You're right that there's more reuse in OpenACS 4.x than 3.x. Bugs in the CR will show up in a variety of clients, for instance. On the other hand, the odds are better that we'll get the CR right than, say, a dozen similar ad-hoc attempts to implement subsets of the functionality for a dozen different packages.
Code sharing usually wins and I see no reason for OpenACS 4.x to not fit this common scenario.
I agree that Luke's made a good point. These separate packages also provide simple-to-use functionality for sites that perhaps don't want to run the CMS system or even ETP. They're a lighter-weight approach.
Jun, if you do have time to work on News let me know. I don't think it would be hard to port. I looked at it, it's a bit messy but relatively simple.
