What’s SMIL and why should we use it?

What's SMIL and why should we use it?OK, here comes a geeky article about elearning and data management best practices. Although the issues are very technical in nature, they require the support of well-informed management so that the most appropriate decisions can be made. Understanding these issues from the outset can save you or your organisation a lot of time, effort and going up blind alleys in the not-too-distant future.

What are the issues? If it isn’t broken, why fix it?

Currently, most elearning developers use so called “rapid elearning development tools”, e.g. Microsoft PowerPoint + Adobe Presenter, Adobe Captivate, Techsmith Camtasia, Articulate and Raptivity,  to create and publish content. They present quick and easy solutions to elearning novices, enabling them to create and deploy multimedia rich, highly interactive learning content on the web without learning a great deal of technical skills or knowledge. However, the vast majority of these tools publish content that is “single purpose” or “single use” and elearning content developers may end up spending a lot of time and effort on creating very impressive content that has little effect on learning outcomes and, in the long run, may make elearning courses difficult and time consuming to manage, maintain and develop. Here are some of the drawbacks:

Proprietary dependencies (lock-in)

Rapid elearning development tools create source files that only their software can read and edit. Often, they’re not forwardly compatible meaning that if you want to edit files from a newer version of the software, you’ll have to buy an upgrade. If you want to edit the source files in other tools, it’s usually a breach of copyright and the End User Licence Agreement (EULA) to do so. Additionally, some rapid elearning development tools publish content in out-dated versions of Flash leading to some unexpected problems for developers. Vendors rely on this to keep users dependent on their software and to make it as difficult as possible to migrate learning content development away from their tools.

Inflexible learning content

Published presentations from rapid elearning development tools generally take the form of single or multiple Flash (.swf) files that present the learning content in predetermined sequences. All the text, images, audio, video and animations are locked away inside the Flash file(s). If you want to change the order of a sequence, you have to go back to the source files and re-author and re-publish the new sequence. There’s no access to the published learning content from other software than can re-use and re-purpose it and Learning Management Systems (LMS) cannot allow teachers and learners to access media from the files and use it in presentations of their own or in discussion forums, wikis, glossaries, etc. Now that group learning (AKA social learning or Social Constructivism) is becoming increasingly popular among learners and teachers, this is a severe drawback.

Narrow range of uses

Pedagogically, presentations, slide shows, simulations, etc. have a narrow range of uses. Regular, old-fashioned HTML web pages often have comparable learning outcomes to rapid elearning tool produced learning interactions with video and multimedia. Furthermore, with all the multimedia, audio, video and animation options available at your fingertips, it’s easy to get carried away and to include too much media and too many different types media simultaneously resulting in cognitive overload and a subsequent drop in learning efficacy.

Inappropriate use of quizzes

Most rapid elearning development tools recommend and encourage the use of quizzes before, during and after presentations. Indeed, they pride themselves on providing the best possible editors, training and support for learning content developers to add quizzes to their presentations. However, according to the US Department of Education’s Evaluation of Evidence-Based Practices in Online Learning,

Elements such as video or online quizzes do not appear to influence the amount that students learn in online classes. The research does not support the use of some frequently recommended online learning practices. Inclusion of more media in an online application does not appear to enhance learning. The practice of providing online quizzes does not seem to be more effective than other tactics such as assigning homework.

Source: ED.gov (PDF, page 18)

Can your Learning Management System (LMS) do it?

Most modern LMS’ have well-developed and designed presentation authoring modules built in. They almost all have quiz and exam authoring modules. The results can be comparable to rapid elearning development tools. It’s worth spending some time with your LMS and seeing what it can do. While LMS’ don’t typically have the best support for multimedia, there are a lot of advantages to this option:

  • Learning resources can be edited and created immediately online.
  • No extra software or development tools are necessary.
  • LMS’ are usually database driven which means indexing, searching and maintaining libraries of learning resources in them is powerful, flexible and simple.
  • Some LMS’ have text filters that can automatically add links and tags to learning content and learner generated content to make make them more closely integrated, such as glossaries, wikis and discussion forums.
  • Some LMS’ provide easy to use tools for embedding multimedia into presentations, quizzes, glossaries, etc.

I’m not advocating abandoning rapid elearning development tools altogether (I think they’re very appropriate for one-off, highly particular presentations and simulations) but I think it’s important to understand their limitations and that, in many cases, there are more appropriate approaches to creating, maintaining and managing learning content.

Another option: SMIL XML

The Synchronized Multimedia Integration Language (SMIL), is a W3C.org recommended XML markup language for describing multimedia presentations. It defines markup for timing, layout, animations, visual transitions, and media embedding, among other things. SMIL allows the presentation of media items such as text, images, video, and audio, as well as links to other SMIL presentations, and files from multiple web servers. SMIL markup is written in XML, and has similarities to HTML.

Source: Wikipedia.org

SMIL is currently most commonly used as a subtitle or text captioning format for online video, otherwise known as SMILText, TimedText or RealText, and for media play lists like those used with the JW Player and the Media Player module for Moodle but, as you’ll see in this article, it is capable of far more than that.

How does it work?

A SMIL XML file contains all the data necessary to organise a play list or learning interaction such as a PowerPoint style presentation or a multimedia quiz. Note that the main constituent parts of the learning interaction are kept separate; the multimedia files, the SMIL data files, any styling and the SMIL player. Software developers call this the Model-View-Controller* (MVC) design pattern which is used in almost all web software, such as Content Management Systems (CMS), such as WordPress, Joomla, Drupal and Mambo, and LMS’, such as Moodle, Sakai, ATutor and ILIAS. This means that each part of the multimedia, data, styling and player can be edited, substituted and recombined separately without “breaking” the learning interaction. Also multimedia, which tends to be costly and time consuming to produce, can be re-used and re-purposed easily for other learning interactions. For example, if all the learning interactions display an image, only one copy of that image needs to be stored on the LMS. If we want to change or update it, we only need to edit or replace this one copy and this will be reflected across all the learning interactions that use it, so there’s no need to go through the laborious task of editing and republishing tens, hundreds or maybe even thousands of files just to change one image, which is the case with typical rapid elearning SCORM packages.

SMIL diagram

* In the case of elearning MVC would be:

  • Model – SMIL XML files and multimedia files. Additionally, SMIL files often contain layout data.
  • View – Any styling, which could include colour schemes, fonts, graphics, backgrounds, logos and branding.
  • Controller – The software that manipulates the model and applies the styling to create presentations and other learning interactions.

Platform agnosticism

Perhaps one of the biggest advantages of using a web standard format, like SMIL XML, is that it’s “platform agnostic“, i.e. it isn’t limited to just one operating system, software platform, runtime or playback device. This means that you can develop applications in any language for any operating system or runtime to play SMIL presentations. Options include but are not limited to: Flash and Javascript (for web browsers), Adobe AIR and Java (for desktops), Android apps (mobile phones and tablets) and iOS apps (iPhone, iPod and iPad). The following media players also support SMIL playback: Apple’s Quicktime player, Windows Media Player (WMP) and RealNetworks RealPlayer.

Flexible and adaptable

In addition to playing SMIL files from start to finish, as slide show presentations, it’s also possible to develop custom applications that can use the presentation data to create new activities, for example games, quizzes and reference aids. I develop Flash Multimedia Interactive Learning Applications (MILAs) that read SMIL files and use them to create a variety of learning interactions. With this approach it’s possible to create an almost unlimited range of activity types to your exact specifications.

When should we use SMIL?

SMIL XML is a potential replacement for presentations typically produced by using one of the many rapid elearning development tools. If you find yourself copying and pasting layouts, content, templates, etc. from one presentation to the next or you find yourself doing very repetitive tasks frequently, then that’s a good case for considering adopting a SMIL based approach. Typical rapid elearning development tools that SMIL can replace are:

* OpenOffice.org, the free open source alternative to Microsoft Office, can publish presentations directly to Flash. Additionally, it’s compatible with MS Office documents so it’s one of the cheaper and easier ways to convert old legacy presentations to Flash for web deployment. I previously wrote an article, Open source for elearning, which lists alternatives to commercial, proprietary software.

Why should we use SMIL?

The advantages

  • Open file format – Your typos, spelling mistakes, wrong images, audio or video, etc. can be corrected in seconds with a simple text or XML editor. (Moodle 1.9 allows you to edit SMIL XML files in the course files repository directly online.)
  • Media files are stored separately – Images, animations, audio and video can be updated without having to re-author and re-publish elearning packages. Also screen recordings in either video or Flash are separate from the main presentation structure and can be re-recorded without completely rewriting the whole project.
  • All the data and media is available at a “granular” level so it can be manipulated and re-purposed with software to create an almost infinite variety of learning resources.
  • Web browsers cache media files and, instead of unnecessarily downloading them multiple times, taking up bandwidth and time, they are re-used from the cache. It’s faster and more efficient.
  • Video file formats preserved – As long as video file formats are supported, they are played directly in their original form. This avoids the inevitable loss in quality caused when rapid elearning tools transcode video files imported into them.
  • Presentations can share files and data – It’s possible to re-use media files such as video saving you server storage space and reducing internet bandwidth usage.
  • SMIL is “platform agnostic” meaning that you can develop/use SMIL player applications for use on any operating system or runtime.
  • Course/Site wide configuration – Groups of presentations can be configured using a single, shared file and changes to courses or even whole sites can be made easily. With rapid elearning development tools, it’s necessary to edit and re-publish every single presentation.
  • Smaller file sizes – Most rapid elearning development tools typically produce unnecessarily large files. A combination of SMIL content files and software SMIL players typically produces smaller, optimal file sizes, therefore learning resources download and start faster.
  • No problems with rapid elearning development software versions – You can update image, animation and video production software without worrying if it’ll be compatible with previous or later versions. Additionally, you’re not tied to using any particular software to maintain legacy presentations.

In short, you get a leaner, meaner, faster, more flexible, more editable and ultimately more efficient way of producing elearning presentations and learning resources.

The disadvantages

  • Initial cost of developing a SMIL player (Almost no free or open source web based SMIL players available so please let me know if you know of any)
  • Knowledge of SMIL XML schemata for authoring and editing is required
  • Generally requires some specialised, skilled IT support

Ultimately, the choice is determined by the number of presentations you’re likely to deploy and maintain on your elearning courses. If it’s a small number, then the software development and inconvenience of training or hiring developers with the necessary skills and knowledge outweigh the benefits. However, elearning and blended learning programmes can quickly accumulate large numbers of multimedia learning interactions, which can become difficult and time consuming to manage and maintain and subsequently place unreasonable restrictions on your curriculum development programmes.

More information about SMIL

Making e-learning resources relevant

EditE-learning resources have a number of advantages over book and paper based resources. In this article, I’m going to write about one in particular: the importance of being able to edit the resources that you use with learners. This article focuses directly on resources for learning English as a Foreign Language (EFL) and English as a Second Language (ESL) as they provide very clear examples of why it is important.

Why would you want to edit learning resources?

Apart from obvious reasons, such as changes to the syllabus, correcting typos, correcting wrong answers and other such errors, it’s a great advantage to be able to edit resources to bring them into line with other resources to create more coherent, better targeted courses.

Frequently, e-learning resources are used in combination with classroom, task-based and other types of learning activities, otherwise known as blended learning. Under these circumstances, it’s desirable to limit the scope of what learners are exposed to in order to help them focus on the core curriculum. In the case of EFL/ESL lessons, you’ll most probably have more success presenting and practising the same 8-12 new words and expressions in the classroom and on-line than if you present one set of vocabulary in class and another, different set on-line.

Good resources, inappropriate content

There are plenty of very good and very effective photocopiable resources and self-study materials for EFL/ESL learners to learn and practise with. The difficulty arises when we try to construct a coherent, well focused lesson or course with a core syllabus using these resources. For example, if a lesson covers comparative and superlative adjectives and transport and travel vocabulary but the photocopiable resources for adjectives cover lifestyle vocabulary, and the resources for transport and travel vocabulary cover the Present Perfect, then we’ve just doubled the amount of new language being simultaneously presented and practised. For many students, having to grapple with two lexical areas of vocabulary and two areas of structure and use at the same time can be overwhelming and confusing.

Lesson objectives: Grammar = comparative & superlative adjectives, Vocabulary = transport & travel

  • Resource #1: Grammar = comparative & superlative adjectives, Vocabulary = lifestyle & leisure
  • Resource #2: Grammar = Present Perfect simple, Vocabulary = transport & travel

Inconsistencies across different resources

Futhermore, for any language that we expect the learner to remember and use correctly, a certain amount of practice and repetition is necessary so we need a series of different types of activity to recycle the target language. Photocopiable resources and self-study materials tend to cover each language point once, therefore  learners usually need more resources from other sources. If we use a photocopiable resource in one instance to learn and practise vocabulary in a particular lexical area, such as transport and travel for Intermediate students, the specific vocabulary can vary from resource to resource since photocopiable resource authors often have different ideas about what the vocabulary for a particular lexical area for a particular level of English should be. This can result in learners being expected to learn more new vocabulary than is reasonable in a given length of time.

If we were to add yet more resources, the amount of language presented quickly becomes unmanageable. Currently, a lot of teachers spend a lot of time adapting or rewriting resources because they are inappropriate for the language points that they are teaching.

How does e-learning have an advantage over other resources?

Books and worksheets, our most familiar learning resources, are published in what’s known as hard copy, in other words, on paper. Unless we get heavily involved with Tippex, scissors, tape and glue, these resources are essentially uneditable. E-learning resources differ from this in that they are stored electronically and can be easily copied and edited, so having your texts and pictures in MS Word or Open Office Writer documents means that you can copy and paste any of them into a new resource and edit them and publish them as necessary. With Google Docs you have the added bonus of having the resources available in one easily accessible place.

To avoid copyright infringement issues, I recommend only using your own original content or Creative Commons content (Wikipedia, Flickr, Wikimedia, etc.) for this and properly acknowledging all authors and contributors.

Data driven learning resources and learning interactions

Electronic documents are somewhat more convenient than hard copy resources but, of course, computers and software are capable of making our lives much easier by generating learning resources that are multimedia and interactive and hosting them on the Internet where learners can have access to them anytime and from anywhere they have an Internet connection. Here, I’m talking about dynamic learning resources supported by learning management systems.

Dynamic simply means that the resources are generated by software automatically so all we need is the resource content, i.e. text, questions, answers, images, audio and sometimes video, in an appropriate format. An e-learning application, usually Flash or Java based, can read the learning resource content and create a unique learning resource. In the world of e-learning this is called a learning interaction. A learning interaction can be as simple as a page of text and images or it can be a highly interactive grammar game or self-correcting dictation.

So we have:

learning resource content (data) + e-learning application (software) = learning interaction (resource)

The beauty of this “data-driven” model is that we only have to author the learning resource content once. After that, we can use it, dynamically (i.e. automatically) to create any number of different learning interactions. All we have to do is substitute different e-learning applications that can read the learning resource content. E-learning applications can be word searches, dictations, multiple matching, multiple choice, true or false, short answer, gap-fill, written answer dialogues, spoken answer dialogues, listening activities, shadow reading activities, subtitled video, tests, etc. From just one set of learning resource content, it’s possible to create any number of learning interactions that focus on developing learners’ skills and knowledge. An added advantage of this model is that it encourages learning resource designers to recycle language more frequently thereby ensuring that learners get adequate practice. You can see an example of this on the demo course on my learning management system (login as a guest). On the demo course, the learning interactions “Word Search”, “Listen and Select” and “Look and Describe” all share the same learning content. If we edit the learning content, the changes are immediately reflected in all the learning interactions. We can also create a new set of learning content by copying and the original set and editing it to suit another purpose. Again, with this new set of learning content, we can dynamically create any number of different learning interactions.

This idea of keeping learning content and learning applications separate and combining them to create learning interactions can reduce course content development time dramatically. With a relatively small library of learning applications and learning content, it’s possible to provide hours of effective, engaging, varied, high-quality learning interactions in a very short space of time. It’s also very easy to copy and edit entire courses to re-purpose the learning content. So here we have an easy way to create lots of appropriate, well targeted and effective e-learning resources!

A quick note about SCORM

The Shareable Content Object Reference Model is an e-learning standard initially commissioned by the US military in 1999. It’s very different to dynamically driven e-learning resources. It’s not easily editable as it requires each learning interaction to be a self-contained package, i.e. they don’t share the same source of learning content, so in order to edit the learning content of a set of interactions, it’s necessary to edit and re-author all the interactions individually which is quite a complicated and involved task. I wrote an earlier article about the pros and cons of SCORM on this blog.

SCORM: The Pros and Cons


With e-learning becoming more widely adopted by schools, universities, governments and private companies for their educational and training needs, SCORM has become the de facto format. But should we really be adopting it?

What is SCORM?

SCORM, or the Sharable Content Object Reference Model, is a widely used web standard for e-learning interactions. It promises cross-platform compatibility and a homogenised approach to deploying e-learning resources but at what cost?

What are the advantages of SCORM?

  • A wide range of 3rd party support
  • You can buy ready made e-learning interactions
  • Supported by most Learning Management Systems or Virtual Learning Environments
  • Learning interactions can, in theory at least, be transferred from one LMS/VLE to another

Since SCORM was adopted by the US military in 2004, a number of agencies, consultancies and organisations have sprung up to offer SCORM resources and support, including software packages for authoring and packaging learning interactions.  Additionally, most Learning Management Systems or Virtual Learning Environments can deploy SCORM compliant learning interaction packages and if, at some point in the future, you decide to change to another Learning Management System, you can transfer them… well, in theory (see disadvantages).

What are the disadvantages of SCORM?

  • Limited selection of types of activities
  • Inherently insecure
  • Difficult to edit and correct ‘typos’
  • Very complicated format
  • ‘Compliance’ doesn’t guarantee that learning interactions will function correctly on your LMS/VLE
  • Learning interactions use a lot of Internet bandwidth and server storage space
  • Learning interactions take a long time to download

Firstly, I consider the main drawback to SCORM being it’s limited selection of types of activities. They are true or false, multiple choice, fill in the gap and multiple matching and a few variations of these types. As a teacher, I don’t expect my students to learn a great deal from such activities and I don’t think that they’re a very effective way of testing learners’ abilities or knowledge either.

Secondly, SCORM is inherently insecure. It requires the learner’s computer to store all the data related to a particular learning interaction, including the answers, usually in the web browser cache. It’s pretty easy to go and find the directory where those files are stored and look at them. On a Windows 2000 or Windows XP operating system, using Internet Explorer, you can change Tools… > Folder Options > View > Hidden files and folders > Show hidden files and folders and you’ll find the browser cache at C:Documents and SettingsusernameLocal SettingsTemporary Internet Files, and  Firefox provides a direct link to its browser cache.

Another security concern is that Adobe Flash abandoned support for SCORM after Flash MX 2004, also known as Flash (version) 6. It’s possible to author SCORM compliant e-learning applications in subsequent versions of Flash but they must be published in the MX 2004 legacy format. Likewise, the majority of software packages for authoring Flash e-learning applications for SCORM publish them in older, pre-Flash 9, formats. In the last few years, a number of high-level security threats have been identified in pre-Flash 9 files, allowing attackers to inject malicious software into users’ computers among other things. Read this article at Adobe.com for more details.

Another thing is, we’re all human beings and we all make mistakes. It’s frustrating when you think you’ve finished authoring a learning interaction and deploy it on a Learning Management System or Virtual Learning Environment and test it only to find that you’ve misspelled a few words, made a few typos or something similar (or your proofreaders have missed something). Most web-authoring tools allow you to easily go back and correct those inevitable mistakes by simply typing in the correct spelling, punctuation, etc. This isn’t so with SCORM. You’re required to re-author the learning interaction package and re-deploy it.

It also isn’t possible for SCORM packages, despite the word ‘sharable’ in the name (Sharable Content Object Reference Model) , to share resources. If you have images, audio recordings or videos that are used by all or several learning interactions, SCORM requires you to deploy copies of those media files in each and every package. This can result in a Learning Management System or Virtual Learning Environment using up many times more Internet bandwidth and server storage space than is necessary and also make updating those resources and long and tedious process. Basically, someone has to re-author every single learning interaction package. Perhaps that would be a job reserved for the “new employee” in the department?

In terms of how cost effective SCORM can be, I conducted an informal survey with administrators and teachers who use SCORM and I discovered that it was quite normal to deploy learning interaction packages of around 80MB – 100MB, the smallest being 15MB,  giving download times of anywhere up to 4 minutes before the learner can start the learning interaction (The average YouTube.com video is around 10MB and, of course, runs on Google’s impressive global data infrastructure – Watch the progress bar to see how long it takes to download). Additionally, Internet bandwidth and storage space on servers are expensive and should be kept as low as possible. If you multiply 80MB by the number packages on a course and by the number of students downloading them, you can see that bandwidth usage, in particular, can get pretty high. Such large file sizes and high bandwidth usage are, in my opinion, unnecessary and wasteful.

Finally, SCORM is incredibly complicated and difficult to understand. Here’s an extract from a user’s post on Moodle SCORM support forums:

“Right now I’m trying to figure out how to do basic SCORM compliancy (e.g. I have an authorware file with pictures of my cats and their names, buttons are there, one even has a quiz question)  How do I go from there to saying “Haha!  This is SCORM Compliant!  These are the steps to follow to ensure our product complies and we get paid!”  I’ve downloaded the tools off of adlnet, and I’m driving myself crazy at this point.”

And the reply:

“I feel your pain. I spent the last two years hoping for such a solution, and the bottom line is that the solution only exists by pouring over the documentation on the ADLnet.org site. Here’s the nutshell, but if you do a project for the navy with only this information, you will run into trouble. The best I can do for you is point you to ADLs “SCORM 2004 Conformance Requirements” pdf file and offer my consulting services. I lived on the ADL product downloads page while I was learning to develop for the SCORM.”

I think that with the amount of time and resources an organisation could spend on adopting SCORM and providing IT support, it might in some cases, be more cost effective to develop proprietry frameworks for authoring and deploying e-learning interactions. It would certainly be less time-consuming to use one of the many alternatives available.

To sum up…

So initially, SCORM seems to offer a great deal in terms of interoperability and support but there are less than obvious drawbacks that can increase costs in terms of bandwidth and IT support. Furthermore it can present serious security risks to your Internet/Intranet servers and users’ computers.

There are better technologies available which are more secure, cheaper to install and maintain and more efficient. For example, it’s possible to deploy all the activities specified by SCORM with Moodle’s native Quiz module which is database driven, which means it’s relatively easy to edit and update,  and relatively efficient. There are also faster, smarter, more efficient technologies on the horizon with Flash, the Flex Framework and Adobe AIR (all version 9 or above) being the platforms of choice for the e-learning applications of the future.