Notes from the CMS conference

Attending the CIDM “Content Management Strategies” was very interesting. Some things that peaked my interested are below. Reader beware - no warrenties on the accuracy.

Globalization
- Machine aided “gisting” (with appropriate disclaimers) can be used for dynamic content. The World Bank is using this. For important content (i.e. those few pages that are used the majority of the time), translation must be done by humans. Text in graphics is a large challenge.
- Metadata and logic for search is different for different cultures
- The largest number of sales of PCs and the greatest utilization of Web search are from China.

Topic mapping and search
-The first level of search on a web site is the tool bar and other menu systems (often designed by programmers).
-Keywords are high maintenance and users must know the list to search. Key words don’t capture the relationship between topics.
- Ontologies – Relate subjects on the topic level and can capture the relationship between topics. A link for more information is http://www.ontopia.net/omnigator/models/index.jsp.

On CMS implementation
- Keep in mind the need to meet deadlines, while learning new tools
- Outsource or have a dedicated team do the conversion into the system, this is a one-time event, so don’t waste time training a lot of people.
- During the implementation, focus on training not process re-engineering.
- For those resisting the change, try to determine what is causing their fear and address.

Collaboration and workflow
- In many cases collaboration and workflow are just new models for pushing a document around.
- Where knowledge comes in is capturing the collaboration and the context of comments.

XML and document components –
- 90% of content is already in other documents
- The challenge is in linking the document components.
- Defining dependencies is important – for example changing data in one document may affect 10 documents. Do you need to allow folks access to the previous versions?
- This still seems to be an art to me, with folks understanding the challenge, but requires a lot of programming to solve.

DITA – There was a lot of discussion on DITA. Can say that I really understand it yet, but here’s a link: http://www-106.ibm.com/developerworks/xml/library/x-dita1/