When my son was a bit younger, we noticed that when we asked him a question, he'd answer with a question. Inevitably this led to the following exchange:
Q: Why do you always answer a question with a question?I remembered that yesterday when somebody asked me: "Who do you read on the topic of REST?" I answered with a question of this form:
A: Why does everybody always ask me that?
/item[contains(.,'REST')]/titleWhich, translated into English, asks: "What RSS items have I read that mention REST?". You can run that query directly here, and you'll get back the most recent 25 matching items. Note that you can also subscribe to that query, using the orange XML icon at the bottom of the page. So not only can I answer the question with a question, but my answer keeps producing new answers.
I realized today, though, that "What RSS items mention REST?" is not quite the same as "Who's talking about REST?" So here's the next turn of the crank in my ongoing quest to query the blogosphere: views of who (in my set of feeds) is talking about what. For example:
who in jon's feeds is talking about REST?
|James Governor's MonkChips||3|
|Stefan Tilkov's Random Stuff||3|
|tbradford.org: Tom Bradford||2|
|Dare Obasanjo aka Carnage4Life||2|
|Making it stick.||2|
|Loosely Coupled weblog||1|
|The Mountain of Worthless Information||1|
|All Things Distributed||1|
|Feedster on: jon udell||1|
|Jon's Radio (full-length descriptions)||1|
|java.net Weblogs by Michael Champion||1|
The links drill down into the set of matching items for each feed. Note, in passing, that Quoderat is much concerned with REST. I've been meaning to mention that. Quoderat belongs to Dave Megginson, an XML god who recently arrived on the blog scene and has been asking and answering important questions about REST design.
Here are some other questions:
The page that answers these questions includes an input box, so you can ask your own questions. In this version, I haven't exposed the full power of structured search, it's just plain text search, but everything I've shown earlier about structured search could apply here as well.
Of course the scope of the answers is only the set of feeds I happen to read. So your interest in these results will be in direct proportion to the overlap between your set and mine.
In the future blogosphere that I want to live in, that restriction will vanish because everyone's feeds will work this way. Then we can ask really interesting questions, like: "What is the unique set of items about REST from all of the feeds subscribed to by people whose thinking about REST I trust?"
Former URL: http://weblog.infoworld.com/udell/2005/03/04.html#a1190