This week I’m in Vancouver at the DocTrain West Conference. I’ll be posting each day on the sessions I attended that day. I’ll cover the highlights and add comments.
Day one offered four pre-conference workshops. I chose Content Engineering: Workshop, presented by Joe Gollner of Stilo International. Joe is an excellent speaker, who kept us engaged for 3.5 hours. It’s impossible to summarize the full seminar in a blog entry, so I’ll hit the high points and add a few comments of my own.
The workshop was a comprehensive introduction to Content Engineering, which Gollner defines as the “application of rigorous engineering discipline to the design and deployment of content management and content processing systems.” He sees Content Engineering as a necessary means for controlling the explosion of both the volume and complexity of the content that organizations must deal with.
Gollner divides Content Engineering into two major activities: Content Management and Content Processing. While many CMS vendors see Content Management as the overriding discipline, and other activities, like Content Processing, as subordinate, Gollner sees Content Processing as an equal, and in many ways more complex, discipline. He also sees Content Processing as a weak link in many CMS offerings.
Following from an engineering approach, Gollner made some other important points:
- Metadata and link information must be treated as “first class” content; no different from any other content.
- This means that metadata and links must be “detachable” from any CMS; i.e., you must be able to export this information in usable, non-proprietary form, something that not all CMS’s support.
- Technology components must be “loosely coupled,” which means that interfaces must depend on the exchange of validated content, rather than depending on component to component interfaces like proprietary APIs.
- Processing rules aka business rules must be treated like content and therefore be expressed independent of any particular technology component.
- In general, must be able to export everything (content, links, metadata, processing rules, etc.) as processable content.
He concluded with some general comments and a “Top Ten” list of guidelines. The general comments centered around the importance of recognizing the content is inherently complex and getting more complex all the time. Effectively processing content requires engineering discipline that covers the entire life cycle. Doing this well is an elusive goal.
In the Top Ten list, without a doubt the most important point was “Don’t invest in Content Management technology too early.” Gollner has seen many projects get “bogged down in molasses” by committing to CM technology to early. Instead, he suggests focusing on Content Architecture and Content Processing first. Having seen exactly the same thing happen, I heartily endorse this recommendation.
Another notable item in the Top Ten was: Take a “Customer Service” focus in delivering tangible benefits to real users. All too often, the people who should be receiving new features and benefits from Content Engineering are forgotten and see little or no direct benefit. It’s important to keep delivering benefits to real users and not just “cool toys” for internal users.
Overall, I found the session valuable and hope it is a harbinger of how the rest of the conference will go.
