Intelligent content has been a buzzword for quite a while now. It is a promise to deliver all the content you need at the right time. For example, you’re a service technician going to maintain that PrettyBigMachine, you’ll need the maintenance instructions for all the components that are up for scheduled maintenance, plus a list of required tools and the time you will need for your work.
While content delivery has always been at the heart of technical communication, the sheer amount of information and complexity of the subject, as well as the capability of IT systems, has increased over the recent years. The demand and expectations of users have to be met by more sophisticated publishing and delivery technologies. Content delivery portals are the new smart websites that integrate data from multiple sources and allow searching, browsing, and delivering of content based on metadata. Metadata makes intelligent information.
But how to bring it all together? To name just a few, how to integrate supplier documentation, company-internal data sheets, technical documentation, marketing material, and support hotline troubleshooting information. Even if all this information came with metadata, it would probably be everything but homogenous. That’s where the International Standard for Intelligent Information Request and Delivery iiRDS comes into play. It is an initiative to standardize metadata for technical communication and a packaging format for archiving and exchange. It’s a potential starting point for developing your own taxonomies and ontologies.
iiRDS was initially developed by a working group of the German Association for Technical Communication tekom. An industry-driven consortium took over in 2018 and is since then responsible for iiRDS. The specification of the metadata vocabulary and package format are published under a creative commons license (CC BY-ND 4.0) and are freely available after registration on https://iirds.org.
iiRDS package and metadata
The iiRDS packaging format specifies a ZIP container with a folder structure and two flavors. Unrestricted iiRDS packages may contain all content formats while iiRDS/A allows only some file formats, for example only XHTML5 and PDF/A for textual content files. Next to the content files, all iiRDS packages contain a metadata.rdf file with metadata about the content. The metadata is based on a standardized iiRDS metadata vocabulary in RDF and RDF Schema, the W3C standards of the semantic web.
The iiRDS metadata vocabulary provides a core vocabulary and domain-specific extensions, currently for machinery and software use cases. The centerpiece of the vocabulary is the iiRDS Information Unit with its subclasses Document, Topic, Fragment, and Package. An iiRDS Information Unit is a set of metadata that picks from a pool of product-related concepts, information subjects, and functional metadata. For example, a topic could be relevant for a particular component, provide conceptual information about component safety, and require a reader with a specific skill level. Additionally, there’s vocabulary for administrative purposes. The metadata set of an iiRDS Information Unit references a file in the iiRDS package. Also, parts within a file can be referenced, for example, a page in a PDF can be an iiRDS Topic with a maintenance task.
iiRDS aims to be a standard vocabulary for taxonomies and ontologies describing user information. As with each standard, it cannot cover all specific needs. That is why it does not contain a vocabulary for product features, product components, or product variants, for example. These are usually company- or industry-specific.
The nature of RDF and the iiRDS metadata vocabulary is its modular expandability. Users can add their proprietary vocabulary to iiRDS by adding their vocabulary to existing iiRDS concepts. For example, MyRatherSmallComponent would be a member of the iiRDS class component. An iiRDS capable system can then list MyRatherSmallComponent on a faceted search page in the content delivery portal.
DITA and iiRDS
While iiRDS can enrich monolithic chunks of technical documentation, it profits from modular content. Smaller chunks of information enriched with metadata allow more granular and precise content delivery. The authors feel that DITA is a particularly good match to iiRDS. An iiRDS package can act as a metadata wrapper around an assembly of native DITA files.
Currently, there are two iiRDS plugins available for the DITA-OT. The first plugin allows to author content with iiRDS metadata. With specialized attributes and data elements, authors can add metadata to topics, maps, and fragments. The plugin specializes document type definitions in Relax NG. The second iiRDS plugin adds a transformation type and generates an iiRDS package with a metadata file in RDF. Both plugins are available on https://github.com/parsempolis/dita-iiRDS.
The following figure illustrates a DITA-based workflow for generating iiRDS packages. Once the iiRDS packages are created, they can be imported and processed by a content delivery application.
The iiRDS Consortium
As iiRDS was a tekom initiative, it is quite present on the German market. Several German CCMS vendors are members of the iiRDS consortium and actively participate in the development of the standard. The first projects that productively use iiRDS metadata for dynamic content delivery have been presented, for example, https://www.dosco.de/1/references/iirds-bei-siemens/
You can become part of iiRDS, too. Get involved, join the iiRDS Consortium, provide feedback about the standard, or contribute to the DITA-OT plugins.