Just taking a step back and organizing some of the threads that have emerged out of comments.
It seems to me there are a couple of different lines to follow here:
1. Plumbing. What are the taxonomies of “news” that make sense; what are the tools to make it easier to tag/sort/put them in databases; what are the human/organizational issues that will help increase the value of such databases? And so on. This includes discussions about microformats, standards, OpenCalais, Document Cloud, ways news organizations can or should partner, share data and formats, etc. Some of this will be heavily technological; some needs to be grounded in taxonomy/story structure; and some in human and organizational relationships.
2. Products. And then what do we do once we get that data together? What kind of products/visualizations/databases can we build on top of that data that makes sense? Does having the location of every car crash in the city yield anything useful? What are the visualizations of people relationships that have value? What are the essential elements that are needed to make a product work, and what’s nice-to-have? Politifact is nice to look at, but it’s the ratings – Liar Liar Pants on Fire, etc – that make it much more user-friendly, and that isn’t necessarily an element many people might have thought of immediately, and so not created that data field. Recognizing that, of course, sometimes you just have to get the data and trust that value will come. (Although it may be harder to get the VC to hand over cash if that’s your business case.)
3. Process. How do we change our day-to-day work processes/habits so we’re more actively creating things that will have longer shelf-lives and more value? What kinds of stories should we do; how should we change the way we do them; should we think beyond stories and more atomizable elements of information? In practical terms, can we start filing current and archive versions of stories, rather than wait for a technological fix to come along that will find and change all date references in stories? What can we do to get reporters/editors to file extra information in their notebooks into a database? And so on.
There are other threads, of course, not least a big question: How do we make some money out of this?
All those Ps and the Q are, of course, interrelated. There’s no point talking about product if you don’t know what data might be available to power it, and that requires understanding newsroom processes as well as technology; and you can’t design the newsroom process without knowing what product you want to build; and there’s no point building a cool product if there isn’t a sustainable way of keeping it alive. (Actually, there’s lots of reasons to build cool things, but it does become a problem if you can’t keep food on the table.)
But I think it’s helpful to keep them slightly separate, simply as a framework to think about where we can increase value in each of those areas – whether in dollar terms, in reader-use terms, or in public interest terms.