grammar. This is the first step towards more rich interaction which can lead to semantic
understanding of learning responses and learning success.
The openness of these interoperability standards to other external applications will also further
contribute to more engaging learning experiences: for example, the Unity Game Engine (Unity, 2013)
for 3D game development which is cross-platform capable, can be used to generate games with
graphic capabilities (gamification) which can set the playful context for the learning elements to be
learned with the MicroLearning paradigm.
As any learning approach, also MicroLearning needs to define Learning Objectives and Outcomes
which are to be achieved by the learner. The quantification of the learning content into small
portions requires also that these learning outcomes are described in the same small scale, as small
“micro” bits. This requires more frequent user assessment and a finer resolution of learning success
measurement. The new standards, in particular CC with its QTI standard, can support this, with
questions that can be dynamically invoked based on various inputs (context, location, etc.).
Learning strategies that are linked to mobile MicroLearning can take into account the learner’s
location, the context of this location and the interaction it offers with regards to the learning
objectives, and the degree of interactive engagement which the learner is engaging or is indeed
engaged with.
Conclusion
In this survey paper we have indicated the importance of interoperability standards. The newly
developed standards Common Cartridge (CC) and Experience API (xAPI, formerly TinCan) which are
going beyond AICC and SCORM, are beneficial for satisfying some of the specific requirements of
MicroLearning. CC and xAPI are not yet supported by all LMSs and authoring systems, but in the near

MicroLearning Conference 7.0, Stift Goettweig (Austria), 26.-27.September 2013
future these standards will also be supported. In the meantime, the developers and learning
managers need to ensure that there is a compatibility between the LMS and the learning objects in
terms of their interoperability. It is a good practice to choose authoring systems which can export
the content to a variety of such standards, so that these learning objects can then be easily adapted
to the growing capabilities of the LMS.
References
Addison, Steve (2009). Cheating on SCORM Courses – It’s Not Difficult. VSSCORM, 31.7.2009.
http://www.vsscorm.net/2009/07/31/cheating-on-scorm-courses-its-not-difficult/
ADL (2013). Website of Advanced Distributed Learning. http://www.adlnet.gov/
AICC (2013). Website of the Aviation Industry Computer-Based Training Committee
http://www.aicc.org
Arroyo, Sinuhe (2006). A Semantic Service-based Micro-Learning Framework.
http://www.cc.uah.es/ie/projects/luisa/papers/2006/SArroyo06.pdf
Baumgartner, Peter (2013). Educational Dimensions of MicroLearning – Towards a Taxonomy for
MicroLearning. To be published in: Designing MicroLearning Experiences - Building up Knowledge in
Organisations and Companies. Edited by Martina Roth Peter A. Bruck and Michael Sedlaczek,
Innsbruck: Innsbruck University Press. http://peter.baumgartner.name/wp-
content/uploads/2013/04/Baumgartner_2013_Educational-Dimensions-for-MicroLearning.pdf
Evans, John (2012). In Five Years, Most Africans Will Have Smartphones. Techcrunch, 2.June 2012.
http://techcrunch.com/2012/06/09/feature-phones-are-not-the-future/
Gonzalez-Barbone, Victor, and Anido-Rifon, Luis (2010). From SCORM to Common Cartridge: A step
forward. Computers & Education 54 (2010) 88–102.
http://lcell.bnu.edu.cn/cankaowenxian/foreign/From_SCORM_to_Common_Cartridge_A_step_forward
.pdf
Hug, Theo; Lindner, Martin; Bruck, Peter A. (eds.) (2006). Microlearning: Emerging Concepts,
Practices and Technologies after e-Learning. Proceedings of Microlearning 2005. Innsbruck:
Innsbruck University Press, 2006.
IMS Global (2010). IMS Global Learning Tools Interoperability™ Basic LTI Implementation Guide.
17.May 2010. http://www.imsglobal.org/lti/blti/bltiv1p0/ltiBLTIimgv1p0.html
IMS Global (2012). Common Cartridge Frequently Asked Questions.
http://www.imsglobal.org/cc/ccfaqs.html
IMS Global (2013). IMS Common Cartridge Specification. http://www.imsglobal.org/cc/
IMS Global LIS (2013). Learning Information Services. http://www.imsglobal.org/lis/
IMS Global QTI (2012). IMS QLC Question and Test Interoperability Project Group.
http://www.imsglobal.org/QTI.html
KnowledgePulse (2012). History of Microlearning and KnowledgePulse.
http://www.knowledgepulse.com/en/history-microlearning-and-knowledgepulse
Laru, Jari, and Järvelä, Sanna (2013). Using Gartner’s Hype Cycle as a basis to analyze research on the
educational use of ubiquitous computing. CSCL 2013.
http://www.academia.edu/3773533/Using_Gartners_Hype_Curve_as_a_basis_to_analyze_research
_on_the_educational_use_of_ubiquitous_computing_CSCL_2013_

MicroLearning Conference 7.0, Stift Goettweig (Austria), 26.-27.September 2013
Orion Partners (2011). Is Mobile Learning fated to follow the familiar technology hype cycle?
http://www.orion-partners.com/is-mobile-learning-fated-to-follow-the-familiar-technology-hype-
cycle/
Pachler, Norbert (ed.) (2007). Mobile Learning towards a research agenda. Institute of Education,
University of London.
Symons, Tim (2011). AICC or SCORM: Which is best for packaging e-learning content? Accelerated
Business Results, 6.September 2011. http://www.acceleratedbr.com/blog/aicc-or-scorm-which-is-
best-for-packaging-e-learning-content/
Tin Can API (2013). SCORM vs The Tin Can API. http://tincanapi.com/scorm-vs-the-tin-can-api/
Traxler, John (2012). Potential of Learning with Mobiles in Africa. World Innovation Summit for
Education (WISE). http://www.wise-qatar.org/content/prof-john-traxler-potential-learning-mobiles-
africa
Unity (2013). http://unity3d.com