Global Learning Systems

Preparing Information Systems for Global Web-based Education

Paul McKey - Managing Director, Redbean Pty Ltd.

Index

Abstract
The Three Elements of Online Education
A Continuum of Information Products
Digital Objects and Systems
Use me like an object but treat me as an Individual!
The role of the LMS - liberation or limitation?
The Learner - the managed or the manager?
The LMS - What is the new role?
The Role of the LMS in Development
Summary
References
Acknowledgements

Abstract

Online Education has three major influencing elements: pedagogy, technology and commerce. Much expertise in the first two exists within universities, but little of the latter.

Taking the student's perspective of an online education system can show that these elements are not working in sync with each other. This is critical if the commercial online enterprise is to be successful.

The development of empowering technologies such as XML (eXtensible Markup Language), along with standard initiatives like the IMS Project, will allow us to develop systems that make use of the strengths of each of the above influencing elements. This though will require a re-engineering of typical organisational information and workflow systems to move away from the current, expensive, 'cottage-industry' of content development and course management.

The major advantages to an organisation of this re-engineering are to:

Provide greater reuse of existing material across subjects;
Provide ease of tracking of all material in the system;
Provide a choice of structures for lecturers and faculty to develop material;
Provide greater responsibility and ownership of information to reside with originator of the information;
Provide greater control of presentation of material;
Provide greater economies of scale via automation;
Provide consistency of quality across all offerings;
Prepare material for commercial sale in a global education market.

The required flexibility of the educational material coincides with the desired flexibility of teaching and learning styles. So, it is necessary to evolve the information systems to support different structures, learning methods and business models.

Top

The Three Elements of Online Education

Given the statement that online education has three major influencing elements; pedagogy, technology and commerce it is apparent that the third element -- commerce -- has only recently achieved sufficient prominence to create this educational triumvirate. Not that commerce isn't already there since it is the harvest of commerce -- taxation -- which currently subsidises education. . Yet most universities see 'online' as a separate and distinct business from the campus university which has research and teaching as its mainstays. Also, in the online market they see the opportunity to experiment with new business models particularly since the rewards come directly to them rather than via the taxation system. Hence, they are entering into different branding alliances with each other, and commercial partners. These partnerships extend the reach of their offerings into markets not previously considered, mainly due to the high cost of a physical teaching presence.

Furthermore, online education accelerates the need to embrace a commercial ethos. Universities are being forced to either "define or be defined" by all sorts of Portal vendors who need little more than a database and Linux box to set up an e-education brokerage business in direct competition with traditional distributors. "The threats to the bricks-and-mortar model of Harvard and other prestigious business schools are distance learning and in-house corporate training". (Christensen, in Jones, 2000)

Peter Drucker predicts that US education spending will increase dramatically beyond the current estimated US$1 trillion but:

"the growth won't be in traditional schools, which currently take about 10% of the GNP (kindergarten through high school, 6%; colleges and universities, 4%). The growth will be in continuing adult education. Online delivery is the trigger for this growth, but the demand for lifetime education stems from profound changes in society". (Drucker, 2000)

Others are saying that the ability to offer technology that "will enable informal learning, making information available and accessible on demand, in increasingly user-friendly ways" is driving the pundits to predict a mature eLearning market which will exceed US$20 billion by 2005. (Giraudo, 2000)

Drucker goes on to argue that in the past we developed skill whereas today around 40% of the developed world's workforce are knowledge-workers and knowledge "is mobile and transferable. It belongs to you, not to your employer or the state. And it is highly marketable today". (Drucker, 2000)

The semester-based course, eventually leading to an award, is the foundation of higher education product at the moment. However, one has to ask if this is the optimum product for universities to be developing and marketing in a commercial environment leaning increasingly to 'knowledge transfer' or Just-In-Time learning solutions. What other products could be offered and more importantly how could existing material and teaching practises be best utilised?

With possible offerings ranging from edutainment, through short courses, to a PhD online, it is apparent that each of the three influential elements - technology, teaching science and commercial viability - will have varying degrees of influence on the design, quality and value of each offering.

Therefore, if commerce is pressing online education for better products and technology is allowing innovative ways to deliver those products then the pedagogy, the developed learning resources, the institutional work practices and the academic rules all need to be continually flexible, and even customisable, to suit the market. However, to date that which has been produced is little more than the re-purposing of existing material for the web, combined with the continuing forcing of students into constraints based upon campus centric work practices and outmoded academic rules. There is seemingly little awareness that some commercial providers with no such constraints are currently building capability to directly challenge the university system. These commercial providers are taking universities' core quality material and extending its usage by repackaging it to make it commercially attractive. The end result of this could be that some universities in the online market will be reduced to content and accreditation providers while all the service elements are delivered elsewhere.

It seems universities do need commercial partners if they are to successfully embrace the online market. However, they then have to understand the business and what is required to work with their partner successfully. While the business elements mentioned above are vitally important, this paper will concentrate on how the development and flow of educational content can influence the nature of the final product - it will outline how online products may differ, how technology standards can aid their development and how the key component of an online system, the Learning Management System (LMS), while possibly ill-suited to the task, can adapt to suit the needs of newly emerging online education systems.

Top

A Continuum of Information Products

If commerce is to have an influence in online education it will be most apparent in the design of the educational offering. The one semester university course is designed as a building block to be assembled into an academic program which leads to an accreditation. But, if Drucker and others are only half right, this once dominant post-secondary market will be increasingly matched by shorter, more relevant, and timely offerings. Some people may just wish to use the information and knowledge gained for personal development while others may also wish for acknowledgment of their effort via recognised accreditation. But without predicting what will be developed let us just look at what could be developed for now.

It is though possible to consider a continuum of information products. I have refrained from calling them educational products as they may in some cases have no intrinsic educational value (if that is possible). They may be simple 'page-views', research reports, short-courses or complete online degrees. These attributes of the product could have a graduated range of values which may be few or many.

Range

[Value]

[Attribute]

[Value]

behavioural

learning view

cognitive

short

time

long

inexpensive

cost

more expensive

none

credential

PhD

self-paced

teacher support

leader-lead

Figure 1: Information Continuum

Figure 1 shows some sample attributes defining the 'product' and suggests some nominal ranges of the values. These are continued below:

How is it packaged [size] ?
dependent on usage [information to high order learning]
or learning style [from behavioural to cognitive]
or showing the level of engagement [shallow to deep]
or cost and time [low cost and short to expensive and long]
or level of credential [none to PhD]
or level of reusability [once-off to repetitive]
or teaching support [self-paced to leader lead]
or tactile usage [explanation to simulation]
or course design [coursework to thesis]
(Coxwell, 1996)(Reeves, 1992)

To date, universities have perceived themselves in the right hand of the continuum whereas the left end has been the domain of 'training' . But from the consumers view the unaccredited training market is blurring with the accredited higher education market as both of them are tending towards the vocational. After all, what are medicine, dentistry, law or an MBA if not vocational? (Dessaix, 1997)

We recognise that the line between training and education has blurred. So too, the line between informal and formal education as 'free' information has been merged into online courses. Information from within educational courses is now often freely available via the web. So, is it information or education? Let the user decide!

One thing we do know is that, in most cases, the current hand-crafted educational material developed inside universities is ill-designed for repackaging in any granule smaller than the semester long course. Because all the internal work practices and academic rules are based around this granule this will, more than any other factor, limit the university's ability to enter the flexible product market and hence gain increased return on investment.

The effort involved in shortening and repackaging these larger products could be lessened dramatically. A redesign coupled with increased discipline in the development methods used would allow new product areas to open up. This redesign should centre around the concept of making the educational resource the "discrete granule" of development, allowing the multiple combinations of these resources to match variables such as those mentioned in the above continuum of informational products. These resources can be either the more concrete types, such as the images and hard data of the science teacher, to the more abstract such as the history teacher's well-honed anecdote.

Merril (1998) calls these discrete granules "knowledge objects". Robson (1999) calls them "learning resources". These "instructional elements", which could be plucked from a "resource pool" (Mckey, 1997) or repository, are defined as "any resource that an instructor makes available to a student for a pedagogic purpose and can be realized in some sort of learning environment" (Robson, 1999). This is very much an educational perspective since if one was to write a learning resource for an Automatic Teller Machine's Help system the instructor would in fact be replaced by a system instructor or virtual instructor. Also, the instructor or instruction is itself a resource and therefore, in a student managed learning environment, may be turned on or off at any time. It does though capture the relationship between purpose and environment, possibly the two factors most influencing an 'object economy'. In education systems of the past both were tightly controlled and predictable, now they are changing as fast as the weather.

Already groups such as the Educational Object Economy Foundation (http://www.eoe.org/ ) are busy packaging learning objects and have a strategy to provide educational content at low cost to educators around the world. The major delivery package for this content is the Java Applet.

In most cases the deliverable product will be defined by the service it is trying to provide. Necessarily this matches the purpose and environment. The following example services are of interest in this context. They are each based on the combination of the learning need (purpose) and method; and the capacity of the technology (environment) to support it and make it 'come to life'. This is a unique combination which has been discussed and imagined for many years but is now coming closer to reality.

Student Managed Learning (SML)
The concept of Student Managed Learning delegates the control of as many elements of the learning process as possible to the student, including elements such as course administration, educational design and technology choice etc.

Many interfaces only allow proactive and reactive access to features and services. Control means one selects, then possibly designs and manipulates the services and interfaces. In this environment the student would have access to 'tune' the learning resource parameters. These could include the educational parameters such as: level of difficulty and associated credential; pace and activity level; tutor support and so on as well as the technical delivery parameters. This control would also be particularly useful in high bandwidth environments for simulation learning, virtual reality, environment modelling and real time analysis etc.

Electronic Performance Support Systems (EPSS)
Typically EPSS has been seen as support for a training environment. It is aimed to provide support just-in-time for any performance in multiple environments. This could be as varied as changing a tyre (delivered via radio to an in-dash device), medical operations and procedures(via conferenced high speed video), or as work flow automation (via expert system libraries) etc. That is, the appropriate learning resources are made available to match as many multiple purpose and environment combinations as are feasible.

Whereas in the past education/information systems existed in a didactic flow of information from teacher to learner, newer technology based systems also allow a shorter cycle of building knowledge by capture and reuse of information and events. This also opens up the possibility of acquiring knowledge objects, modifying them and either sharing or re-selling the 'improved' object. This is the pattern of open source software and the [download, re-mix, upload-for-comment] loop of music sites such as Mixman. (http://www.mixman.com/)

This new cycle becomes - Information > Teacher > Learner > Discovery > Information… as in the "Orbicular Model" (Gilliver, Randall, Pok, 1999). All elements in this cycle could be replaced by 'knowledge-systems' to build a truly adaptive learning environment.

Figure 2 - "The Orbicular Model" (after Gilliver, Randall, Pok, 1999)

The objective is that once these services and technologies are developed, refined and tested within the education environment they can then be generalised for use outside the strict definition of the public and private education systems. Information access and manipulation will be one of the defining elements of the next age. Players need to be positioned to exploit this as a market as the capability and awakening demand develop. The possibility for expansion into corporate, industry and domestic environments, is unlimited.

But regardless of whether universities are developing new products for within the Higher Education sector (such as alliances and cross accredited degrees etc.) or making product for other non-traditional markets (such as the vocational and corporate customisation etc) one factor will be required, that is the easy exchange of both learning resources and students between both aligned and competing systems.

Top

Digital Objects and Systems

Don't try to understand 'em,
Just rope 'em, throw, and brand 'em.
Soon we'll be livin' high and wide.

Rawhide
(Words and Music: Ned Washington and Dmitri Tiomkin in Brooks and Marsh, 1999)

For hundreds of years we have understood the value of branding cattle (bovine objects) with identifiers. More recently though the beef industry has beaten the education industry in the race to use meta-data to describe their product. While meta-data specifications are still on the to-do list for most educators, in Australia, schemes like The National Livestock Identification Scheme (NLIS) are demanding that any cattle sold through European Union markets be tagged to ensure "traceback and traceforward capability". (NLIS, 1999) Microchip tags, rich with relevant data, are mandatory. How many times have you wished for this capability to trace content and associated files when building online courses? If we can label a 500kg cow why is it so difficult to label educational resources?

Figure 3 - Bovine Object with Attributes as meta-data

For a cow, meta-data could include conceptual attributes, such as quality, or event records, such as birth date, or history such as breeding and production or just a simple list of physical attributes as in Figure 3.

If one is developing online educational material in HTML the content markup phase is typically also used for the structuring and presentation of the material. Unfortunately, it then becomes difficult to extract or change the content, the structure, or the format at a later date. Further, since material is embedded by its document and only implicitly described (such as the assessment for lesson 5 in module 2 in the Biology Course) it becomes difficult to find and reuse anywhere else. The above describes unstructured data from a document centric system. The move to data centric systems brings us structured data where, because everything lives in a database, we can label it. This has in turn raised the possibility of systems designed to 'automate' the content development process and finally uncover the Holy Grail of software and content developers the world over; reusability.

Most universities and corporate organisations have faced the challenge and promise of capturing, managing and reusing institutional 'knowledge'. A complete Knowledge Management (KM) System may have something like the seven layers which Tiwana (2000) has described:

" Interface Layer; Access and Authentication Layer; Collaboration Filtering and Intelligence; Application Layer; Transport Layer; Middleware and Legacy Integration Layer; Repositories".

Information though, in its raw state as data, exists in a cycle with only three major nodes described by the relevant action on the information, capture, store and retrieve. Apple Computer and IBM in their VITAL project of the early nineties, considered that these three critical actions form the lifecycle of any information system regardless of its eventual use.

Educational content development and reuse is only one possible domain of a knowledge management system and obviously has many aspects peculiar to it but also similar to other informational uses. However, with few standards to guide us the systems developed so far have ranged from the sophisticated, yet cumbersome, Lotus Notes ® to umpteen in-house developed, structured publishing systems which typically only deliver fairly restrictive, template driven, Data-in, HTML-out environments. Not very exciting visually and certainly not very flexible pedagogically. They are a simplistic attempt to utilise the capture, storage and retrieval cycle. They are usually designed to deliver institutional wide consistency of presentation and efficiency of development work practices but offer little in sophistication, to the learner, in possible comparison to a well-crafted course. Hence the continued dominance of hand-built HTML.

Interestingly, and perhaps because of the last statement, little of the technology developed for corporate KM systems has been successfully utilised in the design of the commercially available Learning Management Systems (LMS). While these actually use a database they mainly still support the use of externally hand-built, unstructured material as opposed to structured data. This is because they have not integrated the capture component of the information cycle, only the storage and retrieval. This is sure to change.

Top

Use me like an object but treat me as an Individual!
Unfortunately though the capture of unstructured data, no matter how clever the system, will not provide us with our goal of reusability. No, some discipline is required from the very outset of the cycle and to achieve this many have proposed the explicit labelling of educational resources as objects, each one of which can be:

Identified - via a UUID (Universal Unique Identifier)
Described - via meta-data
Placed - via XML (eXtensible Markup Language )
Formatted - via XSL (eXtensible Style Sheets)(McKey, 1999)

The objective is to maintain the independence of the three major components of a typical lesson or learning event; the content, structure and format. Or at a higher level, as Robson puts it, "to divorce the teaching and design process from the medium in which it takes place" (Robson, 1999). Once we have the basic labels above it allows the simple restructuring of lessons and reuse of resources in a number of ways. The ability to hand-build sections should also be retained particularly where it is necessary to add 'glue' between components to provide the narrative or context.

Looking at the specific information cycle of educational content and its development allows us to see some weaknesses in current systems. It also shows the demarcation of a number of tasks in the development process (to possibly avoid scenarios such as the $80k professor marking up HTML for days on end). A simple development program would look something like the following with the major roles identified that could be carried out by one or more people or systems.

The Learning Resources Development Cycle:

Capture
- Subject-matter experts analyse their domain and develop content, in conjunction with Teaching and learning experts and System guidelines and capture information as objects.
Storage
- the System or Others label and store the object.
Retrieval
- subject-matter experts in conjunction with teaching and learning experts define optimum structures, sequences and formats for presentation of the material and concepts (captured as XML Document Type Definitions (DTD)).
Presentation
- the system, or Course builder, retrieves resources and uses assembly tools such as DTDs etc to structure, format and present the finished 'lesson' in the LMS.
- alternatively the LMS stores all components and alternative structures and formats as DTDs and XSL which allows the Learner to assemble their own lesson.the System or Others label and store the object.

Note the addition of the Presentation phase which is not actually part of the main cycle but tangential to it. Therefore Presentation is typically repeatable yet ephemeral.

Figure 4 - Information Lifecycle

Some consider standards as stifling innovation. Others see them as a platform for innovation. When one travelled the world in 1990, say visiting clients, you couldn't plug your Laptop into your client's network. They just weren't compatible until we set and implemented LAN standards. Similarly in 2000 why can't my LMS exchange data with any other vendor's LMS? Make them standard at these layers and then let the vendors innovate in the functionality. The functionality is usually where the differentiation of the product is prominent. This also raises the possibility of buying interchangeable standards-based layers from one vendor and your favourite functional environment for them to run in from another.

While it is becoming critical for software vendors to shift to a more open platform, so it is for educators to understand the advantages of this for their students. Educators must support and promote an ability to interoperate. On the contrary it is often found that educators form strong alliances to their current products which leads to LMS wars (similar to computer operating system wars) where they discuss feature lists of competing products ad nauseam. Yet few seem to consider the student 'outside' this environment and hence one feels that the online educators choose systems based more on their needs than those of the students. The vendors also tend to market and promote this self interest and allegiance to their product.

Still the use and usefulness of many of these features are worth debating. For instance I was recently informed by one university it did not support Chat on its system as it did not consider Chat had any pedagogical value. Of course it doesn't. If Chat had pedagogical value that would mean we would have to award a degree to every listener of talk-back radio, a Masters to every barber in the land and at least a PhD to the participants of the Jerry Springer show! No, Chat is about building community and hence it is a highly desirable feature if the designer chooses the activities to be "cooperative, collaborative and narrative-based" rather than "individual, procedural, problem-solving and/or competitive and abstract activities". (Campbell, 2000)

So while functionality is important, if I was spending my institution's money on an online education solution I would be asking some hard questions along the lines of those which Reid suggests:

" Priority 1: Does the system interoperate with all corporate systems? Priority 2: Will the system extend to use by the University as a whole? Priority 3: Is there maximum functionality in the system?

This is a critical shift in the traditional view of educational software, where functionality has in the past been the major determinant for product selection." (Reid, 1999)

It is just these types of questions and priorities which the US Department of Defence had in mind when publishing the "Sharable Courseware Object Reference Model" (SCORM) in order to "provide a sound economic basis for investment" in technology-based learning. These would satisfy requirements for learning content such as "reusability, accessibility, durability and interoperability". (DoD, 2000).

Returning to the three layers above, it is worthwhile investigating some of the commercial and standards development activity occurring within each and to consider some preferred compliance levels.

LMS - Presentation layer:

Since the Presentation layer is where we 'access' the functionality of the system it needs to be highly customisable to suit individual preferences and differing usage patterns etc.

It should therefore be component based and modular.
A change to the look and feel could be either personal or universal.
Interfaces could possibly be 'bolted on' to the layers below since vendors such as Epicentric (http://www.epicentric.com/) develop customisable interfaces which then use emerging standards such as the Information Content Exchange, (ICE) (http://www.icestandard.org/), to dynamically load personalised information such as course enrolments or weather reports etc from conforming information vendors.
Interfaces should be language independent allowing the support of multi-lingual presentation of at least the system functions and also content if it is translated.
Interfaces should be based on student management and not solely on student access principles.

"We are thus in the midst of a structural migration from data access applications to more pro-active enterprise decision support systems (DSSs)" (Kalakota and Robinson, in Levine et al, 1999:269).

LMS - Education Layer:

To achieve our objectives for encapsulation and exchange of learning resources the LMS needs to accommodate many different types of learning resources, from standard MIME types to complete SCORM compliant courses from similar and competitive systems, either down the hall or across the world.

The IMS Content Packaging specifications in its version 1 scope provide rudimentary guidelines for the packaging and exchange of passive content while the version 2 scope will add administration interfaces and methods for use in a run time environment.
The LMS should be agnostic to structure and not therefore force an implicit navigation design by nature of its own structure. (Most LMS currently use passive Information as the primary navigation delimiter whereas Function would be preferable in an 'active' learning environment.)
The LMS should provide tools to 'learn' from its content's meta-data such as provided by IMS and IEEE or EdNa standards. The LMS should support the management of continuous publishing for both teacher and student.
The LMS should support the easy integration of 'best of class' communication tools.
The LMS should support multiple assessment models in addition to those in the IMS Question and Test Specification.
The LMS should be developed on the principles of open standards, using readily available database systems with clearly published Application Programming Interfaces (API) for the customisation of the functionality where necessary.
The LMS must support Unicode and Internationalisation standards.

LMS - Administration Layer:

The Role of the LMS at this layer is to provide the functionality to support a rich experience for the user of the system. Hence it is critical that the LMS can integrate to other systems otherwise concepts such as Student Managed Learning are sunk before they start. This layer has its own information cycle as it captures, stores and allows retrieval of student data, tracking data, interactions, tests, enrolment, content selection, payments and so on and allows both educators and students to manipulate the learning environment.

The IMS Enterprise specification provides standards for the exchange of data between systems such as Human Resource, Corporate Training, Student Administration and Library Management systems. This includes such elements as personal profile data, group and enrolment management and final grade processing as well as security.

LMS - Functional Layer:

The Functional layer provides access to all of the above. Some would suggest this is the job of the Presentation layer but I argue that is only where it happens not how it happens. As suggested this is usually the proprietary domain of the vendor or system developer. Functionality is usually manifested as code programmed as the middle tier between the client and the data schema, in such languages as ASP, JavaScript, Tango Cold Fusion or WebObjects etc.

Code embedded in HTML pages unfortunately further compounds the counter-productive issues of the HTML editor as it ties function to form so that to change the presentation requires a programmer and not just a web (HTML) builder. Sorley (2000) in development of new LMS designs attempts to rectify this by divorcing form and function, hence allowing multiple presentation styles to be rendered on the same function. Since the average LMS already has roughly 90% functional equivalence with its competitors (since they all copy each other) this would possibly allow a system where the academic or student could swap interfaces from WebCT, TopClass, BlackBoard, Prometheus or Learning Space etc. on the fly. If we could define what that 90% was comprised of and capture it in a standard environment then the LMS wars would be over and we could get on with the teaching and learning.

It must be pointed out that there are a number of, sometimes competing, standards bodies involved in development of the functionality of these systems. While the IMS will hardly be the 'ASCII of the Education world' it does provide a starting point for the improved integration of content, people and processes and currently has the broadest international involvement.

(ASCII was the first widespread standard to allow the interchange and display of common data between differing computing systems.)

Top

The role of the LMS - liberation or limitation?

The major terminus for learning resources developed and transported by the above systems is the Learning Management System (LMS), where the teaching and learning and eventual usage of the resources takes place.

The Web liberated the teaching and learning fraternity from the confines of the physical classroom to overcome distance and time and provide new ways to bring people together. It has also been the single focus for a collection of technologies allowing the development of the online education business. One of these, the LMS, also liberated the teaching fraternity from the confines of their own IT departments, by allowing the download of a simple application to publish course material and manage a student cohort. Alas, like the HTML editor, this has probably been counter-productive to the overall goal of the total student experience, but they couldn't wait for IT to wake up. Accordingly the grassroots movement has turned online education into a multi-billion dollar industry yet many of the original innovators will never reap the rewards.

IT did eventually wake up and so did administrators! They are realising that to provide a consistent experience and compete on a service level, the server under the professor's desk is not the optimum delivery method. So, the current popular LMS design of a downloadable, stand-alone, easily configured, loaded, and managed system yet not mission critical and not interoperable (some not even with standard databases) and not capable of supporting multiple, flexible, teaching and learning styles, has had its day.

The current crop of LMS designs are ultimately restrictive. They are typically tightly-coupled, proprietary systems with the features, interactions and pedagogical flexibility defined by the software vendor. What is needed is a loosely-coupled system, based on standards and defined by a learning plan and Learning Service Agreement (LSA) between the education provider and the student with a component based interface for learner control. This is Student Managed Learning.

In a recent strategic analysis report titled "Evaluating an End-to-End E-Learning Infrastructure Provider" (Aldrich and Ross, 2000) the Gartner Group identifies at least 250 requirements of an end-to-end learning infrastructure. Most of these features and functions are not provided by the typical LMS as we know it, nor could they be since most of them involve integration with numerous other systems to provide all of an online student's requirements. For instance, in a true end-to end system I would consider at least four currently discrete systems need to integrate closely to provide a complete service: a Customer Relationship Management System (CRM); a Student Administration System (SAS); a Continuous Publishing System; and a Learning Management System (LMS). Therefore, any LMS that doesn't integrate is near worthless.

The Gartner report also outlines how it sees the development of the e-learning marketplace over the next few years and highlights three major transitions:

1999-2001: Best of Breed

Defined by isolated and proprietary systems needing to form "strategic alliances with other best-of-breed enterprises to provide a more complete solution".

2001-2003: End-to-End Infrastructures and Suites

"Integrated offerings from large organisations will begin to be more compelling than best of breed for many organisations… because of their ability to provide strategy and full implementation services."

2003 and Beyond: Content Is King

"As the e-learning infrastructure becomes commoditized, standards will finally take hold and content will become king. New media-savvy power vendors will dominate e-learning".
(Aldrich and Ross, 2000)

Top

The Learner - the managed or the manager?

The first administrative use of IT in universities (Student Administration Systems) was for many years only concerned with the management of the student. Most of the current crop of online teaching and learning systems still maintain that focus in that they have been developed around a primary design brief of achieving greater efficiency of the administration component of teaching. Some academics would also argue that these systems have actually increased their management load where they are now expected to do tasks once only doable by the IT/Administration departments.

One major flaw is that these systems still emphasise the student as the one being managed as opposed to the manager. This is changing! The modern student, particularly mature adults, see education as just another thing for them to manage in their lives along with job and family etc. This view sees them as more than the passive recipients of instruction at pre-arranged times but rather as active learners constructing an education that is relevant and achievable within their individual constraints such as budget, time and ability.

This then casts the whole education system as a resource to help students achieve their social, political, economic, cultural and spiritual goals.

However, if the LMS is the contact point for the online student, is it satisfying the above goals? Not really, but as stated earlier, it is also the responsibility of the teaching institution to provide the learning resources in a manner that allows the student to manage their personal learning experience. Hence the combination of product (assembled learning resources and practices) designed for student management, and an LMS designed to facilitate their selection and usage, is a necessary first step towards a goal of delivering a total experience to learners. So, what could this LMS look like?

Top

The LMS - What is the new role?

Recently I suggested an online education system has four interdependent layers: Presentation, Function, Education, Administration (McKey, 1999). This architecture is broader than just Information Systems, by considering the integration of learning with work and society.

Looking at the typical LMS, from the somewhat narrower Information Systems perspective, the LMS can be defined along similar lines. In relation to the movement of content and data , different activities occur at each operational layer:

Presentation - for personalised and dynamic content and functionality

Education - for teaching and learning, Learning Resources, etc.

Administration - student data, course statistics, tracking, administration etc

(Alternative names are, front-end, T&L layer, Back-end). The Functional layer is of course ubiquitous and it is the functionality which most consider, some believe mistakenly, before other features of a system.

As the Gartner Group suggest, the LMS as a digital island is not particularly useful. The major requirements of any LMS is that it is open at each of these layers and supports the appropriate standards (which could still be via proprietary interfaces) for exchange of the data appropriate for that layer. As Figure 5 shows, each layer should be capable of standard communication and interoperability, independent of the other layers.

Figure 5 - Independent LMS Operational Layers

So while functionality is important, if I was spending my institution's money on an online education solution I would be asking some hard questions along the lines of those which Reid suggests:

" Priority 1: Does the system interoperate with all corporate systems? Priority 2: Will the system extend to use by the University as a whole? Priority 3: Is there maximum functionality in the system?

This is a critical shift in the traditional view of educational software, where functionality has in the past been the major determinant for product selection." (Reid, 1999)

Returning to the three layers above, it is worthwhile investigating some of the commercial and standards development activity occurring within each and to consider some preferred compliance levels.

LMS - Presentation layer:

Since the Presentation layer is where we 'access' the functionality of the system it needs to be highly customisable to suit individual preferences and differing usage patterns etc.

It should therefore be component based and modular.
A change to the look and feel could be either personal or universal.
Interfaces could possibly be 'bolted on' to the layers below since vendors such as Epicentric (http://www.epicentric.com/) develop customisable interfaces which then use emerging standards such as the Information Content Exchange, (ICE) (http://www.icestandard.org/), to dynamically load personalised information such as course enrolments or weather reports etc from conforming information vendors.
Interfaces should be language independent allowing the support of multi-lingual presentation of at least the system functions and also content if it is translated.
Interfaces should be based on student management and not solely on student access principles.

"We are thus in the midst of a structural migration from data access applications to more pro-active enterprise decision support systems (DSSs)" (Kalakota and Robinson, in Levine et al, 1999:269).

LMS - Education Layer:

To achieve our objectives for encapsulation and exchange of learning resources the LMS needs to accommodate many different types of learning resources, from standard MIME types to complete SCORM compliant courses from similar and competitive systems, either down the hall or across the world.

The IMS Content Packaging specifications in its version 1 scope provide rudimentary guidelines for the packaging and exchange of passive content while the version 2 scope will add administration interfaces and methods for use in a run time environment.
The LMS should be agnostic to structure and not therefore force an implicit navigation design by nature of its own structure. (Most LMS currently use passive Information as the primary navigation delimiter whereas Function would be preferable in an 'active' learning environment.)
The LMS should provide tools to 'learn' from its content's meta-data such as provided by IMS and IEEE or EdNa standards.
The LMS should support the management of continuous publishing for both teacher and student. · The LMS should support the easy integration of 'best of class' communication tools.
The LMS should support multiple assessment models in addition to those in the IMS Question and Test Specification.
The LMS should be developed on the principles of open standards, using readily available database systems with clearly published Application Programming Interfaces (API) for the customisation of the functionality where necessary.
The LMS must support Unicode and Internationalisation standards.

LMS - Administration Layer:

The Role of the LMS at this layer is to provide the functionality to support a rich experience for the user of the system. Hence it is critical that the LMS can integrate to other systems otherwise concepts such as Student Managed Learning are sunk before they start. This layer has its own information cycle as it captures, stores and allows retrieval of student data, tracking data, interactions, tests, enrolment, content selection, payments and so on and allows both educators and students to manipulate the learning environment.

The IMS Enterprise specification provides standards for the exchange of data between systems such as Human Resource, Corporate Training, Student Administration and Library Management systems. This includes such elements as personal profile data, group and enrolment management and final grade processing as well as security.

Code embedded in HTMLpages unfortunately further compounds the counter-productive issues of the HTML editor as it ties function to form so that to change the presentation requires a programmer and not just a web (HTML) builder. Sorley (2000) in development of new LMS designs attempts to rectify this by divorcing form and function, hence allowing multiple presentation styles to be rendered on the same function. Since the average LMS already has roughly 90% functional equivalence with its competitors (since they all copy each other) this would possibly allow a system where the academic or student could swap interfaces from WebCT, TopClass, BlackBoard, Prometheus or Learning Space etc. on the fly. If we could define what that 90% was comprised of and capture it in a standard environment then the LMS wars would be over and we could get on with the teaching and learning.

(ASCII was the first widespread standard to allow the interchange and display of common data between differing computing systems.)

Top

The Role of the LMS in Development

The Chicken, the Egg and Evolution - The Continuous Development cycle

Since the LMS is the focal point of any online education system it makes sense that it should be tightly integrated with any product development process to ensure optimum efficacy of the developed material and system components. The LMS could itself be an information cycle (capture, store, retrieve) used in the larger cycle of continuous development. That is the "design, develop, deploy and evaluate" cycle - a continuous publishing system with an in-built R&D component. To better explain this cycle let me use an example.

The system is a cycle.
The major components are the Chicken and the Egg.
The chicken defines what form and functionality an egg will have and lays it.
The egg provides content which then defines what the features of the next generation of chicken will be and hatches it.
Chicken grows up, cycle starts again.

What this highlights is that both the Chicken and the Egg have equal capacity to influence the other. The Chicken is the LMS (including, Internet technologies, commercial trends etc). The Egg is the Development Program (including latest research into learning theory, education technology etc.).

So which came first?

<joke>

A chicken and an egg were lying in bed. The egg had a very annoyed and frustrated look on its face while the chicken, lighting a cigarette, said "Well that solves that question!".

</joke>

Let's return to the development of educational products in a university or other online-education, content development houses.

In the several months that a semester typically runs combined with the several previous months when the course was being developed/updated equals a long period of continuous change to both Internet technology and market variables while the 'product' is effectively frozen. It is obvious from a competitive viewpoint that the development cycle or product lifecycle is too long.

It is the argument of this paper that the move to a continuous development/publishing environment based on learning resources that are easily identifiable, and reusable digital objects, will have a major influence on shortening this cycle.

However, any advantages these new advances may have will lose their impact in an environment where technology and work practice change may take years to occur. Hence many universities are partnering with a commercial service provider to help with this issue. Partnering is the optimum arrangement so that mutual benefits such as the raising and maintaining of skills via transfer from provider to institution and vice versa is achieved.

There are many actors in this development cycle but for simplicity I will identify the three major roles. (Remember that these are roles and could be fulfilled by one person or hundreds). They are the LMS Developer, the Learning Program Developer , and the Education Consultant. This last role is either provided by the service provider, the institution or a combination of both and it is what I consider is the missing ingredient in the Chicken and Egg story and I like to call it Evolution. (Apologies to Creationists).

Where is the learner in all this? Ideally in a learner centred system all three of these roles are working for the learner but they are represented in the LMS.

Once again for simplicity, and to protect the innocent, I will rename the roles as:

Chicken - (System Developer)
Egg - (Academic/Teacher/program developer)
Evolution - (Education Consultant)

The major task of the Chicken is the development of the system and hence is a fairly technical role. So, they are busy coding and testing, scratching around for new ideas and keeping an eye on the other chickens.

The major task of the Egg is being a subject-matter expert which provides yummy content. So they are busy researching, thinking and writing and developing themselves into a saleable product.

Therefore, the task of Evolution is to be everywhere at once and considering all the options. Evolution has to talk with learners, discover their needs, consider new features with the Chicken and Egg, articulate and educate the Egg on these new technologies etc, inform the Chicken of new teaching methodologies and provide the continuous improvement of the experience for the learner. If a new feature is developed in the Chicken, the Egg has the option of incorporating that feature in the next update/publish cycle (hatching) and vice versa in the next lay.

Hence the importance of evolution - and the need for Educational Institutions to evolve. If universities fall behind in meeting the needs of the online learner, the symbiotic relationship they have enjoyed in the past will lessen in importance and they may well be rendered unnecessary and possibly become extinct. That's how evolution works!

Top

Summary

The current wave of educational technology offerings are becoming more monolithic and proprietary in design as they attempt to be everything to everyone. This has seen the demise of many software products in the past as the level of product innovation is inversely proportional to the number of users of the product. Unfortunately the domination of the technology in the online education environment has shifted the role of the technology from one of liberation to one of limitation, as educators are forced into 'product camps'.

The development of learning resources as independent objects in combination with emerging standards, and less restrictive teaching and learning environments, would allow for more responsive and relevant educational products to be developed satisfying a growing demand for flexibility in offerings. Hence information systems which work to support the dynamism of the relationship between teachers, learners and their goals will provide a better experience for all involved which must also make for better business.

One of the challenges of this era is to provide the reliability, predictability and consistency of the ingredients and experience without losing the 'flavour'. Personalisation and uniqueness of experience are, at the end of the day, more important than automation and reusability yet all have their place. Rather than making the technology omnipresent the major objective should in fact be to make the technology disappear. The complexity in the back end is of little use without a corresponding simplicity at the front end for the user.

References

Aldrich, C., and Ross, R., 2000, "Evaluating and End-to-End E-Learning Infrastructure Provider", Part of Strategic Analysis Report, 01 May 2000. (only available via subscription from The Gartner Group, http://www.gartner.com/)

Brooks, T., Marsh, E., 1999, "The Complete Directory to Prime Time Network and Cable TV Shows", 7th Ed., Ballantine Books

Campbell, Katy, 2000 "Gender and Educational Technologies: Relational Frameworks for Learning Design", Journal of Educational, Multimedia and Hypermedia, Vol. 9 No.2 2000, AACE, Charlottesville.

Coxwell, Joshua,1996,"Workshop Biology Course ", http://biology.uoregon.edu/Biology_WWW/Workshop_Biol/wb.html, (accessed June 2000)

Dessaix, Robert,1997," Hail the Tall Poppy", Sydney Morning Herald, Saturday May 10 1997, Features pp. 7

Drucker, P. "Putting More Now Into The Internet", Forbes Magazine, http://www.forbes.com/forbesglobal/00/0515/0310092a.htm

Gilliver, R.S., Randall, B.J., Yang Ming Pok, 1999, "The Orbicular Model - Cognitive Learning in Cyberspace", Journal of Educational, Multimedia and Hypermedia, Vol. 8 No.4 1999, AACE, Charlottesville.

Giraudo, Amy, 2000, "Emerging eLearning market to exceed $20 billion by 2005", Press Release of SRI Consulting, (http://future.sri.com/BICPR/LOD-PR.eLearn0500.shtml)

Jones, D, 2000, "Are MBAs slowly going out of style?", USA Today, 2000-05-23, http://www.usatoday.com/money/bcovtue.htm"

Levine, R., Locke, C., Searls, D., Weinberger, D., 1999, "The Cluetrain Manifesto", Perseus Books, Cambridge, Ma.

McKey, Paul, 1997, "The Development of the Online Educational Institute" Dissertation for Masters of Education, Southern Cross University, (http://www.redbean.com.au/articles/masters/thesis.html).

McKey, Paul, 1999, "The Total Student Experience", Keynote Presentation at ASCILITE 99, http://www.nexted.com/news/papers/TSE_McKey.html

Merrill, M. David, 1998, "Instructional Transaction Theory (ITT): Instructional Design Based on Knowledge Objects". In press. Chapter 17 in C. M. Reigeluth (Ed.), Instructional-Design Theories and Models: A New Paradigm of Instructional Theory. Mahwah, NJ: Lawrence Erlbaum Associates. On-line at http://itech1.coe.uga.edu/itforum/paper22/paper22a.html.

Reeves, Tom, 1992, "Effective Dimensions of Interactive learning Systems", Proceedings of Information Technology for Training and Education Conference, Sept 29th - 2nd Oct., University of Queensland: Brisbane.

Reid, Ian, "Online Strategies in Higher Education" Paper presented at Ausweb 99 http://ausweb.scu.edu.au/aw99/papers/reid/

Robson, Robby, 1999, "Object-oriented Instructional Design and Web-based Authoring", available online at http://robby.orst.edu/papers/objectoriented.html

Sorley, Scott, 2000 (personal demonstration of current R&D work).

Tiwana, Amrit, 2000, "The Knowledge management Toolkit", Prentice Hall PTR, Upper Saddle River, New Jersey.

US Department of Defence (DoD), 2000 "Shareable Courseware Object Reference Model", available at http://www.adlnet.org

Standards Bodies: