I've been thinking about basic concepts required for Blog-based information services. My thought wanders...

XML feed services for databases
Feeds can include any kind of data that can be put in databases to help in knowledge capturing and reuse (creation). These sources might include news articles, press releases, weblogs, document repositories, email discussion groups. The difficulty in providing access to information services around these feeds should is providing a mechanism for making them findable. The first and best method of providing access is to embedd a link for an XML feed attached to database search results. That is to say, allow people to search any appropriate database and allow the system to create a URL for retrieving those results in XML periodically. An alpha listing for viewing/browsing sources can be available with a reverse-chron. spool of recent entries. Feeds can be listed in categories as well if this makes sense.

Enterprise aggregation services
Info. services around aggregation are complex. This type of service is very different from the newsreading/aggregation service provided by clients for individual consumption. Enterprise use of aggregators for knowledge-based use might require more complex functionalities including archiving collected weblog entries for long-term storage and retrieval; providing a richer set of (automatically entered?) metadata for each entered record; providing some semi-automated means of classification via a controlled vocabulary (taxonomy), e.g. where terms occurring in the blog entry text (title, description) are compared with synonyms, phrases, more complex boolean expressions that can be mapped to terms in the controlled vocabulary.

[I'm wondering on drupal.org how much of the above concept is possible with Drupal]

Enterprise blog creation services
This is perhaps the most difficult service to push. Providing software for users that allows them to easily create blogs is a great idea as long as policy doesn't requires that users only operate that software because it is corporate standard. Pushing a service that is controlled centrally can mean great benefits in terms of controlling metadata at the point of creation, so this avenue should be pursued if viable.