[Sigia-l] Corporate Blogging - the problems

[Michael Angeles]

> So -  to ask a very silly sounding question - can a blog be fed to a 
> whale
> without killing it (the blog that is)?  Anyone been feeding their 
> whales
> blogs? How about dolphins  (Small to Medium enterprise sites?).

I'm actually going to be talking about strategies for making the 
textual data from blogs/k-logs (knowledge logs) usable inside the 
intranet. The type of blogs we're talking about are topical or project 
related weblogs that people keep inside an intranet to share 
information within communities of practice.

I guess, as far as the IA is concerned, for now we're looking at blog 
data as we would any kind of data. The way to findability has, for us, 
much to do with how the system can represent the knowledge contained in 
it -- to express what the data is about when you ask questions of it. 
So how to feed the whale... Our strategy with dealing with any type of 
textual data, from vendor's news feeds to internal document 
repositories, is to capture a rich set of metadata and at some level to 
apply classification. We do this using various filters including 
end-user classification, but mostly relying on textual analysis and 
clustering (by machine) which is altered/approved by manually (by 
humans). Essentially text goes in, algorithms try to apply indexing 
using our subject taxonomy, and then librarians check, modify or add 
appropriate index terms. Most of these librarians are steeped in the 
subject area that they classify documents in.

The strategy is to then find a way to make the aggregated and 
classified data usable to the enterprise by linking people with 
documents, people, projects, etc. based on common areas of interest. Or 
something like that. That's as far as we've thought for now in terms of 
strategy.

-Michael

   . . . . . . . . . . . . . . . . . . . . . . . . .
   michael angeles
   michael at studioid dot com