datablogging Attention Data: Log Types Shared as OPML
11 Hour
32 Minute
AM
New feature for Reger.com users: Share the list of log types that you use via OPML.
If you're using a running log you're interested in running. If you're using a movie log you're interested in movies. If you're using a biking log you're interested in biking. Etc.
You grant these things some of your attention. Some of your time. So a list of your log types is a type of attention data. Another good example of attention data is your blogroll, the list of other sites that you read, or give your attention to.
As of today, Reger.com publishes, for each blog, an OPML file that lists for each log: 1) Log Name 2) Log URL (web) 3) Log Schema
The key is the XML Schema. The XML Schema file defines the format of the log. The microformat. So an aggregator, say PubSub, can read this file, parse the format and know what to expect from blog entries on a particular blog.
Users can read each other's log type attention files to get a sense of what they're actively blogging about... what they give their attention to.
In the datablogging/sb world we need to flesh out the framework a little more. We need to tie the extended data into the world of blogs better.
A first step is publishing, for each blog, those microformats that are in use. Next we need some sort of auto-discovery tag so that search engines can find this list.
From there we need to agree on the format of formats. Whether we describe a format with XML Schema, RDF or something else, aggregators need to know what to expect.
And we need to agree on what we'll describe in those formats. I propose that we only describe the extended data. We already have RSS, Atom and other formats that do a great job describing a blog entry. And we have a robust framework of publishing, aggregating and consuming in place.
We only need new work to flesh out the framework of extended data... structured data. Once we have an agreed-on language to describe a microformat (XML Schema or RDF) then we need to forge links into the existing framework.
Structured blogging does this by publishing extended data hidden in html. Reger.com supports this and is publishing such data today. We also publish extended data in RSS.
We should also extend the metaWeblogAPI to include extended data.
This is the way we should do it: leverage existing frameworks and extend them. Nothing new here in my proposal.
Post-Ironman Melancholy