Managing Metadata for Responsive Web Sites with Subject Schemes
Documentation departments are under increasing pressure to produce more in less time. Customers expect customization and don’t understand why they can’t get it. Out of desperation, many documentation departments are looking to single sourcing and reusing content first, and, eventually, to customizing their documentation.
We recommend taking a very close look at your current process to determine the right tool for your department and its processes. What tool(s) are authors using to create content? What are their deliverables? Are you publishing to the Web? Are you printing manuals? What workflow applications are being used throughout the corporation, and will the tool’s workflow software integrate into the content management system? How are your deliverables produced?
POET Content Management Suite (CMS), which was released in May 1998, seeks to integrate into current documentation processes. Once you create a predefined publications specification, POET CMS automatically breaks XML or SGML documents into components based on their component tags and stores the components in a central repository. The repository is based on POET’s well-known, object-oriented database, where authors can access and reuse components.
There is no theoretical limit to the number of components that can be stored in the repository; hard disk space is the only limiting factor. Many content management systems rely on databases created by other vendors, but by using their own object-oriented database, POET has been able to make refinements that solve issues directly related to content management.
POET CMS reduces the learning curve for those authoring content by integrating the POET CMS repository into familiar authoring applications. Plug ins for Abortext Adept and Epic, SoftQuad Xmetal, and Adobe FrameMaker+SGML create an integrated interface between the authoring application and the POET CMS repository. Authors check out, reuse, modify and update, and check in components from a familiar interface. After the author checks the component back into the repository, POET CMS automatically creates a new version of the component (preserving previous versions) and indexes it to allow searching.
POET CMS includes a robust searching facility using Verity that allows authors to search by full text or metadata or to search for text within specified component tags. Metadata helps authors assign to components additional information that can be used to customize output targeted for specific audiences (novice, intermediate or expert) or specific releases.
Because the repository is XML/SGML based, authors can process the content into multiple outputs using the XML/SGML publishing tools already in place. Because POET CMS integrates into the scripting and workflow software, much of the output can be generated automatically to reduce costs. When used with third-party XML/SGML publishing tools that use style sheets, it is possible to deliver content dynamically to the Web without tedious HTML conversion.
At the completion of a project, POET CMS offers an Editions feature that “takes a snapshot” of the data for archiving purposes. No built-in facility currently exists for archiving components to free up disk space, but POET plans to build more archiving functionality into the product. It is currently possible to archive by making backups and automate the process using scripting.
Documentation departments often have staff off site who need access to the documentation. POET CMS offers plug ins for Cold Fusion combined with security features that allow authenticated users off-site access to the repository from any Web browser. Users can check out, modify, and check in components from anywhere in the world. Theoretically, dozens of people can work in the repository at the same time without affecting performance. Performance can be affected in the unlikely event of more than twenty people attempting to check components in or out simultaneously. However, since at any one time most authors working in the repository will be writing content rather than checking components in or out, this performance limitation should have little impact.
Authors can create “variants” of the referenced information to facilitate translation on the fly. Variants of the original component are linked to the original component. As authors create components, they can automatically let the translators know that the new component is ready. When the original component changes, workflow software can trigger an email to translators, notifying them that there is text in the repository to be translated. In this way, the translators can get started immediately rather than waiting for the entire completed document. Translation memory applications integrate into POET CMS, allowing a completely automated process when POET CMS is used with workflow software.
You can also use the Variants feature to customize documentation. Once you have produced a standard or baseline set of documentation, you can create variants to customize the documentation, while reusing much of the existing content. In this way, you can produce “one off” documents for important customers or target documentation to different skill levels while still getting the cost, quality, and labor efficiencies found with single sourcing. Process Management
To manage the documentation process, POET CMS integrates into workflow applications. POET CMS integrates into Staffware 2000, but can be customized to work with any piece of workflow software. Documentation managers can track project progress through the workflow software and refine their processes through trial and error. The workflow applications permit you to automate much of the process: for example, as steps in the process are completed, the software sends triggers to notify authors, editors, translators, and managers that the next step in the process can begin.
POET CMS Application
Minolta GmbH in Hanover, Germany, is integrating POET CMS into their documentation processes. Minolta knew that they needed to quickly find a tool to fit their documentation process. While documenting photocopier, fax machine, and printer documentation that is translated into several languages, Minolta’s documentation department was creating 50,000-75,000 pages of documentation to maintain every year.
They needed a more efficient way to create their documentation because much of the information is redundant and reusable. They first decided to standardize on SGML because it is vendor neutral, not likely to become obsolete, and accessible from many applications. They then began looking for a tool that would support all their processes using SGML and found POET CMS. With POET CMS, they were able to create the granularity they needed to reuse information, use the Variants feature to support translation, and access and store previous versions using the Editions feature to comply with administrative and legal requirements.
Minolta, as well as many other organizations, discovered that finding a tool that integrated with their process without having to tailor their process to a tool produced the results they were looking for: the ability to reuse information and create custom documentation while reducing production time.