I'm working on an idea for a simple microblogging publishing pipeline: First, I use my Twitter account to keep track of what I read by starting the tweet 'reading…'. Then, each week or whatever, I can use a ruby program and twitter gem to generate a blog post which I can then edit and add more comment to. The idea is that Twitter acts as an instant active bookmark, 'active' in that the tweet is published content itself and not just a bookmark in a web browser or somesuch. There is no 'todo' list, as the items on the list have already been published. Bingo!
I thought that this was a great article as it discusses the practical considerations of running an internal publishing system (intranet, Yammer, etc.). I've seen this when I worked on the intranet at NCsoft. It was fairly successful, but ultimately why would people use in internal system when they can use the internet? Including a holiday booking system and a meeting room booking system ensured that the intranet was a frequent destination and this helped to encourage use of forums, but never really for work discussions, and it was seldom used by management.
These lists of principles smacked of the old mission statements, but they might actually be worth something. They reminded me of Donald A. Norman and Steve Krug: if a user fails to operate an object then its the design that is wrong and not the person. Design should remove the opportunity to get it wrong, e.g. a door that you push should have a plate and not a handle.
New York appear to be 'getting' #opendata