I just returned from a two day thinking session in a monastery, where Enki, Mihi, Smi, Sushimaster, Weezur and me were discussing the future development of MeManage, a knowledge management technology I've been working on since I moved to knallgrau.

MeManage is (becoming) a technology for indexing, analysing, sharing and searching arbitrary resources. Components include

• A spezification for a set of scanners which scan different sources (like filesystems, OPML feeds, databases, websites,...) for arbitrary data blocks
• A spezification for a set of adapters that parse these blocks into a general data structure.
• A spezification for a set of analysers that analyse the structure and content of the gathered data, extract tags, rank them and put them into a a set of arbitrary dimensions (=~ 'collections of tags').
• A Specification for a Meme object, which is a data structure that holds all the dimensions of tags, as well as some other meta information about the data.

Memes are stored in a database and can be shared through a P2P network.
Clients (which communicate with the platform through a webservice interface) can then search the network for memes in different ways (e.g. find memes whose data match a query term, find memes related to another meme,...) and use the metadata from the results (e.g. 'give me a list of people strongly related to the MeManage project').

Our reference implementation (Java) can already create memes and share them through a jaxta network.
There are also some reference implementations of clients in progress that

• create a searchable, dynamic view of an existing network (more or less an ad hoc creation of an intranet).
• plugins for twoday and twodex (our upcoming RSS aggregation tool). Examples include recommending tags for posts, recommending related articles, show items which are somehow connected with a person,...
• a PKM (Personal Knowledge Management) tool, which you can install on your local machine to share documents with groups of people