Food

Standards and Specifications-A Primer

Description
Standards and Specifications-A Primer
Categories
Published
of 6
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
Related Documents
Share
Transcript
  Standards and Specifications - A Primer    What are Standards and Specifications?  Over and over, people creating computer based resources (especially for learning) have suffered as technology evolves and their resources become obsolete - think HyperCard, the Doomsday project etc. Determined that this should not happen again, various interested parties have been working to define a set of 'standards' aimed at facilitating the reuse of learning materials at all levels. Various groupings have developed, but by far the most important is IMS (Instructional Management Specification) Global Learning Consortium. Indeed IMS are so central to the development of educational technology standards that their name has become a byword for the whole field. Who is Involved?  IMS is a body set up by a number of parties within Government, Education, and the Commercial Sector. Although based in the US, their membership is worldwide, with the UK being especially well represented - with 15/54 members at present. CETIS (the Centre for Educational Technology Interoperability Standards) are funded by JISC and represent UK higher Education within IMS. Other UK members include: BECTA, UfI, , Learning & Teaching Scotland and UKeU. Corporate members include Microsoft, Sun, Blackboard, and Fretwell Downing etc. Why the Interest at this point in time?  The IMS Group had one of their regular meetings in Sheffield recently (September 2002). As part of this, a couple of days of public events were planned - and these provided an ideal opportunity for disseminating the current 'state of play' to the practitioner community. In addition, the work done by IMS in the last 18 months is finally beginning to reach a level of maturity where it starts to affect the way people work - more and more you will find people using words and phases like 'Interoperability', and 'IMS Compliant etc'. The JISC has funded two major initiatives over the last 6 months: •   X4L - Exchange for Learning - repurposing existing JISC-funded content for re-use. •   FAIR - Facilitating Access to Institutional Resources - allowing deposit and disclosure of institutional resources. These projects are just starting out now. Why are they called Specifications?   Although we often talk about 'standards', it is important to remember that IMS does not produce standards, only specifications. The difference between a standard and a specification is that a standard will have some sort of quality assurance associated with it - possibly in conjunction with an accreditation program. The following diagram demonstrates the process which occurs to take create a standard from a specification.    As you can see, IMS draws on the user community, technology developments and research areas to define and formulate specifications. This is done within working groups. Specifications go through private and public phases and work quickly. Once a specification has been released as a final version, it will start to be used by the user community and specific agencies such as ADL (The Advanced Distributed Learning Network - http://www.adlnet.org) who start the process of actually using the specifications - providing feedback to IMS for future development, and creating reference implementations that others can work with. Presently, the most widely used of these is SCORM (the Scalable Content Object Reference Model) which seeks to combine the MetaData and Content Packaging specifications with the specific aim of facilitating reuse of learning objects between Learning Management Systems (LMS) such as BlackBoard and WebCT. ADL are currently formulating an accreditation program for SCORM so that LMS vendors can certify that their product is SCORM compliant. A specification only becomes a Standard when one of the approved certification authorities, such as ISO, BSI, IEEE etc. get involved. Obviously, in this fast evolving market we are a long way from stable standards yet. What aspects of Learning Technology are covered?  IMS is at present working on a number of specifications that are at various different stages of completion. Almost every aspect of the educational process is covered - and inevitably, the simpler and more useful specifications (metadata and content packaging) have been finished first. The Metadata and Content Packaging specifications are now sufficiently stable to be used in earnest and these specifications have been/are being used by LearnDirect, BECTa, NLN etc. to create content repositories and describe resources. Most of the other specifications are some way behind, but progress is being made. SPECIFICATION   VERSION   COMMENT MetaData 1.2.1 MetaData effectively means data about data and this specification defines standard ways of describing learning resources and content - this includes conventions for describing educational level, usage restrictions etc. Content Packaging 1.1.2 The Content Packaging specification seeks to provide a standard way of organizing educational content. By providing standards on packaging content (along with metadata describing it) resources can be taken out of one system and easily placed into another. Enterprise 1.1 This specification covers the management of learning systems and provides for management of learners and interchange of data between for  example certification authorities, learner management systems and college MIS's. Learner Information Packaging 1.0 This specification attempts to standardise the way information about a learner is stored. Question and Test Interoperability 1.2 This specification seeks to allow the interoperability of questions and tests between different assessment engines. This specification now supports Question Banks, Random Tests, Multimedia, Selection and Ordering, Scoring, Reporting and Multiple Output Formats (e.g. PDF and web). Simple Sequencing Soon to be 1.0 Sequencing refers to the aggregation of learning content and the specification of routes through that material. The specification supports score roll up, flow choice, adaptive learning, and works by extending content packaging information. Digital Repositories 1.0 by end of 2002 Based on z39.50 and supporting search, retrieval and updating of repositories. Learning Design  Approved for Public Draft Defines roles and participants, resources, and, importantly, the Instructional Design approach, again integrating through Content Packaging. Reusable Competencies 0.01 Imminent This specification defines a skill record for a learner.  Accessibility 1.0 Public draft of Guidelines The accessibility specification is being developed to manage appropriate access to resources through the use of assistive technologies such as screen readers and text to speech engines. The accessibility specification is being extended to support user preferences as described above. What is the Ultimate Vision?  The ultimate aim for these specifications is that the whole area of providing education to learners through electronic delivery works seamlessly. In particular, these efforts seek to ensure that: •   Content created for one system can be re-used in another (cf effort spent formatting content for BlackBoard or WebCT delivery). This is effectively horizontal integration. •   Content created for one syllabus can be used in another (how different is physics, mechanics, engineering etc.) •   Smart content can exist - allowing an LMS to specify initial parameters for a simulation or question choice for an assessment. •   Information produced by one system can be used by another (scores generated by a CAA system fed into an LMS, and from there to the college MIS, or even the government records office). This is Vertical Integration of Systems. What are the Barriers?  The view that educational technology standards can provide an answer to all our  problems is utopian. A number of barriers stand in the way: •   Creating this utopian vision of true absolute interoperability will involve enormous investment from all quarters. And with the technologies evolving all the time, there will be inevitable problems over what is considered the stable state - 'why standardize on this simple model when we can now do this'. Standards will inevitably become the lowest common denominator. •   It is unrealistic to expect perfect adoption of standards across international boundaries - educational philosophies, as well as the infrastructures which support them are sufficiently different between (for example) the US and the UK that a single solution simply will not work. Local standardisation is a much more likely outcome. •   In the short term at least, such massive investment will seem like overkill to many small and independent producers. This is especially true in the small business market where there are established trusted partnerships of content providers and clients. •   There is a danger that if we go too far too fast, then academics will rebel - and reject the standards as being constrictive rather than enabling. One specific area of tension is Learning Design - where there is an attempt to model pedagogic styles and standardise lesson plans - many teachers would argue that standardisation at this level is nonsense. •   Lack of accreditation. Those who are trying to use these specifications and standards in the real world are getting increasingly frustrated by the lack of proper accreditation and conformance authorities - more and more, vendors are saying that they are SCORM compliant or IMS compliant when they are plainly not (you can't ever be compliant with an IMS specification, and SCORM accreditation is not set up yet). •   One of the most significant barriers to all of this is commercial pressure. For a company with an established commercial product (e.g. an LMS) there is very little to be gained from changing your product to make it easier for a client to drop it in favour of a competitor. For instance, whilst some LMS vendors might claim to be compliant with standards, they will try and enhance the extras such that moving to another LMS would result in a significant loss of user experience. Indeed one of the significant by-products of this is that there'll be lots of open source tools and reference implementing created as by-products. What are the Drivers?  There are a number of significant drivers for change: •    All new public money spent on commissioning courseware will specify adherence to standards at some level - so non-compliant systems will be frozen out, quickly. This will be evident first in schools and colleges where central provision of LMS systems is common. Universities will also evolve this way as strategic planning sees standards compliant systems as a route to efficiency. Other factors include increased overlap with the FE sector, award conditions from bodies such as JISC, and a need for interoperability with external resources. Although based in the US, the UK is a powerful force in IMS - providing almost a third of all members. There is also high-level commitment both from the education sector (JISC, UfI, UkeU, NHSu) and government - see the e-GIF (e-Government Interoperability Framework). •   Many commercial suppliers have seen standards as providing a potential commercial advantage and have decided to become involved so that they can actually drive the standards process - can studios (based in Sheffield) and Giunti Interactive Labs (based in Genoa) are two examples of  particularly active companies. Involvement in IMS (as contributing members) is being seen as a quality criteria by those involved in choosing LMS systems. (cf talk by Fred Beshears, University of Berkeley at Sheffield.) •   Legislation on accessibility which will require all course materials to be made available in a variety of formats and will ensure that maximum flexibility is built into any course. Accessibility demands interoperability and reuse. What Next?  Now that MetaData and Content Packaging are mature specifications, focus within the community has moved on to tackling some of the more complex issues - specifically: •   sequencing (allowing the content to contain information about course pre-requisites, suggested routes through materials etc.), •   learning design (thinking about specific pedagogies and how they can be supported - this would start to enable interoperability for discussion forums I think) •   enterprise (to allow integration of disparate systems) •   accessibility/learner information profiles (enhancing the user experience) and •   assessment - (how to share questions and tests between CAA systems). Question and Test Interoperability  On the face of it, the QTI specification seems to be all things to all men - it is split into two sections - basically covering Questions (ASI Selection and Ordering) and Results (Results Reporting). To date, no one has really done much work on the Results Reporting aspect of the specification, more work has been done with the  ASI Selection and Ordering: ASI stands for Assessment, Section, Item - such that a question is an item, a group of questions a section and a whole test an assessment. Absolutely at the bottom of the pile is QTI Lite, which is a basic implementation of just question types and the simplest marking possible to allow exchange of questions between systems. - Many individuals and vendors have implemented QTI Lite. SCORM (currently at v1.2) plans to incorporate the QTI specification into its v1.4 release, due out in late 2003.  ASI Selection and Ordering covers many aspects of test (assessment) design - including various ways in which questions can be chosen - e.g. all, all except, random x from y, always selected, chosen by topic, conditional selection, and repetition allowed/precluded. In addition, item order can be controlled along various lines - fixed, random, random within a group, etc. Items have to contain information not just about the question - question type, all choices, but also about how to mark it and how it should be rendered (displayed), including alternative rendering arrangements for accessibility purposes. A wide range of item types are supported - MCQ, t/f, multiple response, hotspots, fill in blank, slider, drag and drop, order, match etc. but there is no inbuilt support for randomization of variables, steps, and other complex behaviours. This type of functionality would have to be implemented as extensions to the core specification - possibly with inclusion in the specification at a later date. Some more sophisticated marking schemes aren't yet supported by the QTI specification. The CETIS AssessmentSIG meets up 4-5 times a year to discuss the issues surrounding QTI. Within the group there is a direct link to the QTI specification through Steve Ley (UCLES) and a number of active developers keen to share experiences, and solutions. There is a mailing list at: qti-
Search
Tags
Related Search
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks
SAVE OUR EARTH

We need your sign to support Project to invent "SMART AND CONTROLLABLE REFLECTIVE BALLOONS" to cover the Sun and Save Our Earth.

More details...

Sign Now!

We are very appreciated for your Prompt Action!

x