If These Walls Could Talk: Unlocking the Interoperability of SCORM, S1000D, and DITA

By Diane Wieland, DCL News

While at first this topic might sound a bit esoteric, it is important to anyone dealing with technical information and training people to use it. S1000D is a standard defining task-based documents which describe how to perform a particular procedure. DITA defines topic-based documents which are more general and descriptive. SCORM defines educational materials.

In preparing educational materials the need to incorporate task-based informationand topic-based information often occurs. It can be frustrating and costly to constantly rework materials among the different formats. Perhaps more important is the time lag in getting information into the educational process. Frequently the engineering information is the first to be updated, and there's a big a lag until it gets into the training process. The interoperability of these standards would go a long way to smoothing the process among these three inter-related areas.

"We are looking at DITA coming into this mix, not as a competing standard, but an integrated piece"

Tim Tate, Director JTPC

When organizations examine their content creation and management processes,they often find adopting a technical content standard is the first step toward reducing inefficiencies. Content standards provide the consistency often lacking in proprietary content creation approaches and make it possible for content types to be shared, sorted, ranked, rated, and mixed and matched to create a variety of information products. These standards support automation, reduce ambiguity, and make content more user-friendly.

As organizations around the globe begin examining the benefits of content standards, they often find that moving to a standard approach involves adopting a variety of standards and finding ways to make them work together to achieve critical business goals. Finding ways to cooperatively use widely accepted standards such as [URL=http://www.s1000d.org/]S1000D[/URL],[URL=http://www-128.ibm.com/developerworks/xml/library/x-dita1/]DITA[/URL], and [URL=http://www.adlnet.gov/Scorm/index.cfm]SCORM [/URL]can help fill the gaps left by one or the other and offer frameworks that minimizes the impact of change.

In this exclusive DCL News article, Diane Wieland explores how some experts are saying the walls of DITA, S1000D, and SCORM could talk.

What started as a whisper among content management professionals who support interoperability between DITA and S1000D, is growing into an open conversation about how these two standards could work together to provide more content sharing and greater reusability. In fact, it will be a topic of an upcoming S1000D-DITA subcommittee summit hosted by [URL=http://www.adlnet.gov/]Advanced Distributed Learning[/URL] in Alexandria, Virginia, April 10-11.

There has been some interest in bridging the gap between S1000D and DITA in the aerospace industry, but the latest spark of interest in S1000D-DITA interoperability seemed to emerge in two places at once. The [URL=http://www.s1000dscorm.org/home/]S1000D-SCORM Testbed[/URL], and private sector experts appear to have had simultaneous "Eureka! moments" that have led to the first ever [URL=http://adlcommunity.net/course/view.php?id=24]S1000D-DITA Training and Learning Subcommittee Summit[/URL] and an OASIS sponsored [URL=http://lists.oasis-open.org/archives/dita-s1000d-discuss/200702/msg00000.html]DITA-S1000D discussion group[/URL]. (Notice the switch in acronyms?)

SCORM (Shareable Content Object Reference Model) is an XML-based framework used to define and access information about learning objects so they can be easily shared among different learning management systems. The S1000D-SCORM Testbed is the responsibility of Tim Tate, Director of the Advanced Distributed Learning Job Performance Technology Center. As director of JPTC, his task is to look for ways public and private sector organizations can improve job performance by sharing technology research, development, implementation and evaluations.

"Our commitment is to be the 'honest broker,' Tate says. "When asked, we will provide our honest opinion about what is the best solution. The ultimate result for us, because we are in personnel and readiness, is people doing their jobs correctly, efficiently, and effectively."

In order to promote workforce readiness, the JPTC looked for ways to improve on-the-job performance. One way to improve performance is to assure e-learning materials and technical documentation are in synch. Because e-learning material is often created separately from technical documentation, inconsistencies can occur. It's also increasingly difficult to keep both sets of content up-to-date. JPTC wanted to solve this problem by electronically tying the two together. As a result the S1000D-SCORM Testbed Project was born.

"We think it's very important to reconcile these parallel approaches."

Eric Severson CTO, Flatiron Solutions

The Testbed project spent the next four years developing the integration of S1000D technical materials and [URL=http://en.wikipedia.org/wiki/SCORM]SCORM[/URL] instructional solutions. The Testbed project wrapped up its first phase, Testbed One, demonstrating three important things related to their mission. First, it demonstrated how learning materials could be created in S1000D (as Data Modules (DM)) and delivered in SCORM. Secondly, it showed that materials could be created in SCORM using S1000D DMs as auxiliary resources, bringing the two together from a learning-centric approach. In both those scenarios, it was assumed that S1000D technical documentation was the most-trusted source of information.

However, the final use case for Testbed One assumed that the most-trusted content could be from either the technical documentation or the e-learning resources, a realistic scenario in some organizations. So the concept of "proto content" was explored. [URL=http://www.s1000dscorm.org/usecases/Proto%20Content%20Source.asp#_Toc151969803]Proto content[/URL] describes the most-trusted source objects, "raw materials (or raw digital assets) which developers of derivative content could use in the creation process."

The project created a Common Source Data Base (CSDB) of proto content and populated it with XML objects from both SCORM (Sharable Content Objects, or SCOs) and S1000D (DMs). Data Conversion Laboratory participated in the project by converting legacy documentation into S1000D to deliver whatever output was needed, e-learning materials or technical manuals. The use case found that, indeed, interactive technical materials could be created using DMs and SCOs, and web-based learning could also be created using DMs and SCOs from the database.

"But the CSDB doesn't have to be limited to something like S1000D because what we are using is an XML-based repository of data," says Tate. "It can be a logo, 3-D graphics, simulations, or game scenarios." The project wanted to identify the most granular level of information. "For us it is a XML representation at a granular level. S1000D's nuclear granular level is a DM. SCORM's smallest granular level is a SCO."

However, DMs and SCOs are relatively large chunks of information with somewhat limited reusability. If DMs and SCOs can be broken down further, why not do that to increase reusability? "To keep them in the database and keep them updated you have to tag metadata to them," says Tate. "A lot of metadata tagging processes are still done manually, and that creates expense."

In that case why not use something like the [URL=http://dita.xml.org/introduction]Darwin Information Typing Architecture[/URL] (DITA) to organize DMs and SCOs into smaller pieces of content data that can be used by both SCORM and S1000D? One reason may simply be because S1000D and DITA have always been thought of as competing standards. However, several thought leaders may be changing that.

"We are looking at DITA coming into this mix, not as a competing standard, but an integrated piece," says Tate. "That's why we are bringing together the DITA subcommittees and the S1000D subcommittees so they can possibly work together. We are hoping that something under our Testbed Two, or other prototypes we are working, can actually use DITA and see where we can go with that."

Joe Gollner, Vice President e-Publishing Solutions for Silo International happens to think DITA offers some very compelling contributions to the standards integration mix. "DITA fills some very noticeable gaps in S1000D. Chief among these contributions are an extensibility framework that could be used to more actively tailor the S1000D schemas and reference applications to meet individual project requirements more cost-effectively, a set of very simple, and accordingly very functional, content building blocks; and a supporting set of reference documentation specifically oriented to helping authors create reusable content."

Currently there are at least two theories on how DITA and S1000D could work together. One idea is that DITA could be used as a "umbrella" standard that includes S1000D data module types as topics, when highly specialized types of content are needed (such as Illustrated Parts Data or Wiring Data), but also include other topics that can be pulled into document content as well. Or as Scott Hudson, leader of the OASIS DITA-S1000D discussion group puts it, "to graft S1000D type modules onto the DITA type hierarchy."

Hudson is a consultant for Flatiron Solutions, Inc. "My hope in starting the discussion group was to provide a public forum for the two standards groups to discuss the potential for interoperability. Based on feedback from that list, I anticipate that we will form a DITA subcommittee for S1000D specialization."

Chief Technical Officer for Flatiron, Eric Severson, describes the two standards this way. "DITA and S1000D both represent state-of-the-art techniques for managing XML information as a set of standalone, reusable modules that can be flexibly re-combined into PDF, web, Help and other output formats. While similar in concept, however, they come from different backgrounds and points of view.

"What Scott (Hudson) was looking at was 'I wonder if we could actually take S1000D and make it formally a variation of DITA and reconcile the two ideas.' So it actually is possible to think about DITA as having a specialized topic type that happens to relate to the S1000D topics. We (at Flatirons) think it's very important to reconcile these parallel approaches, and are therefore spearheading an OASIS technical subcommittee to attempt to describe S1000D data modules as formal DITA specializations."

Another idea gaining support is the possibility of using DITA as an antecedent, or precursor, to S1000D in order to create smaller bits of reusable information that can be used to create S1000D data modules, and at the same time be used for things like customer support, e-learning, parts catalogs, etc.

Severson agrees, "I think using DITA as a source format for this content makes a lot of sense, even though S1000D will be used as a key part of the delivery environment. DITA's topic orientation provides a more granular ability to re-use content at the source level, and it has the flexibility to support a superset of related information, part of which is used in manuals, part of which is used only in an e-Learning context."

It remains to be seen which-if either-of these theories could end up integrating DITA and S1000D. And because DITA already works well with SCORM, and the Testbed project has found ways for S1000D to work with SCORM (in theory at least) it leads to the possibility that all three of these standards could soon be working together to build the best sources of shareable, dynamic content for technical and learning materials.

DCLNews Editorial
March 2007

原文。