Best Practices for Internal Research Work Product Databases

As a follow up to my prior post on full-text keyword searching versus controlled vocabularies, I am wondering what law firms are doing regarding harvesting and re-using their internal research work product (research memos, client bulletins and newsletters, reasoned opinions and the like).

I know of some firms that actually catalog them in a separate database using a simplified legal taxonomy. I assume the other extreme is doing nothing but making them available on a document management system (DMS) to be searched full-text by keyword.

For those that do some level of profiling or cataloging:

1) What taxonomy works best (and at what level of detail)?

2) Do you rename the document?

3) Do you keep the documents in a stand-alone repository or library (or are the documents in the same repository as client files)?

4) How many staff are involved? Full-time?

5) Do you evaluate documents for quality before they are added? If so, who does that?

6) Do you weed or cull documents after a certain period of time (e.g., after 10 years)?

If any readers of SLAW are willing to comment here or to reply to me offline, I would be happy to later update this post with any some of the best practices that are followed.