From this description it is apparent that teams need a place to share information and a way to publish it for the world to see. In simple terms, these requirements boil down to a social bookmarking system with perhaps a set of internal blogs and an external blog. The external blog might also include a public link blog. There are many ways to fulfill these requirements.

First, we could build it from components. The social bookmarking system could just come from a web service such as del.icio.us as Infoworld did, or we might use scuttle or de.lirio.us for a self-hosted system. We could use any of a number of blogging platforms including wordpress or movable type for self-hosted blogging. We could use typepad or blogger if we did not want to host it ourselves.

The advantage of using preexisting components is that their individual user interfaces and stability benefit from testing over a large user base. But, there is a real problem with this approach: we have to integrate the components. This involves developing a workflow, creating an interface for the integrated components, and building a training program around those. We would have to achieve the training in about 1.5 to 2 hours during the first session of the bootcamp. The workflow would involve going from bookmarking in one system to blogging in another.

This combination of factors led me to conclude that the advantage to be derived from using prebuilt components is outweighed by the issues we would face in building and maintaining an integrated system for just this one purpose.

What the Port Network brings us, along with a few drawbacks

What we are getting with The Port is a micro-community of five to six people with a public interface to the rest of the world. Dan Backus, the Port's CTO, and his team have had to work hard to scale back The Port from its usual size of hundreds of participants with time to learn its many features at a less hectic pace.

Our scaled back version of The Port Network includes:

• A feedreader for tracking sites that publish RSS and atom feeds.
• A bookmarklet for directly blogging articles read in a web browser.
• A team “feedroll” that allows participants to add relevant content to their research database from other sites that have RSS and atom feeds.
• The ability to create blogs at will.

In some ways, the mix of features is close to ideal. The one major drawback is, as it turns out, the reliance on RSS. Most resources for traditional industries like cleaning and restoration do not have RSS feeds. However, there is plenty of information on the web. What we will likely do is designate The Port's blogs to act as link blogs for the purpose of bookmarks. In this regard, tag support would be a positive.

Another issue we will face is training participants on a new system that does not already have a lot of users. Dan Backus has agreed to generate some demo movies, but this is an area where we will have to do a lot of work.